Understanding AC Scaling [Archive] - Giant in the Playground Forums

MaxWilson

2019-01-29, 05:22 PM

I had an initial thought that perhaps +AC items were expected or that spells were expected at higher levels to account for the mathematical difference, but PC attacks against monsters do not have any similar kind of trend. The chance to hit an equivalent level monster remains quite constant at ~60-65% throughout all levels.

As an exercise I've changed AC formulas to be 8 + prof + ability so chain mail is 14 + prof + ability, plate is 16 + prof, studded leather is 12 + prof + ability and the scaling issues almost disappear.

Have I missed something? Is there something that is meant to account for this difference at level ~13+?

=======

Discussion: Please limit the discussion to the math and statistics of the game.

I suspect you would benefit from rereading 5E designer Rodney Thompson's post about bounded accuracy. A copy is archived here: http://bluishcertainty.blogspot.com/2016/06/bounded-accuracy.html

For example, this bit here:

The DM's monster roster expands, never contracts. Although low-level characters probably don't stack up well against higher-level monsters, thanks to the high hit points and high damage numbers of those monsters, as the characters gain levels, the lower-level monsters continue to be useful to the DM, just in greater numbers. While we might fight only four goblins at a time at 1st level, we might take on twelve of them at 5th level without breaking a sweat. Since the monsters don't lose the ability to hit the player characters—instead they take out a smaller percentage chunk of the characters' hit points—the DM can continue to increase the number of monsters instead of needing to design or find whole new monsters. Thus, the repertoire of monsters available for DMs to use in an adventure only increases over time, as new monsters become acceptable challenges and old monsters simply need to have their quantity increased.

This means that AC never goes obsolete in 5E unless the DM is doing something unusual like using exclusively high-CR monsters.

Kadesh

2019-01-29, 05:23 PM

At high level, getting hit by a melee attack is usually much less threatening than getting hit by one at low level, so a binary yes/no to get hit at low level is much better at keeping you alive when the swing of a single d8+4 damage roll can result in your character dropping to 0 or being fine. Taking 5d8+6d6 damage at 20th level with a plethora of healing and Resurrection and mitigation, and negation is much less threatening, so having the same chance to hit at low levels is just meh, for interest.

LudicSavant

2019-01-29, 05:24 PM

After doing some math recently I've realized that PC AC scales poorly at higher levels. ACs scale at about the same rate against enemies of the same CR until around level 13 when AC starts to become less valuable, becoming almost useless by level 20.

https://i.imgur.com/2UKhN0u.jpg

Here you can see that Plate (starting at 16 with Chain mail, swapping to Plate at 5th level) has a 50% chance of being hit by the average equivalent level monster. That rate holds true with minor fulctuations until about level 13 when it starts to be hit a lot more often. At level 20 the chance to be hit by an attack from an equivalent level monster is 80%. I'd like to understand why this is.

I had an initial thought that perhaps +AC items were expected or that spells were expected at higher levels to account for the mathematical difference, but PC attacks against monsters do not have any similar kind of trend. The chance to hit an equivalent level monster remains quite constant at ~60-65% throughout all levels.

As an exercise I've changed AC formulas to be 8 + prof + ability so chain mail is 14 + prof + ability, plate is 16 + prof, studded leather is 12 + prof + ability and the scaling issues almost disappear.

Have I missed something? Is there something that is meant to account for this difference at level ~13+?

=======

Discussion: Please limit the discussion to the math and statistics of the game.

You're missing a lot of sources of defenses that are common in higher level games, where buffs, debuffs, and other abilities that improve defense (or reduce enemy offense) are much more commonplace and can make up a significant portion of your defenses.

You're also missing that not all enemies are equal-CR. Minions and swarms are a thing and AC is important for mitigating them (with even a couple points of AC often providing exponential returns against them).

Unoriginal

2019-01-29, 05:28 PM

AC isn't meant to scale much. HP is.

Most characters' ACs will plateau at lvl 10 at the latest, and in general earlier, while there is quite a huge gap between a lvl 10's HPs and a lvl 16's ones, for example.

Keep in mind that the enemies' AC doesn't scale up much either. Not many go beyond AC 19, no matter their CR, and the few who do are meant to be especially tough to hurt.

Kryx

2019-01-29, 05:31 PM

I suspect you would benefit from rereading 5E designer Rodney Thompson's post about bounded accuracy.

This means that AC never goes obsolete in 5E unless the DM is doing something unusual like using exclusively high-CR monsters.
I've read it, but that doesn't explain the issue. If the math is intended then essentially it's putting high CR enemies off limits as they are far beyond the goldilocks zone of ~60% chance to hit.

AC is practically obsolete at the top tier levels against equivalent CR enemies, not to mention high CR enemies.

Additionally the game has a tolerance built in for magic items, so even if AC did scale with prof bonus to ~21-22 it's not going to break anything that the game doesn't already make readily available in the form of magic items.

==========

At high level, getting hit by a melee attack is usually much less threatening than getting hit by one at low level
The same is true for monsters, but they scale. If HP is the defense at higher levels why even both with AC? 20% chance to not be hit is crazy..

========

AC isn't meant to scale much. HP is.
...
Keep in mind that the enemies' AC doesn't scale up much either.
Enemy ACs do indeed scale. They scale from ~13.6 average to ~19.6 average which is a gain of 6, or the exact same gain as proficiency bonus +2 from ability modifiers.

=====

You're missing a lot of sources of defenses that are common in higher level games, where buffs, debuffs, and other abilities that improve defense (or reduce enemy offense) are much more commonplace
The same things exist for +to hit, but those numbers do not have the same scaling issue.

You're also missing that not all enemies are equal-CR.
The average is a weighted average against enemies within a few CR (~4 each way). Against higher CR enemies the numbers become worse.

LudicSavant

2019-01-29, 05:37 PM

The same things exist for +to hit, but those numbers do not have the same scaling issue. They're not actually the same things though, and you didn't include any such factors in your math. PCs have an awful lot of tools at their disposal at higher levels to help them make enemies miss, and AC aids this.

For a great many characters, your ability to dodge attacks does not stop scaling as soon as you buy a suit of full plate. That's what you're missing in your chart.

Kryx

2019-01-29, 05:40 PM

PCs have an awful lot of tools at their disposal at higher levels to help them make enemies miss, and AC aids this.
PCs also have the same tools to help them hit, but to hit does not have the same issues.

If you want the math for PC chance to hit:
https://i.imgur.com/jvdQmXq.jpg

Why does +to hit scale into high levels at the same rate? Because it scales at the same rate as proficiency bonus + ability score scaling. Primary ability scores generally add +2 to the modifier and proficiency bonus adds +4 over the 20 levels, equaling the ~+6 AC scaling of monsters.

Kadesh

2019-01-29, 05:46 PM

==========

The same is true for monsters, but they scale. If HP is the defense at higher levels why even both with AC? 20% chance to not be hit is crazy..

========

You need to explain this point better.

Kryx

2019-01-29, 05:48 PM

You need to explain this point better.
Which point?

80% chance of being hit being unreasonable?
or
Monsters have higher HP and AC, not just higher HP. So why should PCs only have higher HP and not AC that scales at the same rate?

LudicSavant

2019-01-29, 05:49 PM

PCs also have the same tools to help them hit

They are different abilities, do different things, and have different mathematical results.

If you're just going to keep saying they're the same despite that I don't know what to tell you.

Kryx

2019-01-29, 05:50 PM

They are different abilities, do different things, and have different mathematical results.
A typical AC boost is Shield of Faith for +2 AC.

A typical +to hit boost is Bless for +~2.5 to hit.

There are plenty of tools for raising AC and +to hit. This is not a reason for the math to not scale for one of them.

DMThac0

2019-01-29, 05:53 PM

One thing I've yet to see you take into account is the abilities of the PCs to mitigate many of the threats of a CR equivalent creature.

I stun the creature = advantage = ~+5 to hit.
I have Rakish Audacity = advantage (when used properly) = ~+5 to hit
I cast Foresight = advantage

I cast Foresight = Disadvantage on attacks toward me = ~+5 AC
I have a Ring of Protection +2, Plate +1, each increasing my AC.

There are far more items, spells, and abilities available to high level PCs to level the playing field so that the discrepancy is actually a misnomer.

Unoriginal

2019-01-29, 05:58 PM

CR X means "medium encounter against 4 PCs of X level/easy encounter against 5 PCs of X level".

6 points of scaling isn't much when you consider the number of attacks an high-CR monster will face each round, both due to the number of PCs and the Extra attacks.

PhoenixPhyre

2019-01-29, 05:59 PM

MaxWilson

2019-01-29, 05:59 PM

I've read it, but that doesn't explain the issue. If the math is intended then essentially it's putting high CR enemies off limits as they are far beyond the goldilocks zone of ~60% chance to hit.

AC is practically obsolete at the top tier levels against equivalent CR enemies, not to mention high CR enemies.

Yes, against a high-CR enemy other defenses like resistance (Absorb Elements) or mobility are much better. So what? You won't be fighting enemies of high CR all the time. If you fight fifty Githyankis and one Ancient Red Dragon (per discussion on another thread), AC does you very little good against the red dragon and lots of good against the Githyanki, whereas Regenerate does you very little good against the Githyanki and lots against the Red Dragon.

In general, 5E's designers made sure that it's a lot easier to boost defense than offense. E.g. you can spend a feat on Elemental Adept to boost your Fireball damage output by something like 10%, or you can learn a single first-level spell like Absorb Elements to double your defense against fireballs and red dragon breath (and blue dragon breath and black dragon breath and white dragon breath, etc.). You can boost your defenses further with Protection From Evil, Sanctuary, Blur, Blink, Antipathy/Sympathy, Contingency (Dimension Door), Phantom Steed, Regenerate, Death Ward, Foresight, Shapechange, etc., many of which don't even require concentration.

You asked what you were missing, and here's what you're missing: AC is not supposed to scale with level. Nothing is. There's no numbers treadmill here.

Man_Over_Game

2019-01-29, 05:59 PM

Kryx

2019-01-29, 06:00 PM

There are far more items, spells, and abilities available to high level PCs to level the playing field so that the discrepancy is actually a misnomer.
I'm not understanding this point. I believe I've addressed it a few times now. Let me try again: High level PCs have access to many items, spells, and abilities that can benefit them. Those benefits can equally benefit the ability to hit as the ability to defend against attacks (AC).

This does not seem like a justification for one to scale and not for the other.

PhoenixPhyre

2019-01-29, 06:02 PM

I'm not understanding this point. I believe I've addressed it a few times now. Let me try again: High level PCs have access to many items, spells, and abilities that can benefit them. Those benefits can equally benefit the ability to hit as the ability to defend against attacks (AC).

This does not seem like a justification for one to scale and not for the other.

The bolded part is assumed without evidence. There are a lot more spells and abilities that give defensive benefits (which may or may not include actual AC bumps, but may include resistances, shielding effects, disadvantage, etc) than there are that actually give numerical to-hit bonuses.

Keravath

2019-01-29, 06:10 PM

I'm not understanding this point. I believe I've addressed it a few times now. Let me try again: High level PCs have access to many items, spells, and abilities that can benefit them. Those benefits can equally benefit the ability to hit as the ability to defend against attacks (AC).

This does not seem like a justification for one to scale and not for the other.

On the defense front, players have access to magic items and spells that can be expected to increase their AC in the level 13 to 20 range .. scaling up to +3 armor, +3 shield, rig of protection and cloak of protection.

On the attack side, PCs get improved attack abilities and most of the opponents you are discussing get lair and legendary actions.

The increased monster to hit scales against the increasing AC of the PCs due to magic and spells.

The point is both do scale .. a +14 to hit Cs AC 24 from magic and armor .. nets 10 to hit .. similar to lower levels.

ad_hoc

2019-01-29, 06:13 PM

Unoriginal

2019-01-29, 06:15 PM

Thank you.

People think the CR system is broken because they fight 1 solo per long rest.

For examples of it working well just run a published adventure.

TBF the published adventures don't necessarily have the fights CR-based.

Kryx

2019-01-29, 06:18 PM

Kryx, the key, faulty assumption is that the proper band for calculations is a static CR = level +- 4. That means you're only facing solos (or at most duos, for deadly encounters). That's so far off the guidance (and thus the underlying math) that of course your results will look screwy. That's fundamental to bounded accuracy. At higher levels, you're facing a lot of the same (relatively) low-CR things you did earlier, just a lot more of them.
Hey, thanks for engaging in the math discussion! Super appreciated!

5e is set up around mass fights. Roughly 1 enemy per PC or more. If you use Xanathar's ratio-based encounter guidelines, the weighted averages for medium mass encounters (max 2 enemies per PC) go like:

level 1: average CR 0.25
Level 5: average CR 1
Level 11: average CR 5
Level 17: average CR 8
Level 20: average CR 9
Lets look at each of these tiers:

level 1: average CR 0.25 enemy has ~12 AC so our PC has ~70% chance to hit. Average enemy has ~4.1(4) to hit, so our Plate PC has ~45% chance to not be hit.
Level 5: average CR 1 enemy has ~13 AC so our PC has ~75% chance to hit. Average enemy has ~4.3(4) to hit, so our Plate PC has ~45% chance to not be hit.
Level 11: average CR 5 enemy has ~14.8(15) AC so our PC has ~70% chance to hit. Average enemy has ~6.6(7) to hit, so our Plate PC has ~50% chance to not be hit.
Level 17: average CR 8 enemy has ~15.2(15) AC so our PC has ~85% chance to hit. Average enemy has ~6.7(7) to hit, so our Plate PC has ~50% chance to not be hit.
Level 20: average CR 9 enemy has ~16 AC so our PC has ~80% chance to hit. Average enemy has ~8.5(9) to hit, so our Plate PC has ~60% chance to not be hit.

In this scenario we're looking at ~15% scale for both, far more reasonable numbers. Though 70-80% for chance to hit is far outside the goldlilocks zone.

Since 5e came out I've often heard that the encounter building guidlines are quite far off. Boss battles were often using double or tripple deadly. Double deadly for example would be 4 CR 15 monsters against 4 level 20 characters.
This shouldn't be all battles as I'm a huge proponent of short rests and the adventure day, but worth a mention for those epic battles.

under those assumptions the math works beautifully, with constant miss chances at all levels on both sides (both player-> monster and vice versa). This is strong evidence that these are the correct assumptions for monster CRs, namely the ones the system math was done for.
Indeed, I believe it as you say that the math was calibrated for lower CR enemies, though it's not constant by my numbers above.

Kryx

2019-01-29, 06:20 PM

AC is not supposed to scale with level. Nothing is. There's no numbers treadmill here.
There is definitely a numbers treadmill in 5e. It's not to the same degree as it was in 4e, but it is still there. Remove proficiency bonus from your main weapon and you'll see it rather quickly. ;)

MaxWilson

2019-01-29, 06:21 PM

Kryx

2019-01-29, 06:24 PM

This is not true. Boosting defense is more cost effective in 5E than boosting offense. It's really hard to double a fighter's DPR with spells but easy to double his rounds-until-death.
Max, I appreciate the disccusion but please focus on the math aspect. You're twisting my words here to go beyond AC to overall defense. It's not the same and is not relevant to the discussion.

PhoenixPhyre has pointed out why the math is the way it is. Average enemy CR is meant to be much lower.

ad_hoc

2019-01-29, 06:26 PM

Kryx

2019-01-29, 06:28 PM

If to hit bonuses go up while AC stays the same that isn't a treadmill.
The AC doesn't stay the same. Look at the AC of enemies using PhoenixPhyre's CR numbers above. Enemy AC goes from ~12 to ~16. PC AC goes from ~16 to ~18 or ~14 to ~17.

PhoenixPhyre

2019-01-29, 06:29 PM

Hey, thanks for engaging in the math discussion! Super appreciated!

Lets look at each of these tiers:

level 1: average CR 0.25 enemy has ~12 AC so our PC has ~70% chance to hit. Average enemy has ~4.1(4) to hit, so our Plate PC has ~45% chance to not be hit.
Level 5: average CR 1 enemy has ~13 AC so our PC has ~75% chance to hit. Average enemy has ~4.3(4) to hit, so our Plate PC has ~45% chance to not be hit.
Level 11: average CR 5 enemy has ~14.8(15) AC so our PC has ~70% chance to hit. Average enemy has ~6.6(7) to hit, so our Plate PC has ~50% chance to not be hit.
Level 17: average CR 8 enemy has ~15.2(15) AC so our PC has ~85% chance to hit. Average enemy has ~6.7(7) to hit, so our Plate PC has ~50% chance to not be hit.
Level 20: average CR 9 enemy has ~16 AC so our PC has ~80% chance to hit. Average enemy has ~8.5(9) to hit, so our Plate PC has ~60% chance to not be hit.

In this scenario we're looking at ~15% scale for both, far more reasonable numbers. Though 70-80% for chance to hit is far outside the goldlilocks zone.

Since 5e came out I've often heard that the encounter building guidlines are quite far off. Boss battles were often using double or tripple deadly. Double deadly for example would be 4 CR 15 monsters against 4 level 20 characters.
This shouldn't be all battles as I'm a huge proponent of short rests and the adventure day, but worth a mention for those epic battles.

Indeed, I believe it as you say that the math was calibrated for lower CR enemies, though it's not constant by my numbers above.

I've done the math with the actual average numbers (see the spreadsheet in my signature). And it's within 10% either way. That gets swallowed up real fast by the actual fights involved.

The big reason people say the encounter guidelines are off is that they're only running 1-3 encounters per day. I'm running a project of simulated random battles (to test homebrew) and finding that doing a MM/H/MM day (/ are short rests, M is medium and H is hard, which exactly meets the adventuring day XP thresholds) is really really tough on the PCs, especially at low levels. That, and focusing on solo/duo encounters. That's horrible encounter design in 5e, because it requires going way hard and risking a TPK to overcome the action advantage. Even legendary monsters benefit from minions. They really should make that very clear--thou shalt have most of thine fights with 1-2 enemies per PC.

Now of course if you're lavishing magic items (especially +X to hit/AC) or doing heavy optimization, the guidelines will be quite low. That's by design, because you're outside what they're built for (no +X items, no multiclassing, no feats). The numbers are quite solid otherwise.

Edit: another thing to examine is the variance in those numbers. It's huge--often a standard deviation of 2-3 (10-15%) in attack bonuses within a CR. And a 1-2 AC standard deviation. Which totally swallows any scaling trends just due to monster selection (which isn't random).

Kryx

2019-01-29, 06:35 PM

I've done the math with the actual average numbers
My numbers are from the monster manual and volos as well, they are actual numbers. I have 711 creatures. Old sheet that I haven't maintained as well, but you can see it: https://docs.google.com/spreadsheets/d/14Ia7HQk7-NOojTW0FWGT8wMvDSKbnN1PGDW3RtblCu4/edit#gid=251686708

I'll have to update my rolling averages to be much lower CR as you recommend. Sincerely, thanks again!

finding that doing a MM/H/MM day ... is really really tough on the PCs, especially at low levels.
Is MM/H/MM a typical adventuring day for you? I'd be curious for a general recommendation from someone who has obviously delved deep in to the math.

Or is there a different route that you'd recommend? Stick to xp budget over a whole day split over 3-6 encounters?

PhoenixPhyre

2019-01-29, 06:40 PM

Os is there a different route taht you'd recommend? Stick to xp budget over a whole day split over 3-6 encounters?

On mobile, but yes. I did this pattern because it's easy to generate and hits the benchmarks without difficulty.

KorvinStarmast

2019-01-29, 07:01 PM

This does not seem like a justification for one to scale and not for the other. OUr tier three group has been fighting a lot of giants. My Champion is frequently hit, despite having AC 23-25. (Yes, I have defensive fighting style, +1 RoP and +1 Cloak of P, ...)
And a crit from a stone giant or fire giant giant puts me not too far away from sleepy time. Current CON is 16, will probably raise that to 18 on the next ASI. In one fight, in one round, two stone giants versus me, four swings, two crits, and sleepy time for me.

At high level, the PC's need to shape the battlefield. (As a party).
Straight up Attrition, as the CR bonuses go up, can be lethal. What you are seeing via number crunching is what I have experienced with a high AC character.

If I added a cloak of displacement, though, the crit problem goes away somewhat.

When the party shapes the battlefield, though, we've noticed it is a whole lot less difficult lethality wise.

MaxWilson

2019-01-29, 07:01 PM

Max, I appreciate the disccusion but please focus on the math aspect. You're twisting my words here to go beyond AC to overall defense. It's not the same and is not relevant to the discussion.

PhoenixPhyre has pointed out why the math is the way it is. Average enemy CR is meant to be much lower.

*puzzled look* That's what I said, and I'm not the only one who said it, but apparently you needed to see it as a spreadsheet first so okay. At least now you know.

Don't just focus on the mean CR, look at the monster to-hit distribution over the whole range of likely opponents. Fighting a bunch of CR 3 githyanki and then a CR 23 ancient dragon is not the same thing as fighting two CR 13 Beholders in a row, despite having the same average. Don't just pull monsters of one CR for your analysis, use a wide range or you'll make inaccurate predictions.

Kryx

2019-01-29, 07:22 PM

That's what I said
Your initial post was the most insightful of the first posts, but it did not say the same thing as PhoenixPhyre. Later you were reinforcing others who were trying to explain away the difference in numbers via features, spells, and other buffs. As PhoenixPhyre has shown - features, spells, and other buffs are not the explanation at all. The explanation is the expected CR of enemies is not CR =~ level +-4. But lets move past it.

====

Don't just focus on the mean CR, look at the monster to-hit distribution over the whole range of likely opponents. Fighting a bunch of CR 3 githyanki and then a CR 23 ancient dragon is not the same thing as fighting two CR 13 Beholders in a row, despite having the same average. Don't just pull monsters of one CR for your analysis, use a wide range or you'll make inaccurate predictions.
While what you suggest is a good way to model things, it isn't something that can be analyzed at the statistical level in that way without huge swaths of calculations.
If someone has found a good way of calculating the variety I'd definitely want to see it and understand it. But if that doesn't exist yet then average CR will suffice as a group could feasibly fight within a few CR of that number (say CR 9) and be within the encounter rules.

KorvinStarmast

2019-01-29, 07:28 PM

Or is there a different route that you'd recommend? Stick to xp budget over a whole day split over 3-6 encounters? That's what I do. I work backwards from the day's budget ... and the point on variance was one well made. Thanks, as ever, for your love of number crunching. :smallsmile:

MaxWilson

2019-01-29, 07:39 PM

While what you suggest is a good way to model things, it isn't something that can be analyzed at the statistical level in that way without huge swaths of calculations.
If someone has found a good way of calculating the variety I'd definitely want to see it and understand it.

Monte Carlo sim is probably the easiest way. The hard part isn't the sim implementation, it's choosing what assumptions to use for encounter building. If you just did something simple like selecting Medium difficulty homogeneous groups from a uniform distribution of all MM/Volo's/Mordy's monsters with CR equal to or less than the PCs' level it would take less time to write the sim than to write up all the monsters. Couple of hours for each maybe.

Sigreid

2019-01-29, 07:39 PM

This is pure conjecture so take it as that. I've always assumed the intent was that the AC scales the way it does so that while the low CR mobs still pose some threat en mass, it is mitigated significantly by the AC whereas the to hit bonuses of the monsters outpacing the AC increase as they increase in CR is to ensure that the higher CR monsters are perceived as the greater threat.

The example would be a red dragon supported by 20 kobolds. The 20 kobolds add to the threat because they can hit the high AC character and do some damage. Say the character has an AC of 21. Just a fighter with the defensive style in normal full plate with a shield. You'd expect about 3 of those kobolds to hit per round if you discount pack tactics. They're tolling resources, but in a pretty minor way. The dragon is a much bigger threat because he not only hits a lot harder, but can almost be counted on to hit every round. In 3.5 by the time you're going after a serious dragon those kobolds will only hit on a crit, meaning they don't really add anything to the fight.

By having the AC scale slower than the to hit bonus, you can use the weaker mobs to make a more complete feeling encounter as well as fine tune the difficulty, set up a situation where the party needs to deal with a multi directional threat, etc.

To me, it's an advantage that you can never completely ignore the mooks.

Kryx

2019-01-29, 07:56 PM

That's what I do. I work backwards from the day's budget ... and the point on variance was one well made. Thanks, as ever, for your love of number crunching. :smallsmile:
Thanks for your input here - I'll try this method next time I run a game (will be a while though)

===============

level 1: average CR 0.25
Level 5: average CR 1
Level 11: average CR 5
Level 17: average CR 8
Level 20: average CR 9
Looking at these numbers again we end up with the following AC levels:
https://i.imgur.com/UiO3Qz4.jpg

AC shows that +prof to armor is not a good idea:
https://i.imgur.com/WjlPG2P.jpg

10% scale isn't such a problem - that is a good level. The AC levels are quite ok too - going from 45% chance to be hit to a 55% chance to be hit is at a good level.

However something seems wrong with attacks:
https://i.imgur.com/Dv7eJlP.jpg

There is a 10% scale here as well which is ok, but the starting value is 70% which is too high. I've read design principles in the past that aim for a lower ~60% chance of success. I'd be fine with 60% scaling to 70%, but 70 to 80 feels too high and makes options like advantage much less valuable.
I don't expect this to be an issue of variety. Perhaps I've missed something else here.

I'm off to bed for the night, thanks for the discussion that ended up in the math of the system and hopefully I can understand what's going on with chance to hit tomorrow.

KorvinStarmast

2019-01-29, 08:09 PM

Thanks for your input here - I'll try this method next time I run a game (will be a while though)

===============

AC shows that +prof to armor is not a good idea:
https://i.imgur.com/WjlPG2P.jpg

How about half prof added to AC, for Fighter class only, somewhat like the Bard's half prof for "jack of all trades".

I've been toying with that idea ...

LudicSavant

2019-01-29, 08:09 PM

I'm not understanding this point. I believe I've addressed it a few times now.

You haven't addressed it. You just keep making this same claim:

Let me try again: High level PCs have access to many items, spells, and abilities that can benefit them. Those benefits can equally benefit the ability to hit as the ability to defend against attacks (AC).

Which PhoenixPyre addressed. (http://www.giantitp.com/forums/showsinglepost.php?p=23670813&postcount=20)

Not only that, but your point isn't especially relevant to whether or not, as you claim, AC becomes "almost useless" as you level up. This is because monsters tend to not actually have a lot of ways to boost their attack bonuses. Sure, that Ancient White Dragon has a +14 to hit, but that's it. They don't boost that. Given how easy it is for the likes of an Eldritch Knight to be pushing AC around 30 + Disadvantage at that level, there's a very real possibility for the Ancient White to be struggling there.

MaxWilson

2019-01-29, 08:29 PM

You haven't addressed it. You just keep making this same claim:

Which as PhoenixPyre rightly pointed out, you simply assumed without evidence.

Not only that, but your point isn't especially relevant to whether or not, as you claim, AC becomes "almost useless" as you level up. This is because monsters tend to not actually have a lot of ways to boost their attack bonuses. Sure, that Ancient White Dragon has a +14 to hit, but that's it. They don't boost that. Given how easy it is for the likes of an Eldritch Knight to be pushing AC around 30 + Disadvantage at that level, there's a very real possibility for the Ancient White to be struggling there.

Another example:

Pit Fiend has +14 to hit, but a pretty standard Paladin/Sorcerer can easily have AC 21 (+5 for Shield) even without any magic items, plus (Quickened?) Protection From Evil (Devotion paladins get this for free even without a spell, from Purity of Spirit), which means the Pit Fiend has a 51% chance of missing outright (needs a 7+ at disadvantage) and only a 20% chance of actually hitting through the Shield (requires two rolls of 12+). The Paladin will have other defenses too (Greater Steed) but Kryx has requested that we not talk about things that aren't related to AC. It is readily apparent that the defensive boosts are stronger and more cost-effective than most of the Paladin's offensive boosts, which are mostly limited to Smite spells or Divine Smite.

Even if you do factor in magic items, the ceiling for defensive boosts is higher than offensive boosts: with a Cloak of Displacement and Plate Armor +3 and a Shield +3, your AC goes up by +6 and all your enemies without blindsight have disadvantage. But with a Greatsword +3 you get only +3 to hit, and with an Oathbow you get advantage and +3d6 extra damage against one enemy per day--I don't know of a canonical way to get +6 to-hit and advantage on all your attacks. Offense is more tightly controlled than defense in 5E.

Dragons are actually harder to impose disadvantage on than most high-CR monsters because they both have blindsight and are not vulnerable to Protection From Evil, but the Pit Fiend profile of +11-14ish to hit with darkvision + extra-planar nature is pretty common, and even cheap spells like Shield and Protection From Evil are pretty effective against that profile.

PhoenixPhyre

2019-01-29, 08:40 PM

However something seems wrong with attacks:
https://i.imgur.com/Dv7eJlP.jpg

There is a 10% scale here as well which is ok, but the starting value is 70% which is too high. I've read design principles in the past that aim for a lower ~60% chance of success. I'd be fine with 60% scaling to 70%, but 70 to 80 feels too high and makes options like advantage much less valuable.
I don't expect this to be an issue of variety. Perhaps I've missed something else here.

I'm off to bed for the night, thanks for the discussion that ended up in the math of the system and hopefully I can understand what's going on with chance to hit tomorrow.

Here are the numbers I get, assuming standard array and maxing attack stat as quickly as possible.
The columns are CR values, the rows are PC levels. The values are hit-chances as decimals. I believe I was using the DMG values for the ACs, which my data says are pretty good (+- 1 or 2) at most CRs above 0.5 and below 20. CRs above 20 tend to be above guidance for ACs, pretty badly.

https://www.admiralbenbo.org/images/misc/hit-chance.jpg

As you can see, the starting range is 65% (which I feel is just right). There's a banded structure, with levels 1-3, 4, 5,6-8, 9-12, 13-16, and 17-20 being separate bands with no movement. Each band (except levels 4, 6-8, and 17+) is 3 CR wide. Band 17+ is the rest of the CRs, because the DMG values don't increase above 19. This would change if using the real numbers. My expectation is that they intended for "vanilla" parties (no feats/multiclassing/+X items) to mostly stick to CRs <= 20, with anything above that being really really hard by intent.

Spreadsheet link (https://1drv.ms/x/s!AjKe-YTGxfZWw3ZeCi2AxTkOY7uT) from where I pulled that--there's also defensive data (being-hit chance [1-miss chance]) for the Basic Rules characters I used as test-beds).

Kadesh

2019-01-29, 09:29 PM

Which point?

80% chance of being hit being unreasonable?
or
Monsters have higher HP and AC, not just higher HP. So why should PCs only have higher HP and not AC that scales at the same rate?
You could say that given that an AC of 18 is possible by level 1, that PCs are already ahead of the curve at that point.

You need to explain why you think Monsters and PC's need to be build the same.

Malifice

2019-01-30, 01:03 AM

Which point?

80% chance of being hit being unreasonable?
or
Monsters have higher HP and AC, not just higher HP. So why should PCs only have higher HP and not AC that scales at the same rate?

Because the game is designed so that low CR monsters remain a threat to even 20th level PCs.

Its not designed so that high CR monsters can be threatened by hordes of low level PCs.

Under your system PC ACs of 25-30 will be common at mid to high levels (even higher in the case of magic item heavy campaigns). A 20th level unarmored Barbarian with Bracers of defence, a +1 shield, and a Dex and Con of 20 is rocking an AC of around 30. His Archmage buddy in robes of the Archmage, at will shield spell and a Dex of 20 is spitting out an AC of about the same (and likely with disadvantage due to Foresight). If he also happened to be a bladesinger...

Meaning that low-medium CR'd monsters are back to needing a natural 20 to hit the PCs.

The game is designed so that at mid to high levels, AC becomes less meaningful as a method of defence for high CR monsters, as opposed to HP scaling. It retains its usefulness vs low to medium CR monsters however.

MaxWilson

2019-01-30, 11:28 AM

Because the game is designed so that low CR monsters remain a threat to even 20th level PCs.

Its not designed so that high CR monsters can be threatened by hordes of low level PCs.

Yes it is. Rodney Thompson's article mentioned this scenario explicitly. Low level PCs can fight high CR monsters by e.g. rallying the low level town guard. This is one of the explicit benefits of bounded accuracy (which is really just another name for old-school lack of number treadmills--hordes of creatures were threatening in AD&D too).

Malifice

2019-01-30, 12:43 PM

Yes it is.

No, its not.

The design intent was so that low CR monsters remain a viable threat (simply by increasing numbers) even as you advance in level.

AC is designed to be largely static as one advances in level, with HP inflation being the primary defensive 'advantage' one gets for advancing in level.

This takes much of the swingyness out of combat (beyond the first few levels, where AC is king), and makes encounters (and encounter design) much easier to predict.

The inverse being partially true is not central to the design of the game. Sure; in 5E 100 commoners/ skeletons/ goblins with bows could take on a Dragon thanks to the same phenomena, but the game isnt designed from the perspective of 100 commoners/ skeletons or goblins.

Removing the relatively static nature of AC means we wind up back at 1-3E where under CR'd threats rapidly become trivial in nature, and easily ignored (routinely needing natural 20's to hit).

The fact that high CR monsters tend to hit more often than not isnt a problem at high levels, as all classes have much more HP at high level, plenty of resources to recover HP (more HD, plus more healing magic and potions than they had at low level), and are much harder to outright kill from a large damage effect (needing at least 100-200 point of damage to kill from a single attack, and even then only if already badly wounded).

What the OP sees as a bug, is actually a feature, and a key pillar in the design of the system.

MaxWilson

2019-01-30, 01:15 PM

No, its not.

The design intent was so that low CR monsters remain a viable threat (simply by increasing numbers) even as you advance in level.

AC is designed to be largely static as one advances in level, with HP inflation being the primary defensive 'advantage' one gets for advancing in level.

This takes much of the swingyness out of combat (beyond the first few levels, where AC is king), and makes encounters (and encounter design) much easier to predict.

The inverse being partially true is not central to the design of the game. Sure; in 5E 100 commoners/ skeletons/ goblins with bows could take on a Dragon thanks to the same phenomena, but the game isnt designed from the perspective of 100 commoners/ skeletons or goblins.

Removing the relatively static nature of AC means we wind up back at 1-3E where under CR'd threats rapidly become trivial in nature, and easily ignored (routinely needing natural 20's to hit).

The fact that high CR monsters tend to hit more often than not isnt a problem at high levels, as all classes have much more HP at high level, plenty of resources to recover HP (more HD, plus more healing magic and potions than they had at low level), and are much harder to outright kill from a large damage effect (needing at least 100-200 point of damage to kill from a single attack, and even then only if already badly wounded).

What the OP sees as a bug, is actually a feature, and a key pillar in the design of the system.

I'll let the 5E designers speak for themselves here (emphasis mine):

It opens up new possibilities of encounter and adventure design. [B]A 1st-level character might not fight the black dragon plaguing the town in a face-to-face fight and expect to survive. But if they rally the town to their side, outfit the guards with bows and arrows, and whittle the dragon down with dozens of attacks instead of only four or five, the possibilities grow. With the bounded accuracy system, lower-level creatures banding together can erode a higher-level creature's hit points, which cuts both ways; now, fights involving hordes of orcs against the higher-level party can be threatening using only the basic orc stat block, and the city militia can still battle against the fire giants rampaging at the gates without having to inflate the statistics of the city guards to make that possible.

The fact that low-level creatures can threaten high-level creatures is a deliberate design feature of 5E.

What this means in practice is that you can stop worrying about the 3E mentality of constructing balanced encounters, focus on giving PCs a lot of agency and things to do, and let PCs manage their own difficulty. I've seen novice 5E players playing 3rd level characters successfully take on the equivalent of a 20th level encounter (64,000 XP IIRC worth of Neogis and Umber Hulks) due partly to this phenomenon, by recruiting a couple dozen NPC helpers and leading them in battle, exactly as Rodney Thompson predicted. It works as intended and it's lots of fun for the players.

MaxWilson

2019-01-30, 01:30 PM

There is definitely a numbers treadmill in 5e. It's not to the same degree as it was in 4e, but it is still there. Remove proficiency bonus from your main weapon and you'll see it rather quickly. ;)

When I say there is no numbers treadmill, I mean what Rodney Thompson says: there is no DM-side expectation that player to-hit bonuses continually rise as they increase in level. You'll notice that even Ancient Dragons tend to max out at around AC 22 (sans magic), low enough that you can still hit them even without proficiency bonuses. If you're fighting with a non-proficient weapon, you aren't a very good fighter, but you're still approximately as effective against an AC 22 Ancient Dragon as you are against the equivalent XP worth of AC 8 zombies. (I.e. not very.) And even at high levels you may fight either.

That's what I mean when I say there's no numbers treadmill. When you get better at doing things in 5E, you actually get better. The goalposts don't shift just because you went up in level.

What do you mean by claiming that there is a numbers treadmill? Are you just saying you use exclusively high-CR monsters?

Malifice

2019-01-30, 01:38 PM

What this means in practice is that you can stop worrying about the 3E mentality of constructing balanced encounters

Yeah, nah.

You play a different game to me.

KorvinStarmast

2019-01-30, 01:49 PM

A 20th level unarmored Barbarian with Bracers of defence, a +1 shield, and a Dex and Con of 20 is rocking an AC of around 30. 25 or 27.
Dex 20: +5
Con 20: +5 or +7 (A straight up barbarian turns a 20 into a 24 at level 20, right?)
Bracers; +2
Shield +1: +3

Add in a Ring of Protection +1, and a Cloak of Protection +1, and you get 27/29.

Primal Champion

At 20th level, you embody the power of the wilds. Your Strength and Constitution scores increase by
4. Your maximum for those scores is now 24.

Vogie

2019-01-30, 01:50 PM

The fact that low-level creatures can threaten high-level creatures is a deliberate design feature of 5E.

What this means in practice is that you can stop worrying about the 3E mentality of constructing balanced encounters, focus on giving PCs a lot of agency and things to do, and let PCs manage their own difficulty. I've seen novice 5E players playing 3rd level characters successfully take on the equivalent of a 20th level encounter (64,000 XP IIRC worth of Neogis and Umber Hulks) due partly to this phenomenon, by recruiting a couple dozen NPC helpers and leading them in battle, exactly as Rodney Thompson predicted. It works as intended and it's lots of fun for the players.

Matt Colville has a great video about this - I believe it's the one where he's talking about the lessons that 4e taught WotC.

The reason for bounded accuracy was NOT a math one... rather an active design choice for the player experience.

The general idea is that the actual act of hitting is where the fun lies. Missing, on the other hand, is not fun.
The way 5e was designed is that there are very few changes to your AC or Attack rolls on a level to level basis, staying relatively static, while hit points inflate very quickly. Even in the posts above, the AC does fluctuate with monsters, but only within a couple points.

In 4th Edition, it was the introduction of the Minion mechanic. If applied to 5e, it'd be like this: an Ogre is a CR 2 monster with +6 to hit, 11 AC and 59 hit points, but an Ogre Minion has +6 to hit, 11 AC and... 1 hit point. They do the same damage, are equally as dangerous when standing in the background throwing javelins, but as soon as a PC engages them, they start to fall quickly, giving the player very cinematic feeling of general bad@ssness.

In 5th Edition, there's something very equivalent. For example, I'm running a group through Waterdeep: Dragon Heist, a published adventure, and the initial fight, with level 1 PCs is against a Troll, which is a CR 5 monster, as well as 3 Stirges. It would be impossible, but the troll starts at Half health, and they are assisted by a strong NPC who can fight the troll alone to a standstill, requiring the party to assist him. At the end of W:DH, if played as written, there is a massive fight that will be all but impossible for the PCs to win, unless they've followed the political intrigue and Renown-gathering of the various Waterdhavian factions, which turn the tides in the favor of the Party using droves of named and unnamed NPCs.

MaxWilson

2019-01-30, 02:24 PM

Yeah, nah.

You play a different game to me.

I know. I play CAW and you play CAS.

25 or 27.
Dex 20: +5
Con 20: +5 or +7 (A straight up barbarian turns a 20 into a 24 at level 20, right?)
Bracers; +2
Shield +1: +3

Add in a Ring of Protection +1, and a Cloak of Protection +1, and you get 27/29.

Primal Champion

Defensive Duelist can add another +6ish, as long as you're only taking three-ish attacks per round. (Any more than that and you probably won't have your reaction free to block the extra hits, reducing your average effective AC bonus from +6ish to +4ish or less.)

In 5th Edition, there's something very equivalent. For example, I'm running a group through Waterdeep: Dragon Heist, a published adventure, and the initial fight, with level 1 PCs is against a Troll, which is a CR 5 monster, as well as 3 Stirges. It would be impossible, but the troll starts at Half health, and they are assisted by a strong NPC who can fight the troll alone to a standstill, requiring the party to assist him. At the end of W:DH, if played as written, there is a massive fight that will be all but impossible for the PCs to win, unless they've followed the political intrigue and Renown-gathering of the various Waterdhavian factions, which turn the tides in the favor of the Party using droves of named and unnamed NPCs.

Heh. I just wanted to remark on the absurdity of WotC writing a troll, of all things, to start a fight at half health. How contrived does the situation have to be for that to be true? Did the troll just take a bath in acid?

Unoriginal

2019-01-30, 02:44 PM

Heh. I just wanted to remark on the absurdity of WotC writing a troll, of all things, to start a fight at half health. How contrived does the situation have to be for that to be true? Did the troll just take a bath in acid?

DRAGON HEIST SPOILERS

There is nothing absurd about it, and "how contrived" the situation is: very little. The troll was attacked by a whole bunch of striges and they started sucking his blood without him being able to defend himself efficiently (due to being busy climbing the inside of a deep pit) , just before he shows up.

The adventure makes clear that if the PCs don't take care of the regeneration power ASAP the troll gets to full health quickly and then things get ugly.

Trolls can take beatings too, and even without that their regeneration doesn't require them taking "a bath in acid" to be blocked, anyone with a torch could do it.

MaxWilson

2019-01-30, 02:52 PM

DRAGON HEIST SPOILERS

There is nothing absurd about it, and "how contrived" the situation is: very little. The troll was attacked by a whole bunch of striges and they started sucking his blood without him being able to defend himself efficiently (due to being busy climbing the inside of a deep pit) , just before he shows up.

The adventure makes clear that if the PCs don't take care of the regeneration power ASAP the troll gets to full health quickly and then things get ugly.

Trolls can take beatings too, and even without that their regeneration doesn't require them taking "a bath in acid" to be blocked, anyone with a torch could do it.

Yeah, that's pretty contrived. The troll would have to have fallen in the hole and been attacked by stirges just minutes before the PCs showed up.

Unoriginal

2019-01-30, 02:59 PM

Yeah, that's pretty contrived. The troll would have to have fallen in the hole and been attacked by stirges just minutes before the PCs showed up.

No, because the troll is climbing up the hole, with the stirges attacking him while he climbed, and he reaches the place where the PCs are while still under attack by the stirges (reaching said place being the troll's goal).

NPCs have events happening to them independently of the PCs' presence, and their agenda can lead them into troubles which weaken them or into favorable circumstances that make them more dangerous. That the PCs are only one of many moving pieces of the plot, and the others aren't waiting for them, is one of Dragon Heist's themes, by its construction.

Max_Killjoy

2019-01-30, 03:04 PM

No, because the troll is climbing up the hole, with the stirges attacking him while he climbed, and he reaches the place where the PCs are while still under attack by the stirges (reaching said place being the troll's goal).

NPCs have events happening to them independently of the PCs' presence, and their agenda can lead them into troubles which weaken them or into favorable circumstances that make them more dangerous. That the PCs are only one of many moving pieces of the plot, and the others aren't waiting for them, is one of Dragon Heist's themes, by its construction.

Here's the crucial question:

Why is the troll climbing up the hole? Is there a trigger related to the PCs being there, or is it just a coincidence that of all the infinite moments ever, he just happens to climb up at the moment the PCs are there?

E: Just realized this sidebar is barely related to the thread topic, thought I clicked on something else, so maybe we should drop it.

MaxWilson

2019-01-30, 03:05 PM

No, because the troll is climbing up the hole, with the stirges attacking him while he climbed, and he reaches the room where the PCs are while still under attack by the stirges (reaching said room being the troll's goal).

NPCs have events happening to them independently of the PCs' presence. That the PCs are only one of many moving pieces of the plot, and the others aren't waiting for them, is one of Dragon Heist's themes, by its construction.

Yes, because it doesn't take long to climb out of a hole, nor to recover from a stirge feeding. If a troll falls in a hole and gets eaten alive by 30 Stirges, well, guess what? Ten minutes later all the stirges are full, the troll is fine, and he's no longer in a hole. If there are hundreds of stirges he will take longer to recover, but in that case he won't be at half health either, he'll be at zero.

To quote a certain web comic (https://www.shamusyoung.com/twentysidedtale/?p=615), "That is a very specific level of tired."

So yes, it's contrived.

Unoriginal

2019-01-30, 03:16 PM

Here's the crucial question:

Why is the troll climbing up the hole? Is there a trigger related to the PCs being there, or is it just a coincidence that of all the infinite moments ever, he just happens to climb up at the moment the PCs are there?

Mostly a coincidence. The troll was climbing to get a meal out of the people in the room above the pit, but the PCs are far from the only ones present.

Yes, because it doesn't take long to climb out of a hole, nor to recover from a stirge feeding. If a troll falls in a hole and gets eaten alive by 30 Stirges, well, guess what? Ten minutes later all the stirges are full, the troll is fine, and he's no longer in a hole. If there are hundreds of stirges he will take longer to recover, but in that case he won't be at half health either, he'll be at zero.

The troll didn't fall into a hole, and it was never implied. He was in the cave below the pit and climbed up to reach where the PCs are, and got attacked by the stirges in the pit.

Also, what "ten minutes later"? I literally said that the stirges are still attacking the troll when he reaches the room at the top of the pit and the fight with the PCs has a chance of happening. It's immediate.

It's no more contrived than any doomsday clock scenario happening when the PCs are close enough to intervene or any bandit group deciding to ambush the one caravane the PCs are guarding.

Anyway, I'm stopping talking about this from now on. Sorry for the off-topic.

Malifice

2019-01-30, 03:24 PM

No, because the troll is climbing up the hole

https://www.youtube.com/watch?v=CtOEig1l8SA

Couldn't resist.

MaxWilson

2019-01-30, 03:40 PM

Also, what "ten minutes later"? I literally said that the stirges are still attacking the troll when he reaches the room at the top of the pit and the fight with the PCs has a chance of happening. It's immediate.

I said "The troll would have to have fallen in the hole and been attacked by stirges just minutes before the PCs showed up." and you said, "No, because [reasons]." What am I supposed to infer from that if not that it wasn't a matter of minutes?

Unoriginal

2019-01-30, 03:47 PM

I said "The troll would have to have fallen in the hole and been attacked by stirges just minutes before the PCs showed up." and you said, "No, because [reasons]." What am I supposed to infer from that if not that it wasn't a matter of minutes?

You were supposed to infer that is wasn't a matter of minutes when you read the

and he reaches the place where the PCs are while still under attack by the stirges

part of what I wrote.

Again, it's not any more contrived that "the adventurers walk on the road and see someone getting attacked by bandits/monsters, what will they do?". The only difference here is that the one who gets attacked was also an enemy and it happened while he was moving toward the PCs.

Anyway, I won't comment on this subject in this thread again, for real this time.

KorvinStarmast

2019-01-30, 04:02 PM

Defensive Duelist can add another +6ish, as long as you're only taking three-ish attacks per round. I did a somewhat detailed examination of Unarmored Barbarian at 20 AC a few years back (https://rpg.stackexchange.com/a/67223/22566) ... but it isn't the final word. I went with the barbarian Unarmored, others tried a different approach.

Astofel

2019-01-30, 05:41 PM

The reason for bounded accuracy was NOT a math one... rather an active design choice for the player experience.

The general idea is that the actual act of hitting is where the fun lies. Missing, on the other hand, is not fun.
The way 5e was designed is that there are very few changes to your AC or Attack rolls on a level to level basis, staying relatively static, while hit points inflate very quickly. Even in the posts above, the AC does fluctuate with monsters, but only within a couple points.

This is basically it. Hitting is always more fun than not hitting. While sure, the players would like to not be hit, the DM is part of the game too and it's not fun for them if their monsters are only hitting 50% of the time. Also, adding proficiency to AC is a bad idea. Let's apply that to your typical sword-and-board fighter with the Defense fighting style wearing full plate. He has AC 27 at 20th level. That's high enough that he can now only be hit by a crit from lower-level monsters. It's also higher than anything in the Monster Manual, and even the Tarrasque itself will miss 35% of the time. If that fighter also happens to be an eldritch knight with the shield spell then Big T's miss chance goes up to 60%, and iirc Big T has the biggest to-hit bonus in the game. This fighter, who has no AC-boosting magic items and has only made a few simple, reasonable choices about his build, is now basically untouchable to direct attacks from most of the things he'll fight. Sure, maybe his saves aren't the best, but now remember that the same level of AC could also be achieved by, say, a paladin/sorcerer with charisma to saves and access to Shield of Faith for even more ridiculous.

5e AC gets less effective at high levels by design. As you get better at fighting things, you can fight stronger things, and those stronger things are better at hitting you than the things you were fighting before. If AC scaled so you always had a ~60% chance of being hit there'd be basically no point in leveling up, high-level monsters would cease to be threatening and become necessary to even touch the party, and the came would become another numbers treadmill.

I actually ran this encounter not too long ago and I fail to see how it's contrived. The encounter takes place in the Yawning Portal, which is famously built on top of a dungeon. The troll hasn't fallen into the hole prior to the encounter, it lives down there and has decided to come out to snack on taverngoers. This is apparently a regular enough occurrence that Durnan keeps his greatsword under the bar just in case. On its way up, the troll is attacked by an unknown number of stirges (it arrives with 9, but could easily have been attacked by more on its way up). Either because it's climbing or because it has regen and doesn't care the troll doesn't attack the stirges and allows itself to be drained to half health, at which point it finally makes it into the Yawning Portal and gets its butt handed to it by Durnan.

I guess you could argue that it's contrived because the PCs just so happen to be there when the troll shows up, but by that logic every story ever is contrived because the characters just so happen to be where the story is.

MaxWilson

2019-01-31, 01:45 PM

PCs gain a minimum of +4 to-hit over their career from proficiency bonus. Proficiency bonus is the lesser treadmill.

I feel like we're talking past each other. You're misunderstanding what I said.

Kryx

2019-01-31, 01:50 PM

Sorry for the delay, had an unexpected board game event last night.

=================

As you can see, the starting range is 65% (which I feel is just right).
I'm not 100% sure how to read your scale, but it seems to show the same issue on yours as mine. If you look at the left column (which I presume is level) at 20 and the first row (enemy CR I assume?) at 9 the chance to hit is 80%. Using your CR of 9 at level 20 results in the exact same issue as the one I point out with my image that you quoted. That being 80% chance to hit is too high.

While I think my CR expectation of +-4 was flawed, I believe the CRs you've chosen are also not the best option. I'm not too concerned with trash mobs and the balance of fights against them - they'll mostly be stomps anyways. I'm more concerned with more scripted fights who will likely have higher than average CRs. Enemy CR equal to level*0.75 seems like a better basis to determine balance.

In that system and without rounding to hit to the nearest 5% we end up scaling from 67% to 72% chance to hit. Plate scales from 45% chance to be hit to 69% chance to be hit. Chance to hit is good. Chance to be hit is slightly too high, but likely not worth tinkering with.

===================

there is no DM-side expectation that player to-hit bonuses continually rise as they increase in level.
PCs gain a minimum of +4 to-hit over their career from proficiency bonus. Proficiency bonus is the lesser treadmill.

===============

Bracers; +2
Shield +1: +3

Add in a Ring of Protection +1, and a Cloak of Protection +1, and you get 27/29.
5e has no magic item expectations. So we have either 1 of 2 results:

In a non-magic game player ACs become too low
In a magic game player ACs become too high

Both can't be at an equal math level.

PhoenixPhyre

2019-01-31, 01:57 PM

Sorry for the delay, had an unexpected board game event last night.

=================

I'm not 100% sure how to read your scale, but it seems to show the same issue on yours as mine. If you look at the left column (which I presume is level) at 20 and the first row (enemy CR I assume?) at 9 the chance to hit is 80%. Using your CR of 9 at level 20 results in the exact same issue as the one I point out with my image that you quoted. That being 80% chance to hit is too high.

While I think my CR expectation of +-4 was flawed, I believe the CRs you've chosen are also not the best option. I'm not too concerned with trash mobs and the balance of fights against them - they'll mostly be stomps anyways. I'm more concerned with more scripted fights who will likely have higher than average CRs. Enemy CR equal to level*0.75 seems like a better basis to determine balance.

In that system and without rounding to hit to the nearest 5% we end up scaling from 67% to 72% chance to hit. Plate scales from 45% chance to 54% chance to be hit. That seems completely acceptable.

===================

There are no trash mobs in 5e. That's an MMO conceit. I harp on this because this is the fundamental mindset-shift you need to make to DM 5e well. The first commandment of 5e encounter balance is thou shalt have at between 0.67 and 2 enemies per PC. The second, by the way, is "thou shalt have multiple short rests per long rest and average more than one encounter between rests.

Group fights are much more dangerous than solos, even if the solos are legendary and CR = level + X. If you can't accept that, then the rest of your analysis is on shaky ground. Action advantage is just that strong. And the system math does not assume solo fights at all. They're not part of the calculations. So analyzing them is totally pointless and in fact distorts the picture. It's like analyzing a fork as a jackhammer. You're asking it to do something it wasn't designed to do.

Big strong solo enemies are more often curbstomps than are group encounters. A group of level 15s can take an ancient white dragon pretty easily, especially if it's the only fight that day. And when they're not curbstomps they're total routs or TPKs due to the necessary difficulty.

A good game can go 100% with never fighting things of CR = Level or higher.

My calculations are exactly pulled from the guidance from Xanathar's on encounter balance. That is, they are the assumptions of the designers.

From the picture I shared, a level 20 character has a 65% chance to hit anything of CR 17+. Less than that, because the higher level monsters depart from the AC guidance upward. As a side note, worrying about levels 17+ is rather pointless. Few games spend more than a few sessions at those levels if they get there at all.

Hitting isn't bad. Hitting is fun. The whole point of being able to use lower level mobs is that you can hit them easily (which makes you feel powerful), but you can't kill them in one hit and they can still hit you, so large packs of them are still dangerous. It's asymmetric on purpose. Because hitting is fun. At high levels you should feel different than at low levels, otherwise you're on a treadmill. So against those CR 9's you've been fighting since level 8 or so, you should feel quite a qualitative difference. But they're still a threat.

Kryx

2019-01-31, 02:10 PM

My calculations are exactly pulled from the guidance from Xanathar's on encounter balance. That is, they are the assumptions of the designers.
And yet players can fight monsters higher than those guidelines. I'm far more concerned about those fights than random monsters that don't put up much of a challenge.

As a side note, worrying about levels 17+ is rather pointless. Few games spend more than a few sessions at those levels if they get there at all.
And here is one of the cruxes of the problem. I'm concerned about high level balance which is where this problem is most manifest. Others never get there. For those games there isn't a problem.

Hitting isn't bad. Hitting is fun.
Hitting is fun, I agree. Hitting 80% of the time is not fun. Nor is being hit 80% of the time. Your CR guidelines lead to math that is something that I do not consider fun.

So against those CR 9's you've been fighting since level 8 or so, you should feel quite a qualitative difference. But they're still a threat.
And those lower CR monsters would still be a threat if PCs had +2 AC at 17-20 which would align with magic items and signifcantly improve being hit 80% of the time.

Snails

2019-01-31, 02:26 PM

Hitting is fun, I agree. Hitting 80% of the time is not fun. Nor is being hit 80% of the time. Your CR guidelines lead to math that is something that I do not consider fun.

That non-low level heroes sometimes fight very martially powerful creatures that hit most of the time has been a D&D thing since forever. That PCs who specialize in martial skills hit most of the time has been a D&D thing since forever. AC superstacks have been a theoretical path in all editions, but something that never showed up in most campaigns, for many practical reasons.

More than any other edition, 5e gives the DM and the Players a pallet of choices. For example: if there is one very strong monster and some mooks, how do the PCs match up? Does the high AC PC do a holding battle against mooks, while everyone else focuses on the big monster? Or do the PCs clear out the mooks and hope the big monster does not corner one PC and shred him up? Or do you match up ACs to the perceived threat to spread the risk, and hope average luck wins you the battle?

PhoenixPhyre

2019-01-31, 02:27 PM

And yet players can fight monsters higher than those guidelines. I'm far more concerned about those fights than random monsters that don't put up much of a challenge.

That's where you're wrong. Totally, absolutely, 100% wrong. Those "random" monsters are much more threatening than the big ones. Not alone, but in the proper numbers. Solo fights are curbstomps on one side or the other. They're not part of the system math at all. They're literally outside the scope of the design. Yes, you can fight them. No, they're not important. At all.

And here is one of the cruxes of the problem. I'm concerned about high level balance which is where this problem is most manifest. Others never get there. For those games there isn't a problem.

Hitting is fun, I agree. Hitting 80% of the time is not fun. Nor is being hit 80% of the time. Your CR guidelines lead to math that is something that I do not consider fun.

And those lower CR monsters would still be a threat if PCs had +2 AC at 17-20 which would align with magic items and signifcantly improve being hit 80% of the time.

The PC -> monster and monster -> PC trends are asymmetric on purpose. At level 20, a high-AC PC should only get hit 70% of the time by CR 30 monsters. And those are once in a dozen campaigns. Even ancient dragons (CR 24) are only hitting them 65% of the time. At those levels, PCs are hitting everyone 65% of the time or better.

Yes, the low-AC characters are getting hit 80% of the time. That's by design.

Not only that, but after CR 10-ish, attack rolls become less important (relatively) compared to saves.

Unoriginal

2019-01-31, 02:28 PM

And yet players can fight monsters higher than those guidelines. I'm far more concerned about those fights than random monsters that don't put up much of a challenge.

The assumptions of the designers are "random monsters that don't put up much of a challenge"? Dismissive much?

Also, if you are calculating for monsters higher than the guidelines, then it is perfectly normal for the AC to not match the guidelines.

You're deliberately choosing to analyse something that is out of the expected numbers by design, and your conclusion is "it is out of the expected numbers, they messed up".

GlenSmash!

2019-01-31, 02:44 PM

What this means in practice is that you can stop worrying about the 3E mentality of constructing balanced encounters, focus on giving PCs a lot of agency and things to do, and let PCs manage their own difficulty.

For what it's worth I take a CAS mentality and still stopped worrying about this. I stopped comparing encounters to dmg guidelines or using one of those handy online encounter builders. I eyeball it.

Some of my encounters end up being really easy for the party, some not so much. And it's rarely the ones I thought it was gong to be.

I've just accepted that Players can make poor choices which lead to negative consequences of which I find PC death to typically be the least interesting. So i jsut try to make those negative consequences still lead to interesting choices for the players.

Ultimately the change in philosophy has been very freeing for me.

To quote a certain web comic (https://www.shamusyoung.com/twentysidedtale/?p=615), "That is a very specific level of tired."

I'm due for another re-read of DMotR. It's brilliant.

Theodoric

2019-01-31, 02:52 PM

There's a big edition-gap here. One can't apply 3e logic to 5e situations. Survivability in 5e is much more about hit dice than it is about other stats, as they just don't scale as much (including AC; as a DM you need to watch out when it comes to giving players gear to exceed ACs of 22, as they'd be rivalling Demon Lords at that point). Even basic CR 1/2 Hobgoblins stay a meaningful threat in decent numbers up to level 10.

Kryx

2019-01-31, 02:57 PM

Yes, the low-AC characters are getting hit 80% of the time. That's by design.
And plate wearers getting hit 80% of the time in a scenario against an equivalent level monster is also by design. The same is not true until level 13. Being hit 80% when your character is designed around not getting hit is not something I'd consider good design.

The PC -> monster and monster -> PC trends are asymmetric on purpose. At level 20, a high-AC PC should only get hit 70% of the time by CR 30 monsters.
Your numbers are way off here. A CR 30 monster has 19 to hit. It literally only misses on a 1 against plate armor and on plate armor+a shield.

==============

Also, if you are calculating for monsters higher than the guidelines, then it is perfectly normal for the AC to not match the guidelines.
PhoenixPhyre is using averages which take a whole adventuring day and result in an average of CR = 9. The PCs will fight above and below that average. In my experience monsters far below that average are not going to make or break an encounter and will not largely impact a player's enjoyment of the game. However when PCs enounter things above that average or even greatly above that average those events are generally far more memorable. If the main design of your class breaks down at that point then there is a problem.

Kryx

2019-01-31, 02:59 PM

There's a big edition-gap here. One can't apply 3e logic to 5e situations. Survivability in 5e is much more about hit dice than it is about other stats
Survivability in 5e is very much about armor until enemies reach about CR 13 when PC AC starts to mean significantly less as the CR raises.

So the issue is not an issue for all tiers of 5e, only high tiers.

MaxWilson

2019-01-31, 03:19 PM

PhoenixPhyre is using averages which take a whole adventuring day and result in an average of CR = 9. The PCs will fight above and below that average. In my experience monsters far below that average are not going to make or break an encounter and will not largely impact a player's enjoyment of the game. However when PCs enounter things above that average or even greatly above that average those events are generally far more memorable. If the main design of your class breaks down at that point then there is a problem.

I want to make sure I understand what you're actually saying here because this claim rings untrue. Either I'm misunderstanding what you're saying or it's just wrong, so let's get some clarity here with a specific example.

If you've got 4 15th level PCs having an average of (say) four Hard encounters in a one-day adventure, you could have one or more of any of the following:

Two CR 6 Invisible Stalkers and two CR 6 Vrocks (18,400 adjusted XP).
A CR 7 Giant Ape, a CR 8 Green Slaad, and a CR 7 Mind Flayer (19,400 adjusted XP)
A CR 17 Adult Red Dragon (18,000 XP)
Ten CR 3 Githyanki Warriors (18,800 XP)

If I understand your claim correctly, it looks like you're saying that because CR 3 is much less than level 15, the Githyanki warriors do not present a challenge, so the fact that AC 21 plate armor is good against them is nigh-irrelevant. I think you are saying that only the CR 17 Adult Red Dragon presents a real challenge for the players? If that's what you are saying, it's very wrong, but I can't see what else you could be saying instead.

It looks like you're conflating monster CR and difficulty, but difficulty is a product of both CR and quantity. If your analysis pretends that AC 21 is not useful at 15th level because CR 17 Adult Red Dragons have +14 to hit, your analysis is wrong.

I keep asking you how you construct your encounters--do you use exclusively high-CR monsters? You haven't answered that but it's starting to look like the answer is yes. Try using some hordes.

Survivability in 5e is very much about armor until enemies reach about CR 13 when PC AC starts to mean significantly less as the CR raises.

So the issue is not an issue for all tiers of 5e, only high tiers.

High tiers != high-CR solo encounters.

A dozen fire giants is a high-tier encounter, and so is a hundred Githyanki Warriors led by four Gishes. AC is useful in both cases, especially if you pair it with e.g. Defensive Duelist + the Blur spell. (You go from taking 55.3 points of damage per round per Fire Giant at AC 13 to taking only 5.15 DPR per Fire Giant at AC 26 + disadvantage. AC is cutting your damage taken by 90%!) As long as your high tier adventures have a mix of encounter types instead of all solos, AC will remain useful in many encounters.

To the extent that AC becomes useless at high levels in your games, it's because you created that problem by the encounters you construct. It's like you're claiming that Intelligence saves are overpowered because you use lots of Mind Flayers. Yeah, maybe it's an issue in your game, but it isn't in general.

Kryx

2019-01-31, 03:47 PM

If I understand your claim correctly, it looks like you're saying that because CR 3 is much less than level 15, the Githyanki warriors do not present a challenge, so the fact that AC 21 plate armor is good against them is nigh-irrelevant. I think you are saying that only the CR 17 Adult Red Dragon presents a real challenge for the players?
No, this is not what I've said. I'll use your example to illustrate:

Githyanki Warriors have 50 hp. They generally aren't going to live long against level 20 characters who do ~50-60 DPR. I don't want Githyanki to be useless, but unless used in large numbers their impact will be relatively small.

An Adult Red Dragon has 256 hp and is surely a larger and generally more memorable experience for PCs.

The Adult Red Dragon hits a Plate wearing Fighter 85% of the time. At that point the Fighter has lost a significant portion of its identity. That's a problem.

I keep asking you how you construct your encounters
I generally port encounters from pathfinder and then clean them up based on encounter calculators.

To the extent that AC becomes useless at high levels in your games, it's because you created that problem
I'm sorry, should I not use any monster above CR 13? That's what it sounds like you're suggesting.

It's like you're claiming that Intelligence saves are overpowered because you use lots of Mind Flayers. Yeah, maybe it's an issue in your game, but it isn't in general.
Oh, I guess I can't use monsters with Intelligence saves either. I thought this was the argument of more monsters, not less.

the fact that AC 21 plate armor is good against them is nigh-irrelevant
21 AC is beyond what a Fighter or Barbarian (who doesn't max con) can achieve without magic items.

I think it's great that we focus on the issue though: What max AC is acceptable? I see all kind of mid 20s levels talked about in this thread, and yet without magic items 20 is barely possible. It seems many here are playing with +X magic items which surely solves the issue.

When not using magic items, a system that increases AC automatically could be used. That could lead to 20 AC for Fighters or 22 with a shield. Far within bounded accuracy.

EDIT:

AC 26 + disadvantage
Ya, you're definitely playing with +X magic items. Then this discussion is not really relevant for you. I'm discussing how the math works without magic items.

Arzanyos

2019-01-31, 03:55 PM

A fighter can hit 21 AC without magic items. Full Plate, Shield, Defense Fighting Style. And the number of 26+Disadvantage isn't with magic items either. Disadvantage comes from the Blur spell, as mentioned, and the extra 5-6 AC comes from Defensive Duelist, which let's you apply prof to AC as a reaction for one attack.

JoeJ

2019-01-31, 04:04 PM

Githyanki Warriors have 50 hp. They generally aren't going to live long against level 20 characters who do ~50-60 DPR. I don't want Githyanki to be useless, but unless used in large numbers their impact will be relatively small.

An Adult Red Dragon has 256 hp and is surely a larger and generally more memorable experience for PCs.

15 githyanki warriors is the same level of difficulty as an ancient red dragon for a party of 4 20th level characters; they're both hard encounters. Add 4 more to the githyanki (for 19 total) and the encounter becomes deadly. And that's assuming there are no higher CR leaders.

Unoriginal

2019-01-31, 04:05 PM

Kryx

2019-01-31, 04:07 PM

A fighter can hit 21 AC without magic items. Full Plate, Shield, Defense Fighting Style.
You're right, I forgot about defensive fighting style. However we shouldn't assume most fighters take Shield & Defense. I'd wager that's ~20 of fighters.

And the number of 26+Disadvantage isn't with magic items either. Disadvantage comes from the Blur spell, as mentioned, and the extra 5-6 AC comes from Defensive Duelist, which let's you apply prof to AC as a reaction for one attack.
Non-EK Fighters do not have access to blur and expecting casters to cast a concentration spell on a Fighter is not a reasonable assumption. Defensive Duelist is quite niche. Quite strong, but generally more of a Rogue feat than a Fighter feat.

I concede what you've written (minus blur) is possible, but very unlikely for the Fighter. I'm more concerned about the regular paths than niche edge cases like this.

Kryx

2019-01-31, 04:09 PM

How is the dragon stripping the Fighter of their identity by hitting them?
One of the Fighter's identites is being well armored. Against higher CR creatures armor means very little without magic items and spells.

Alternatively, a Barbarian does not have this issue as it's identity is not based just on armor, but also resistances. So when armor fails for it then its resistances push it through.

Well, it's not surprising that you think there is something wrong with 5e encounters, then.

5e works fundamentally differently than Pathfinder, and the encounter design is very different.

If you port a 5e module into Pathfinder, the encounters would be all wrong, too.
I think you've missed the part where I said "I clean them up". Encounter calculators are used to adhere to 5e. It's entirely unrelated to this post.

Unoriginal

2019-01-31, 04:09 PM

Non-EK Fighters do not have access to blur and expecting casters to cast a concentration spell on a Fighter is not a reasonable assumption.

How is it not a reasonable assumption? The game assumes you are playing a team of adventurers working together.

One of the Fighter's identites is being well armored. Against higher CR creatures armor means very little without magic items and spells.

Alternatively, a Barbarian does not have this issue as it's identity is not based just on armor, but also resistances. So when armor fails for it then its resistances push it through.

Also, Fighters stay well-armored even if powerful monsters hit them often. They're still hit less often than the not-well-armored people.

I think you've missed the part where I said "I clean them up". Encounter calculators are used to adhere to 5e. It's entirely unrelated to this post.

Depends what you mean by "cleaning them up", but cleaning up using an encounter calculators usually don't imply that you're changing the nature of the encounter, just the numbers. And the nature of PF encounters work for PF, not for 5e.

Kryx

2019-01-31, 04:12 PM

How is it not a reasonable assumption? The game assumes you are playing a team of adventurers working together.
So you're telling me that a Fighter is only viable at one of its identities against high CR enemies if it gets buffed from its allies?

Come on...

What about the other party members that are pretty much auto-hit as well? Should all the casters blur all the martials? That's the solution?

This issue doesn't exist at all at CR <13.

Unoriginal

2019-01-31, 04:16 PM

So you're telling me that a Fighter is only viable at one of its identities against high CR enemies if it gets buffed from its allies?

Come on...

This issue doesn't exist at all at CR <13.

I'm telling you that a team of adventurers fighting a solo powerful monster as a team is an assumption of the game and its designers.

You dismissed a perfectly valid tactic for difficult encounters as not likely to happen... for some reasons. Please explain why, rather than accusing others of saying the Fighter can't help without buff (which is ridiculous).

PhoenixPhyre

2019-01-31, 04:17 PM

Kryx, here's (another) thing you're missing.

Being hit isn't in and of itself a problem. Expected damage intake is what matters for balance purposes. Epic monsters should hit you all the time, for a few reasons.

1) If they didn't hit non-magic-armor folks all the time, they'd be a joke against anyone with magic armor. There's your +2 AC--the designers figured that if the DM is sending you against threats way above your "normal" power level, he'll also let you get equipped properly. Sending a no-magic party against an Ancient Red requires them to not face it head on. And that's CLOSED_WON'T_FIX: BY DESIGN.

2) Expected DPR is tricky. If they didn't hit very frequently, they'd need to hit like trucks when they did hit[1]. Which causes rocket tag issues--whiff-whiff-whiff-dead. And 5e's designers wisely (IMO) decided that was a stupid idea. 5e is built around a lot of little hits. The big difference here with the big solo monsters (CR 20+, of which there are 13 in the MM and Volo's, 8 of those are the ancient dragons, and one (the lich) is only that high in his lair and doesn't use attack rolls much at all) is that they don't have a group of buddies to smooth out the damage intake. Steady damage doesn't kill (within one fight). Damage spikes do. It's why in MMO theorycrafting, avoidance is much less useful for effective health than mitigation, because mitigation is always on.

3) If you're facing something way above you in power, you should expect to miss a lot while they hit easily. The same goes in reverse--strong PCs vs weak monsters should hit a lot and get hit less frequently. So to feel epic, these big monsters better hit you more than someone on your same CR. That's entirely by design.

You're assuming that the 65% number is some decreed "design goal". It's not. It's a nice benchmark for multi-monster encounters. It doesn't apply to big solos (which CR 20+ are by necessity). Don't try to balance around those big solos--you'll end up breaking the rest of the game (which is the part that matters for 99.99999999999999999999999999% of the groups that at most ever face a single one of those, and usually have significant magic items when they do).

[1] Take the Ancient Red Dragon, the toughest thing you'll fight that's not a puzzle monster (like Big T).

He is only an offensive CR of 24. Because otherwise, he'd run a very real risk of one-round-KO a full-health, raging barbarian. Offensive CR maps very closely to the ability to one-round KO from full health a wizard of level X - 1 if everything connects. So CR X + 1 (+6 DPR, or +15 for CR 20+) can threaten to ORKO a "tanky" player. CR X + 4 is threatening to one-round kill a weak PC, from full health. And lucky streaks happen, even with only a 15% hit chance. So if you drop the expected hit rate to 65%, you'd have to pump up the worst-case DPR by 1.37x to keep the expected damage intake the same. And that means pumping out enough damage to insta-kill someone. And being insta-killed, no way out, is no fun. They abandoned rocket tag for a very good reason. Messing with this would only bring it back, and only against the players.

Kryx

2019-01-31, 04:19 PM

no one in this thread but you is rambling
This thread is a harsh reminder of why I rarely post here. People like you who treat other people like this.

I've had enough of the way you choose to treat other people trying to discuss and understand.

MaxWilson

2019-01-31, 04:27 PM

No, this is not what I've said. I'll use your example to illustrate:

Githyanki Warriors have 50 hp. They generally aren't going to live long against level 20 characters who do ~50-60 DPR. I don't want Githyanki to be useless, but unless used in large numbers their impact will be relatively small.

An Adult Red Dragon has 256 hp and is surely a larger and generally more memorable experience for PCs.

The Adult Red Dragon hits a Plate wearing Fighter 85% of the time. At that point the Fighter has lost a significant portion of its identity. That's a problem.

I don't feel like you illustrated anything at all--you didn't even answer the question. Do you or do you not feel that ten githyankis are a reasonable Hard encounter? When you say "large numbers" it's unclear whether you're saying "you need more githyanki" or "I usually don't use this many githyankis." Which one is it?

I'm sorry, should I not use any monster above CR 13? That's what it sounds like you're suggesting.

Er, no. I gave you an example of an adventuring day which included a CR 17 monster, which is > CR 13. How do you go from there to "don't use monsters above CR 13"? But 75% of the encounters were not > CR 13. There was a mix of encounter types.

Oh, I guess I can't use monsters with Intelligence saves either. I thought this was the argument of more monsters, not less.

*baffled look*

21 AC is beyond what a Fighter or Barbarian (who doesn't max con) can achieve without magic items.

False.

I think it's great that we focus on the issue though: What max AC is acceptable? I see all kind of mid 20s levels talked about in this thread, and yet without magic items 20 is barely possible.

This is just wrong. AC 21 is commonplace for anyone who cares about AC; AC 23 isn't too hard to reach with a first-level spell like Shield of Faith; AC 26-28 comes when you stack Shield on top of that; Defensive Duelist requires more feat investment than Shield but has unlimited uses.

It seems many here are playing with +X magic items which surely solves the issue.

You're jumping to false conclusions here. For example...

Ya, you're definitely playing with +X magic items. Then this discussion is not really relevant for you. I'm discussing how the math works without magic items.

Wrong. I am not referring to magic items. I barely use them in my games at all in fact.

=========================

A fighter can hit 21 AC without magic items. Full Plate, Shield, Defense Fighting Style. And the number of 26+Disadvantage isn't with magic items either. Disadvantage comes from the Blur spell, as mentioned, and the extra 5-6 AC comes from Defensive Duelist, which let's you apply prof to AC as a reaction for one attack.

Correct. Other ways to get to AC 26 + Disadvantage include Plate + Shield + Defense Style + Defensive Duelist + Dodge (if you're relying on Sentinel to prevent enemies from ignoring you, though that puts a double load on your reaction), or Plate + Shield + Defense Style + Warcaster + Shield + grapple/prone. Or you can impose disadvantage via Wrathful Smite, or Fear from a buddy.

Unoriginal

2019-01-31, 04:27 PM

GlenSmash!

2019-01-31, 04:31 PM

...alright, can you guys help me?

Not asking to take a side or anything, but where my posts in this thread insane?

I'm a poor judge of anyone's sanity, but I will take it as a good sign that you are willing to ask. I see few jerks with that kind of self awareness.

Still it's easy to get heated in an disagreement. (edit: for all parties involved)

Kryx

2019-01-31, 04:33 PM

You're assuming that the 65% number is some decreed "design goal". It's not.
I've tried to communicate this several times, so I'll try again:

If I, as a GM, wanted to have equivalent level CR monsters as all of my fights then I could do so without issue until around level 13 and CR 13. I don't play that way, but the math of 5e allows for that without issue. A PC hits for ~65% (varrying 5% at some levels) against those enemies.

Once CR 13 enemies hit that paradigm is lost and no longer possible. That's the entire thread in a nutshell.

Epic monsters should hit you all the time, for a few reasons.
At level 1 a CR 1 enemy does not hit you beyond a normal amount. Nor does a CR 3 enemy. The same is not true at the higher levels of the game where even enemies with CR = your level go way beyond the norm.

3) If you're facing something way above you in power, you should expect to miss a lot while they hit easily.
By "above your Power level" I assume you mean the CR guidelines you put above. I'm going to sound like a broken record here, but this is not true before CR 13.

MaxWilson

2019-01-31, 04:34 PM

Non-EK Fighters do not have access to blur and expecting casters to cast a concentration spell on a Fighter is not a reasonable assumption. Defensive Duelist is quite niche. Quite strong, but generally more of a Rogue feat than a Fighter feat.

This is backwards. Rogues already have a lot of demand on their reaction (Uncanny Dodge + strong opportunity attack), so if anything it's more of a fighter thing than a Rogue thing.

It is indeed pretty niche, but that's partly because there are better defenses than AC (including kiting with the Mobile feat or Polearm Master shove), and you requested we not talk about those other methods in this thread. In a thread about AC-focused builds at high levels, Defensive Duelist is not niche at all.

Unoriginal

2019-01-31, 04:36 PM

Well in any case, I apologize if I insulted you by calling bringing up identities in this manner and this context "rambling", Kryx. Hurting you was not my intent, and I am sorry I did it.

I am not dismissing the points I raised, still.

MaxWilson

2019-01-31, 04:40 PM

...alright, can you guys help me?

Not asking to take a side or anything, but where my posts in this thread insane?

I can admit I wasn't the kindest, but I don't see what I did that would have warranted that reaction.

Nope. The word choice "rambling" was a bit rude, but overall your posts in this thread have been mild and on-point. And I normally think you're kind of a jerk--I just don't think you're being a jerk right now, especially not in this post I'm quoting here.

PhoenixPhyre

2019-01-31, 04:42 PM

I've tried to communicate this several times, so I'll try again:

If I, as a GM, wanted to have equivalent level CR monsters as all of my fights then I could do so without issue until around level 13 and CR 13. I don't play that way, but the math of 5e allows for that without issue. A PC hits for ~65% (varrying 5% at some levels) against those enemies.

Once CR 13 enemies hit that paradigm is lost and no longer possible. That's the entire thread in a nutshell.

You can, but you're not following the system design assumptions. So blaming the design for what happens when you go outside those assumptions is rather uncharitable.

At level 1 a CR 1 enemy does not hit you beyond a normal amount. Nor does a CR 3 enemy. The same is not true at the higher levels of the game where even enemies with CR = your level go way beyond the norm.

That's because level 1 PCs are squishy enough already. Any increase in effective DPR would be lethal. At higher levels, this changes because the PCs have
1) much higher health to soak hits.
2) WAY more defensive abilities and spells. And no, this isn't symmetric, because monsters have fixed values. But pumping the DPR (rather than the hit rate) to counter this would lead to rocket tag, except where only the monsters had rocket launchers.

CR 13+ are supposed to be rare creatures. Note that you start getting legendary creatures at about this CR. These are the earliest solo's you're expected to fight, and then you're supposed fight them only occasionally. This is where your CR = Level assumption (which is totally alien to 5e, by the by) fails. That it hasn't failed earlier is just a fluke. It's been an invalid assumption all along.

By "above your Power level" I assume you mean the CR guidelines you put above. I'm going to sound like a broken record here, but this is not true before CR 13.

No, I meant CR >= level.
-----------------

Like it or not, 5e was not designed around solo monsters. Not at all. Not in any way. They're a decorative topping. The core math of 5e is in multi-monster encounters. Failing to accept that means that your entire assumptions are off base and not relevant as a criticism of the design. They may inform your (non-standard) play-style (which is fine), but they're not something wrong with the game. No more than there's something wrong with a pitchfork if it doesn't dig holes well.

Kryx

2019-01-31, 04:44 PM

you didn't even answer the question.
I'm trying to contain the thread via my responses. There are several points that I've brough up that are unanswered:

The math fundamentally changing around CR 13
Magic item expectancy from players (not necessarily designers). It seems many people use magic items.

20 Githyanki are surely a challenge, but that's not what the thread is about at all. The way you're asking this question seems to be trying to put words in my mouth that a Githyanki should not be a threat. That's not my belief at all. Githyanki is in a fine place.

What isn't in a fine place is the math fundamentally changing around CR 13, hence why I'm trying to focus the thread.

This is just wrong. AC 21 is commonplace for anyone who cares about AC
...
In a thread about AC-focused builds at high levels, Defensive Duelist is not niche at all.
This isn't about AC-focused builds. I'm seeking to understand the norm. The norm for a Fighter is one of several options:

TWF (very bad by RAW at higher levels)
GWM
Polearm
Polearm+GWM
S&B

S&B is a thing, but it is likely around 1/4 of fighters or less. Defensive duelist would work for the TWF and S&B, but not GWM and Polearm and Polearm+GWM. A Fighter shouldn't be expected to go full defensive to keep up with the math of the early game.

MaxWilson

2019-01-31, 04:46 PM

MaxWilson

2019-01-31, 05:01 PM

I'm trying to contain the thread via my responses. There are several points that I've brough up that are unanswered:

The math fundamentally changing around CR 13
Magic item expectancy from players (not necessarily designers). It seems many people use magic items.

I answered both of these already, the latter in the very post you quoted from: no, it doesn't change around CR 13, it already doesn't match your claim; and no, it's not about magic items. Quote:

Wrong. I am not referring to magic items. I barely use them in my games at all in fact.

You can hardly claim that your questions are going "unanswered" when you're quoting from a post which answers one of them.

Moving on...

20 Githyanki are surely a challenge, but that's not what the thread is about at all. The way you're asking this question seems to be trying to put words in my mouth that a Githyanki should not be a threat. That's not my belief at all. Githyanki is in a fine place.

*pulls out hair* I didn't say 20 Githyanki; a Hard Githyanki encounter at 15th level is only ten Githyanki. Twenty Githyanki would be a Deadly x2 encounter. You said something that sounded wrong to me, so I tried to ask for clarification by giving an example of four different Hard encounters using 5E math, and asking if you felt that the Githyanki one was too weak as you appeared to. You keep changing the question instead of answering it.

I think from your last two sentences that you are conceding that ten githyanki are a fine ("Hard"-which-means-not-really-hard) encounter for a 15th level party. So what's the beef then? If at level 15 you sometimes fight CR 17 dragons, and you sometimes fight CR 3 Githyanki, the AC 21 guy plays his role as the guy who's an armored tank against the psychic ninja githyanki, while the dragon gets to play the role of the fantastic beast which can cut through steel like a hot knife through butter--against a dragon, they either need to buff the fighter some more with spells, or change tactics.

The fact that the fighter's armor still works just fine is exactly what makes the dragon's ability to overcome it impressive!

You think this is a bad thing, but 5E's designers didn't. If you want to change it, then change it, but the simplest way to do that isn't to mess with the PCs' AC, it's to decrease the dragon's to-hit bonus. Your OP suggests giving PCs +Proficiency to AC, but since you're assuming monster CR = PC level, that's really just equivalent to not giving +Proficiency to monsters' to-hit. Rewrite your dragons to use raw STR bonus with no proficiency and your problem is solved.

What isn't in a fine place is the math fundamentally changing around CR 13, hence why I'm trying to focus the thread.

Pointing out that this assumption is wrong seems relevant. A first level Fighter PC with chain, shield, Defense style, a rapier, and Defensive Duelist has effective AC 21, so goblins and other first-level threats like skeletons are hitting him only ~20% of the time. As the monsters grow in CR that percentage grows from 20% to 70% or more. There's no fundamental change at level 13. It's been growing all along.

This isn't about AC-focused builds. I'm seeking to understand the norm. The norm for a Fighter is one of several options:

TWF (very bad by RAW at higher levels)
GWM
Polearm
Polearm+GWM
S&B

S&B is a thing, but it is likely around 1/4 of fighters or less. Defensive duelist would work for the TWF and S&B, but not GWM and Polearm and Polearm+GWM. A Fighter shouldn't be expected to go full defensive to keep up with the math of the early game.

If you're seeking to understand the norm, start with understanding the norm of how encounters are actually constructed. Your preference for using strictly high-CR solos is not the norm, and is the source of most of your discomfort.

PhoenixPhyre

2019-01-31, 05:03 PM

The paradigm (1) doesn't hold in the first place, (2) doesn't reflect this thread. The thread is about the PCs getting hit, not PCs hitting; and it's very easy for a PC to ensure (even at low levels) that monsters hit him far less than 65% of the time.

If you think monster to-hit can't be reduced below 65% past CR 13 ("that paradigm is lost and is no longer possible") I will tell you that's wrong. Look into ways to impose disadvantage, and look into defense-boosting reactions like Cutting Words, Shield, Bend Luck, Defensive Duelist, etc. Against "equivalent level CR monsters" (and therefore presumably solo monsters, unless you're going for Deadly+ encounters) those kinds of reactions are very strong. You opened this thread by asking "what am I missing?" and apparently one thing you're missing is all the defensive bells and whistles that high-level PCs can have, even without magic items.

Exactly. Kryx, you're not even missing the forest for the trees. You're missing the forest for a single leaf from a different forest entirely.

You're analyzing a tiny part and following a "trend" that only holds when it does because your assumptions have nothing to do with the game design assumptions. And then getting mad because those fall apart (when they only held together by happenstance anyway up to this point).

As another note, as someone with lots of experience here--solo monsters below level 13 are a joke. Really. Even if several CRs above the party. We fought a CR 5 as a level 2 party. Sure, if it hit it really hurt, but it died before it got its second action. The same goes for most things. Action economy is king. Trying to do CR = level is just plain foolish, and not part of the "supported" regime.

Just like you can play tennis with a two-by-four, but no tennis league in the world is going to support such a thing, 5e is under no obligation to support your "only CR = level solo monsters" play-style. Doing so is foolish for all concerned, and makes for boring fights that are either curbstomps or TPKs, with very little room between them.

Kryx

2019-01-31, 05:24 PM

blaming the design for what happens when you go outside those assumptions is rather uncharitable.
I agree with your sentiment, but I'm not suggesting breaking the design. I'm using it as a tool to highlight the issue. I'll do the same without breaking the design:

4 level 20 PCs can fight 1 CR 20 monster as a medium difficulty encounter. According the DMG 84 the adjusted XP per day per charcter is 40,000. 40,000*4 = 160,000. A CR 20 monster is 25,000 adjusted xp. That's 6.4 medium encounters in a day.

In these encounters, that are within the design system, the AC of PCs does not hold up to the +to hit of monsters as it does at earlier tiers.

That's because level 1 PCs are squishy enough already. Any increase in effective DPR would be lethal.
This is true at super low levels, which I'd consider poor design, but that's outside this topic. This is not so true as PCs advance in level. Level 5 for example. PCs have significantly more HP at level 5 or level 10 and the issue doesn't exist.

These are the earliest solo's you're expected to fight, and then you're supposed fight them only occasionally.
...
Like it or not, 5e was not designed around solo monsters.
Is this stated in the rules? I've primarily gone by the DMG rules and I don't see it on a quick skim, though it's been a long time since I've read them (I use encounter calculators), so I could've missed it.

MaxWilson

2019-01-31, 05:32 PM

Is this stated in the rules? I've primarily gone by the DMG rules and I don't see it on a quick skim, though it's been a long time since I've read them (I use encounter calculators), so I could've missed it.

It's stated by the designers, e.g. that Rodney Thompson post on bounded accuracy.

It's not stated in the rules because it's a design parameter, not a rule.

As I said before, if you want to use solos and you want them to stay at 65%ish to hit, you can just remove proficiency from high-level monster attacks, and increase their damage or HP to compensate, using the DMG tables.

PhoenixPhyre

2019-01-31, 05:35 PM

Is this stated in the rules? I've primarily gone by the DMG rules and I don't see it on a quick skim, though it's been a long time since I've read them (I use encounter calculators), so I could've missed it.

Stated? Not directly AFAIK. But it's the inevitable and obvious consequence of the design. That is, the design works great for multi-monster encounters and poorly for solos. Thus, we can assume either
a) the designers wanted you to fight mostly solos but were incompetent
OR
b) the designers wanted you to fight mostly group encounters and were competent.

The second makes much more sense, fits with the designer's pronouncements and philosophy and fits with how the published modules work as well. Why should we assume the uncharitable when the charitable fits the facts better?

Note that a solo monster gets an adjustment factor of 0.5. This means that solos are less powerful than the CR would otherwise indicate. What gets a modifier of 1.0? having 3-6 monsters. That's 0.6 to 1.5 monsters per PC (for 4 or 5 PC parties, which are the explicit design parameters).

Everything fits one way, and nothing fits except by accident the other. I think the design is clear, and it works for what it's designed for.

This is something you have to change, coming from 3e or PF. The underlying assumptions of combat design are very very different even though they use the same words (mostly). Let the system speak, don't force your assumptions onto it and think it's failing.

Edit: I'll also check XGtE when I'm back with my books. IIRC they have pretty explicit guidance that solos should be few and far between in their new "encounter balance" section (where I got the tables of PC:Monster(CR) ratios).

MaxWilson

2019-01-31, 05:44 PM

Note that a solo monster gets an adjustment factor of 0.5.

That's only for large parties, IIRC six or more PCs. For three to five PCs the multiplier for solos is 1x, two monsters is 1.5x, and three is 2x.

PhoenixPhyre

2019-01-31, 05:54 PM

That's only for large parties, IIRC six or more PCs. For three to five PCs the multiplier for solos is 1x, two monsters is 1.5x, and three is 2x.

You very well could be right. I'm AFB, so I can't check.

Kryx

2019-01-31, 05:57 PM

The thread is about the PCs getting hit, not PCs hitting
I was using PC chance to hit as a counterpoint in that PC chance to hit doesn't have the same scaling issue. My apologies for the confusing wording. Lets ignore it.

If you think monster to-hit can't be reduced below 65% past CR 13 ("that paradigm is lost and is no longer possible") I will tell you that's wrong. Look into ways to impose disadvantage, and look into defense-boosting reactions like Cutting Words, Shield, Bend Luck, Defensive Duelist, etc.
I agree that those are all great tools, but the vast majority of those tools are available at low levels. All of those options are available <= level 4. So they are not an explanation for potential higher level math problems (which both you dispute, I understand, but that's what the discussion is about).

it doesn't change around CR 13, it already doesn't match your claim
Compare chances to be hit of a level 12 vs a CR 12 and a level 20 vs a CR 20. Both can easily occur in a 4 PC vs 1 monster as a Medium difficulty encounter. It seems to exist to me.
(see the end of this post for other examples of CR being similar to level where the problem occurs)

You can hardly claim that your questions are going "unanswered" when you're quoting from a post which answers one of them.
Threads have a talent for branching out. I tried to bring up magic items several times and many people in the first page were talking about them and using them as explanations. The thread was moving quite quick when I made the most recent post about them not being answered. You don't use them, correct, but it seems several others do. And in that case this issue isn't as much of an issue.

You keep changing the question instead of answering it.
My apologies, that was not my intention at all. Ten Githyanki are a challenge, sure. But again I don't believe that is really relevant to the discussion as the issue I'm trying to discuss is against high CR monsters. I think you're trying to approach it from the encounter design angle, but high CR monsters can be appropriate encounter design.

Your OP suggests giving PCs +Proficiency to AC, but since you're assuming monster CR = PC level, that's really just equivalent to not giving +Proficiency to monsters' to-hit.
My OP uses it as a comparison. My OP was aimed to be against the wrong CR of enemies though so please disregard that.

Rewrite your dragons to use raw STR bonus with no proficiency and your problem is solved.
I don't want to rewrite every monster > CR 13 to have lower to hit. I'd rather fix the AC scaling if I use any fix. That depends on this discussion though. I want to make an informed decision - that's why I'm continuing the discussion.
AC formula - 1 + half prof doesn't seem too bad. It bumps PC AC by 1 at 9th and 17th level. That would mitigate the issue enough and wouldn't be any more of a math problem than +X magical items are, which many people use and the game is supposedly designed to accomodate for.
But I'm not set on that, I'm trying to be fully informed first.

====================

This is something you have to change, coming from 3e or PF. The underlying assumptions of combat design are very very different even though they use the same words (mostly). Let the system speak, don't force your assumptions onto it and think it's failing.
I think my admittance of porting PF modules (while fixing them up!) and the scenarios I used as examples (CR=Level) have misled how I actually play. I rarely use high CR monsters, but when I do use them I want the battle to be a good experience. I've had several of these battles where martial characters haven't enjoyed them due to the AC issues.

That is, the design works great for multi-monster encounters and poorly for solos.
The problem exists for solo+other creatures as well. Example:
4 level 20 PCs vs 1 CR 18 monster and 2 CR 3 monsters. The CR 18 still has the same problem. The same exists for 1 CR 17 and 3 CR 5s or CR 6s.

What I'm hearing here is: Never use solo monsters and almost never use monsters that are close to the party's level. The first is sad as that is why legendary actions exists. The second should be an option.
Though maybe I'm hearing wrong. Again, I'm not suggesting higher CR as a regular occurrence, but occasionally.

PhoenixPhyre

2019-01-31, 06:05 PM

The problem exists for solo+other creatures as well. Example:
4 level 20 PCs vs 1 CR 18 monster and 2 CR 3 monsters. The CR 18 still has the same problem. The same exists for 1 CR 17 and 3 CR 5s or CR 6s.

What I'm hearing here is: Never use solo monsters and almost never use monsters that are close to the party's level. The first is sad as that is why legendary actions exists. The second should be an option.
Though maybe I'm hearing wrong. Again, I'm not suggesting higher CR as a regular occurrence, but occasionally.

But what I'm denying is that the monsters hitting plate 80% of the time (which they don't, FYI--the average for CR 18 is undefined because there's only one CR 18 which has no attacks...So let's look at CR 17: +13 ATK (65% hit rate vs AC 20). CR 19: +14 (70% hit rate vs AC 20). It's not until you get to CR 22(!) that you start having these problems) is a problem in the first place.

1) At those levels, AC is not nearly as important as it was before, because most monsters can do something to force saving throws.
2) If they didn't hit most of the time, they'd have to do epic spike damage which is much less fun. Miss, Miss, Splat is bad gameplay.
3) The PCs have a host of defensive measures that negate or mitigate this potential, even unoptimized. And the monsters can't optimize to match.

I've run a level 20 game. They murdered a Kraken (beefed up even) without a second thought, barely taking any damage. Oh, and no plate wearers or high-AC people. And that was after a full adventuring day. Then the next day they blew threw 5 deadly encounters with only one short rest...and with a monk and a warlock in the party. Granted, they had good gear. But virtually none of it was +AC stuff.

Edit: you can use solos or boss + minions. They work fine. But only as part of a full adventuring day, never as a single fight in a day. Those will get curb-stomped even if they hit every attack. Adventuring day policing and watching out for action economy (legendary actions help, but not as much as they seem) are the keys to balance in this edition, not number crunching math. And the encounters that will threaten the party most are the hoards of monsters ones. The waves of creatures will test your party to the limit, while solos are eye-candy. Think of them as popcorn fights or catharsis--they're flashy and fun (and stress-relieving) , but they're not the real threat. This is 180 degrees from MMOs (or 3e/PF, from what I've gathered), where the mooks are just chaff that clutters the room.

Kryx

2019-01-31, 06:26 PM

But what I'm denying is that the monsters hitting plate 80% of the time [...] So let's look at CR 17: +13 ATK (65% hit rate vs AC 20). CR 19: +14 (70% hit rate vs AC 20). It's not until you get to CR 22(!) that you start having these problems) is a problem in the first place.
AC 20 is not the norm. Above I outline the options a Fighter has, 1/5 of which is using a Shield. Those options also do not include the longbow, Heavy Crossbow, or Hand Crossbow, all of which are great viable options. So 1/8 Fighter viable fighter options use a shield (though TWF isn't really so viable at high level, so maybe 1/7).

AC 20 is an outlier. AC 18 is the norm across fighter options.

CR 17 has the following Adult Blue Dracolich (+12), Adult Gold Dragon (+17), Adult Red Dragon (+17), Androsphinx (+12, half spellcaster though so that offsets this a bit), Androsphinx (+11, half spellcaster though so that offsets this a bit), Dragon Turtle (+12), Goristro (+13).

Spellcasting offsets the average. With more creatures I expect the average for this CR is somewhere around 13. Far above DMG's Attack bonus of +10 listed for CR 17 so don't give the designers too much credit with monster design.

+13 vs 18 is 80% chance to hit
+17 vs 18 is 95% chance to hit. Though this is an outlier so much less important. Still problematic though

1) At those levels, AC is not nearly as important as it was before
AC shouldn't just stop mattering.

2) If they didn't hit most of the time, they'd have to do epic spike damage which is much less fun. Miss, Miss, Splat is bad gameplay.
Lower levels don't have this rocket tag issue. Why do high level monsters suddenly have to have large bonuses to hit that are far beyond the DMG recommendations for building a monster? As above DMG 274 has a recommended +to to hit at CR 17, not +13 or +17. A few are outliers, but the implementation of monsters is a trend.

3) The PCs have a host of defensive measures that negate or mitigate this potential, even unoptimized. And the monsters can't optimize to match.
PCs have many of the same tools at low levels. Most defensive options are available quite early.

you can use solos or boss + minions. They work fine.
The two scenarios above are not what I'd consider "fine", hence this thread.

EDIT: I have to head to bed. Thanks for the calmer and more fruitful discussion. I'll pick up tomorrow.

PhoenixPhyre

2019-01-31, 06:34 PM

AC 20 is not the norm. Above I outline the options a Fighter has, 1/5 of which is using a Shield. Those options also do not include the longbow, Heavy Crossbow, or Hand Crossbow, all of which are great viable options.

AC 20 is an outlier. AC 18 is the norm.

CR 17 has the following Adult Blue Dracolich (+12), Adult Gold Dragon (+17), Adult Red Dragon (+17), Androsphinx (+12, half spellcaster though so that offsets this a bit), Androsphinx (+11, half spellcaster though so that offsets this a bit), Dragon Turtle (+12), Goristro (+13).

Spellcasting offsets the average. With more creatures I expect the average for this CR is somewhere around 13. Far above DMG's Attack bonus of +10 listed for CR 17 so don't give the designers too much credit with monster design.

+13 vs 18 is 80% chance to hit
+17 vs 18 is 80% chance to hit. Though this is an outlier so much less important. Still problematic though

AC shouldn't just stop mattering.

Lower levels don't have this rocket tag issue. Why do high level monsters suddenly have to have large bonuses to hit that are far beyond the DMG recommendations for building a monster?

PCs have many of the same tools at low levels. Most defensive options are available quite early.

The two scenarios above are not what I'd consider "fine", hence this thread.

You're still only looking at the to hit portion. And that's only a tiny fraction of the whole story. Note that the ones that have the high ATK bonuses are legendary (ie solos). As a hint, dragons are over-CR on purpose. You should always ignore them when calculating things--they're exceptions and exceptions don't make good touchstones. The rest are ATK 13-ish, and put out really crappy physical DPR. Note that the median DPR for CR 17 is only 100, which is significantly below design guidance. You can't look at things in isolation. Calculate the effective DPR, and you'll find that the curve is much more smooth.

And anyone who wants to be a tank (ie whose key selling point is high armor) will have a shield (and thus 20 AC, if not higher). Yes, "DPS-types" will suffer. By design. If you don't want to get hit, take defensive measures. That's why we have parties, after all. Barbarians, for example, are supposed get hit.

Lower level parties do have defensive measures, but they're not nearly reliable enough to count on without a huge health pool to back it up. I'm doing simulated combats right now, and even raging barbarians with good AC (ie scale + shield at level 2) still get knocked down pretty easy. With a much larger health pool, you can soak a few hits just fine, even if you're a wizard.

AC doesn't "stop mattering". But if you only have 18 AC and you're face-tanking heavy solo-types, you're doing it wrong. Unless you're a barbarian, and even then... That's like asking why the rogue who gets surrounded and ganked couldn't survive--he didn't use his class features right.

MaxWilson

2019-01-31, 06:37 PM

What I'm hearing here is: Never use solo monsters and almost never use monsters that are close to the party's level. The first is sad as that is why legendary actions exists. The second should be an option.
Though maybe I'm hearing wrong. Again, I'm not suggesting higher CR as a regular occurrence, but occasionally.

I'm certainly not saying that. Go ahead and use CR 13+ solos if you want--just don't expect them to hit the PCs only as often as CR 3 swarms do.

But the swarms will do more damage in the aggregate anyway. Those ten githyankis will do 10 x 2 x (hit on 17-20 for for 4d6+2) = 78 DPR against AC 21, whereas the Adult Red Dragon will do (hit on 7-20 for 2d10+8+2d6+8+2d8+8+6d8+24) = 74.10 DPR against AC 21, using all legendary actions for Tail Attacks. The Githyankis have a lower accuracy but higher overall DPR, not to mention twice as many aggregate HP (490 HP vs 256 HP on the dragon). The dragon has +2 AC (19 vs. 17) but overall the Githyankis are a greater threat.

However, AC 21 isn't a real tank. If you're something like e.g. an Alert Eldritch Knight using Darkness as a pseudo-Blur, plus Shield, you'll do much better against those Githyankis even if you're not really a melee specialist. E.g. a Sharpshooter EK 13 with Archery style, Dex 20 and Mage Armor has "only" AC 18 when using his longbow, but he can still put up a Darkness spell and use Shield spells (while shooting arrows). The Githyankis need an 18+ at disadvantage to hit him, so he takes only 7.90 DPR from them when Shielding, but the dragon is almost unaffected: it has blindsight and in unaffected by Darkness, and it can hit AC (18 + Shield) = 23 on a 9+, so it still does 63.9 DPR against him*. The githyankis are better against AC 21 but the dragon is better against a real tank or even a semi-real tank like the hypothetical Sharpshooter EK.

Of course in real life the Sharpshooter won't stick around to get hit with all three Legendary tail attacks, so the actual damage will probably be more like 40 DPR.
The Adult Red Dragon is more vulnerable to Wall of Force, but the Githyankis are more vulnerable to Antipathy (Githyankis). That doesn't mean Wall of Force is useless, and it doesn't mean that Antipathy is useless--it means that each of them is sometimes useful. High AC is not useless in high-tier play either--it's exactly the same. Sometimes useful.

The fact that different monsters are strong against different things including AC is a feature, not a bug, even for high-tier play. It's a team game, and if the same teammate always played the same role in every fight against every monster, it would get boring.

JoeJ

2019-01-31, 06:40 PM

AC 20 is not the norm. Above I outline the options a Fighter has, 1/5 of which is using a Shield. Those options also do not include the longbow, Heavy Crossbow, or Hand Crossbow, all of which are great viable options. So 1/8 Fighter viable fighter options use a shield (though TWF isn't really so viable at high level, so maybe 1/7).

How did you get 1/5? There are 3 options for a melee fighter's off-hand (leaving out doing nothing at all with it, which is logically possible but not likely): they can use it to help wield a two-handed weapon, or wield a second weapon, or use a shield. If you're declaring TWF as non-viable, then that leaves only two possibilities.

mephnick

2019-01-31, 07:18 PM

What I'm hearing here is: Never use solo monsters and almost never use monsters that are close to the party's level. The first is sad as that is why legendary actions exists. The second should be an option.

Unfortunately, Lair and Legendary actions basically do nothing to help a solo monster against a group of PCs. It's still one health pool and one movement (sometimes a Legendary action will add movement, but it's not enough). You need to bloat the HP pool to hell and give it a ton of free actions and movement to give it a chance. Or, as is probably intended, have it be the last fight after a grueling adventure against a nearly dead party with no resources. At that point, it becomes more rocket tag than satisfying boss encounter.

The only thing I've found successful is Angry DM's trick of running multiple monsters in one body, weakening it or strengthening it as the HP pools fall away.

MaxWilson

2019-01-31, 07:28 PM

Unfortunately, Lair and Legendary actions basically do nothing to help a solo monster against a group of PCs. It's still one health pool and one movement (sometimes a Legendary action will add movement, but it's not enough). You need to bloat the HP pool to ---- and give it a ton of free actions and movement to give it a chance. Or, as is probably intended, have it be the last fight after a grueling adventure against a nearly dead party with no resources. At that point, it becomes more rocket tag than satisfying boss encounter.

Alternately, you can play the monster with hit-and-run tactics and more cunning. Normally I avoid playing monsters too smart because it's metagamey, but in the case of a high-CR, high-intelligence creature like a dragon it's completely appropriate to e.g. spend one Attack action snatching up a PC in its claws (and biting it a bit too) and then flying off with it to kill the PC in isolation.

That changes it from rocket tag to puzzle game: "how do we get [PC] away from the dragon before it vanishes into the darkness next round?" Maybe the captured PC can Misty Step away; maybe the wizard will cast Wall of Force; maybe the Battlemaster can use a Disarming attack to force the dragon to drop the PC; maybe the Champion can make a mighty leap and climb aboard the dragon so that there's more than one PC it has to deal with.

Then even after you resolve the kidnapping attempt, you still have to finish dealing with the dragon (prevent it from fleeing + resting up + getting some minions + smashing you with everything it's got), so the whole scenario winds up having a lot of dramatic tension despite the dragon's relative fragility in a cage match.

You can only do this a limited number of times before it gets old though.

Rukelnikov

2019-01-31, 07:56 PM

I haven't read the last 2 pages, I'm falling asleep, but thing is, AC is a survivability stat, don't calc how less likely to be hit a single point of AC grants at any given level, what you need to calc is the increase in expected survivability in rounds a point of AC grants.

In a damage calculation every point of AC is applying a percentual reduction to the expected damage of a given enemy, compare the AHP/DPR ratio at different levels, I haven't done the numbers, but only then can we draw a conclusion of exactly how useful or useless does AC become at higher levels.

Arzanyos

2019-01-31, 08:03 PM

Another thing to note. XGTE has encounter guidelines where they matchup how many pcs of a given level is worth how many monsters of a given cr in an encounter. According to them all monsters with CR11 or higher are worth at least 2 pcs. They're all elites, and as such, hitting things more often than standard monsters is their shtick.

PhoenixPhyre

2019-01-31, 08:26 PM

Theodoxus

2019-01-31, 08:35 PM

Didn't read the thread... but to address your idea of adding proficiency bonus to AC? I just did that - I've been working on a massive mod taking 4th ed concepts and melding them with 5th ed BA. It's worked really well. None of my players have high across the board defenses, so I can threaten them all with a mix of monsters. If the high AC tank manages to tick off the beastie targeting AC that had been trying to eat the sorcerer, all the better for the party. But it'll probably bring the ire of the monster that's tossing out attacks that target reflex.

All in all, it's taken the game up a notch. I can throw higher CR critters at the party, getting some fun attacks and unique abilities that a straight 5E game wouldn't be able to handle.

PhoenixPhyre

2019-01-31, 08:50 PM

Another piece of evidence that the DMG's guidance isn't absolute is looking at real monsters. For all CRs other than 0, 8, and 12, the median attack bonus is higher than the guidance by at least half a point. On the other hand, the DPR (the other half of offensive CR) is generally lower than or on guidance, and the attack bonuses that are on-guidance are also on DPR. For all CRs:

Below guidance: 15 (including all CRs 16+).
On guidance: 7 (CRs 1-3, 6-8, and 12).
Above guidance: 7 (0-0.5, 13-15).

But the variance is huge in DPR. For example, CR 13 (median 94 DPR vs guidance of 87-92) has a standard deviation of 61.9, while CR 14 (median DPR 128, higher than CR 15 even, with guidance of 93-98) has a standard deviation of 68.8. Partly that's because in those CR ranges there are only 9 (CR 13) and 6 (CR 14) creatures. And that's where the adult dragons mostly fall (for the weaker colors), and dragons punch notoriously above their weight in CR.

Also note that CR 13 and 14 legendary creatures are the natural "end bosses" for level 10/11 campaigns, where lots of campaigns end.

Also note that every CR is below guidance for raw HP. Some by a little, most by a lot.

MaxWilson

2019-01-31, 10:11 PM

Another piece of evidence that the DMG's guidance isn't absolute is looking at real monsters.

*Snip*

Also note that every CR is below guidance for raw HP. Some by a little, most by a lot.

Not to mention the fact that the DMG guidance itself outright tells you that it doesn't matter what the to-hit bonus is of a given monster of CR XYZ, as long as the damage or HP or AC is increased to compensate.

The MM numbers you quote here bear that out, but the DMG explains that the deviations are deliberate. No individual statistic matters, only the total stat block.

Skylivedk

2019-02-02, 05:26 AM

AC shouldn't just stop mattering.

Lower levels don't have this rocket tag issue. Why do high level monsters suddenly have to have large bonuses to hit that are far beyond the DMG recommendations for building a monster? As above DMG 274 has a recommended +to to hit at CR 17, not +13 or +17. A few are outliers, but the implementation of monsters is a trend.

PCs have many of the same tools at low levels. Most defensive options are available quite early.

First of: thank you for everything, you have contributed to the community. I, and my players, have tremendously enjoyed the changes (splitting feats, changing TWF and GWM and many more) to our D&D which are a direct result of a lot of your good work. I still want to try a game with your houserules fully implemented (or 5,5e as my brother and I jokingly call it).

Let me address a few points here:
1) AC doesn't stop mattering; it requires a much higher investment or team effort (buffs, debuffs) to matter the way it did earlier. I could understand that from a design perspective: as the game progress you want the challenge to increase. Since HP generally scales faster than damage, AC cannot do it and still allow for a rewarding feeling as a tank.

2) what do you mean they don't have the rocket tag issue? I'd almost say the opposite: due to the lower player hp, the monsters have a much more frequent one round KOs. Statistical variance is more dangerous. Naturally depends on which level we're talking about exactly(and I have a feeling we both feel the game is atrociously designed for level 1 and doesn't really get into its groove before 3-5).

3) yes, the players have a lot of the defensive options, but using them is relatively more costly. My Wizard had both Shield and Mage Armour at level 1-2, but he'd rather take hits than not have the AoE Win button called sleep. Later on, you're happy to burn low level slots for survival, since the slots are suddenly only a fraction of your daily power expenditure.

How did you get 1/5? There are 3 options for a melee fighter's off-hand (leaving out doing nothing at all with it, which is logically possible but not likely): they can use it to help wield a two-handed weapon, or wield a second weapon, or use a shield. If you're declaring TWF as non-viable, then that leaves only two possibilities.

He's not declaring it. The math is. I think for this discussion to work and be fruitful, everybody has to have the willingness to surrender to the evidence of the math. That's the focus.

Didn't read the thread... but to address your idea of adding proficiency bonus to AC? I just did that - I've been working on a massive mod taking 4th ed concepts and melding them with 5th ed BA. It's worked really well. None of my players have high across the board defenses, so I can threaten them all with a mix of monsters. If the high AC tank manages to tick off the beastie targeting AC that had been trying to eat the sorcerer, all the better for the party. But it'll probably bring the ire of the monster that's tossing out attacks that target reflex.

All in all, it's taken the game up a notch. I can throw higher CR critters at the party, getting some fun attacks and unique abilities that a straight 5E game wouldn't be able to handle.

Can you link that? For the last couple of months, I've been looking for something that makes combat in my roleplaying games more interesting. I might end up building my own system inspired by the Witcher's stamina game, 13th age and one of the wound systems. Essentially, I'd like a range of mechanically supported more or less equally viable combat options for all classes throughout the game.

MaxWilson

2019-02-02, 11:16 AM

He's not declaring it. The math is. I think for this discussion to work and be fruitful, everybody has to have the willingness to surrender to the evidence of the math. That's the focus.

Math is a product of assumptions. The assumptions have to be either justified or mutually-assumed. JoeJ is asking where Kryx's assumption comes from. It's relevant to the math.

PhoenixPhyre

2019-02-02, 11:25 AM

Math is a product of assumptions. The assumptions have to be either justified or mutually-assumed. JoeJ is asking where Kryx's assumption comes from. It's relevant to the math.

And making assumptions that differ from the assumptions that define the underlying system is a good way to have things feel off or not fit right. Not because the system is at fault, but because the assumptions are (often unintentionally) incongruous.

The math can still be useful as a test of the robustness of the system to alternate assumptions, but it doesn't serve as an effective critique of the system's core. The assumptions made dictate the outcome, often very strongly.

MaxWilson

2019-02-02, 01:10 PM

And making assumptions that differ from the assumptions that define the underlying system is a good way to have things feel off or not fit right. Not because the system is at fault, but because the assumptions are (often unintentionally) incongruous.

Yes. Another way to look at it is: if you make a bunch of assumptions, and then find that your math predicts something other than what you actually observe, your assumptions are probably wrong. Kryx's OP basically asks, "Why doesn't high-level 5E behave the way I expect it to?" and the answer is "your assumptions about encounter composition are incorrect, or at least unusual."

JoeJ

2019-02-02, 01:38 PM

He's not declaring it. The math is. I think for this discussion to work and be fruitful, everybody has to have the willingness to surrender to the evidence of the math. That's the focus.

Where did you see the math? I just saw a number, with nothing to indicate how that number was derived.

Skylivedk

2019-02-02, 02:49 PM

Where did you see the math? I just saw a number, with nothing to indicate how that number was derived.

I don't want this to devolve into a discussion about TWF. According to the calculations shown to me repeatedly, amongst others by Kryx', TWF is worse on fighters than dueling later in the game.

Arzanyos

2019-02-02, 02:54 PM

I don't want this to devolve into a discussion about TWF. According to the calculations shown to me repeatedly, amongst others by Kryx', TWF is worse on fighters than dueling later in the game.

I think I see the issue here. The assumption JoeJ was asking about wasn't if TWF was viable, it was about only 1/5 of fighters using shields.

JoeJ

2019-02-02, 02:54 PM

I don't want this to devolve into a discussion about TWF. According to the calculations shown to me repeatedly, amongst others by Kryx', TWF is worse on fighters than dueling later in the game.

I'm not disputing that. I'm asking where the claim that only 1/5 of melee fighters use shields came from. Logically, eliminating TWF from consideration should make shield use more common, not less.

edit: ninja'd. Yep, Arzanyos is correct.

LudicSavant

2019-02-02, 03:19 PM

I'm not disputing that. I'm asking where the claim that only 1/5 of melee fighters use shields came from.

It comes from Kryx claiming that the "norm for the Fighter" consists of a list quoted below, and that only one of the 5 options on his list uses a shield.

Notably, this list does things like list two-handed polearms twice, and two-handed weapons overall three times (counting various feat combinations as separate options), but does not list polearm-and-shield (quarterstaff/shield or spear/shield), or count rapier/shield as a separate style from S&B, or count various feat combinations as separate options for S&B.

The second post does something similar for archery, listing heavy crossbow, longbow, and hand crossbow as all being separate options, while S&B is still crowded under a single label that does not represent the diversity of viable shield builds.

This isn't about AC-focused builds. I'm seeking to understand the norm. The norm for a Fighter is one of several options:

TWF (very bad by RAW at higher levels)
GWM
Polearm
Polearm+GWM
S&B

S&B is a thing, but it is likely around 1/4 of fighters or less. Defensive duelist would work for the TWF and S&B, but not GWM and Polearm and Polearm+GWM. A Fighter shouldn't be expected to go full defensive to keep up with the math of the early game.

Above I outline the options a Fighter has, 1/5 of which is using a Shield. Those options also do not include the longbow, Heavy Crossbow, or Hand Crossbow, all of which are great viable options. So 1/8 Fighter viable fighter options use a shield

JoeJ

2019-02-02, 04:31 PM

It comes from Kryx claiming that the "norm for the Fighter" consists of a list quoted below, and that only one of the 5 options on his list uses a shield.

Notably, this list does things like list two-handed polearms twice, and two-handed weapons overall three times (counting various feat combinations as separate options), but does not list polearm-and-shield (quarterstaff/shield or spear/shield), or count rapier/shield as a separate style from S&B, or count various feat combinations as separate options for S&B.

The second post does something similar for archery, listing heavy crossbow, longbow, and hand crossbow as all being separate options, while S&B is still crowded under a single label that does not represent the diversity of viable shield builds.

I see. Yeah, that list is just a confused mishmash of weapon choice, fighting style choice, and feat choice. Those are not independent variables, so the list is not valid. (He's also assuming that each choice on the list is equally likely, which would be problematic even if the list were valid.)

Kadesh

2019-02-02, 06:03 PM

I disagree that including a list of things a Fighter can do with their Bonus Action is confusing, just because some things have a varying amount of investment in the class.

MaxWilson

2019-02-02, 06:17 PM

I see. Yeah, that list is just a confused mishmash of weapon choice, fighting style choice, and feat choice. Those are not independent variables, so the list is not valid. (He's also assuming that each choice on the list is equally likely, which would be problematic even if the list were valid.)

By "valid" do you mean "exhaustive"?

Cybren

2019-02-02, 06:46 PM

I think joe meant “comprised of analogous elements”

JoeJ

2019-02-02, 07:24 PM

By "valid" do you mean "exhaustive"?

No, I mean composed of analogous elements.

I disagree that including a list of things a Fighter can do with their Bonus Action is confusing, just because some things have a varying amount of investment in the class.

Since we're talking about shield use and AC, it's not a list of things a fighter can do with a bonus action, but a list of things a fighter can do with their off hand that's relevant. What action they use, if any, doesn't matter.

LudicSavant

2019-02-02, 07:47 PM

Kadesh

2019-02-02, 08:01 PM

Since we're talking about shield use and AC, it's not a list of things a fighter can do with a bonus action, but a list of things a fighter can do with their off hand that's relevant. What action they use, if any, doesn't matter.

Eh. Whatever. I dropped in, but it changes my argument nothing. It gives a list of what a fighter can potentially do with their off hand.

Eric Diaz

2019-02-02, 08:04 PM

The answer is relatively simple IMO:

- Magic items that actually increase your AC*.
- Even if you get hit more often, you have more ways (spells, abilities, etc.) to avoid damage (and other effects of being hit) or make it irrelevant.

* the magic itens thing is also simple, +AC magic items are more common than +to-hit ones, and easier to stack. You only attack with one weapon at a time, but you can use armor, shield, ring, cloak, etc., all at the same time.

But what if we are not using magic items at all?

I am convinced that the game ASSUMES you'll get some magical items regardless of what the designers say, and I don't think it is possible to say a 15th level paladin with a +2 shield and a holy avenger is as powerful as one with no magical weapons. There is just no comparison.

PhoenixPhyre

2019-02-02, 08:17 PM

But what if we are not using magic items at all?

I am convinced that the game ASSUMES you'll get some magical items regardless of what the designers say, and I don't think it is possible to say a 15th level paladin with a +2 shield and a holy avenger is as powerful as one with no magical weapons. There is just no comparison.

The system makes no such assumption. Sure, a blinged-out character is more powerful (and thus can face harder foes). But the math is calibrated (and the guidance is written assuming) no accuracy or AC increasing items. And it works, assuming you actually follow those guidelines.

Magic items are merely gravy, effectively letting a level X character face threats more suited for a magic-item-less character of level X+Y. They're not expected or required in any way by the base system math. This is entirely unlike 4e or 3e (or even 2e), where a character without proper +X items would be unable to hit "level appropriate" enemies except in a 20.

JNAProductions

2019-02-02, 08:22 PM

Eh... The game generally assumes you'll have a weapon that counts as magical around late tier II, early tier III.

It doesn't have to have any properties other than "magical", but there's that.

Unoriginal

2019-02-02, 09:05 PM

Eh... The game generally assumes you'll have a weapon that counts as magical around late tier II, early tier III.

It doesn't have to have any properties other than "magical", but there's that.

The whole PC-monster combat math was calculate without any magic items on the PC side.

That you can get a magic weapon if the magic items'/loot's suggested guidelines isn't relevant to the balancing/scaling of the math.

JNAProductions

2019-02-02, 09:07 PM

The whole PC-monster combat math was calculate without any magic items on the PC side.

That you can get a magic weapon if the magic items'/loot's suggested guidelines isn't relevant to the balancing/scaling of the math.

Check how many higher-tier monsters have resistance to non-magical damage.

I do understand that a mob of mooks is still a very viable threat, but at the same time, there's even a lot of MOOKS that have that. (Notably outsiders.)

If I were to run a game with no magic items, I'd definitely manually add in silver or adamantine or cold iron vulnerability, or whatever material was appropriate. Otherwise, weapon-users would have their output gimped pretty damn heavily.

Edit: Again, this is not a +X weapon. This is literally a sword that counts as magic. No special bonuses, just magic.

Maybe it doesn't rust or something.

Unoriginal

2019-02-02, 09:35 PM

Check how many higher-tier monsters have resistance to non-magical damage.

I do understand that a mob of mooks is still a very viable threat, but at the same time, there's even a lot of MOOKS that have that. (Notably outsiders.)

Otherwise, weapon-users would have their output gimped pretty damn heavily.

Indeed. Yet, as the DMG indicates, monsters with that kind of resistance are considered as having twice as many HPs in CR calculations.

So what is the most likely? That they decided to make monsters that count as having twice as much HPs unless if a certain condition is met, but expected said condition to be met all the time or near, making it just a "gotcha!" in case low level PCs went playing with the big boys too early? Or that they actually put this condition because they expected people to not met as part of their combat calculations?

Because they could just have given the monster twice the HPs and be done with it. But the fact is, no matter what caster-supremacy-defenders try to pretend, 5e casters (as they currently are) *couldn't* compete in term of damage output with martials if those "resistances to damage except magic" didn't exist and the designers had not went the "if you have resistance your HPs count double for CR calculation" way. What would a caster do against an Ancient Dragon with 500 HPs? When the most powerful straight-damage-spell, that a top-level caster can cast once a day, deals on average 120 damages if it hits? Martials would have troubles, but they'd eventually be able to grind the monster down, but since they're limited in number of spells most of the casters would run out and have to rely on cantrips, dealing only a fraction of what the martials deal with their regular attacks, which would be deeply unsatisfying.

The Resistance-to-nonmagic system let the game keep a reasonable combat length, and let both casters and martials contribute meaningfully. Magic weapons are the wonderous weapons of greatness they're supposed to be because they give that much of a boost.

If I were to run a game with no magic items, I'd definitely manually add in silver or adamantine or cold iron vulnerability, or whatever material was appropriate.

That's how you'd run, but that's not how the game designers calculated the game.

Doesn't make their way or yours wrong, but magic items are *supposed* to be extraordinary game changers in this edition. Not cheap character upgrades disguised as equipment of wonder. And as such they were added after the designers finished the calculations for encounter balance.

Edit: Again, this is not a +X weapon. This is literally a sword that counts as magic. No special bonuses, just magic.

Maybe it doesn't rust or something.

I know. The Moon-Touched Sword is pretty great for that.

JNAProductions

2019-02-02, 09:38 PM

It's double for lower tiers.

Higher tiers decrease the multiplier.

Unoriginal

2019-02-02, 09:44 PM

It's double for lower tiers.

Higher tiers decrease the multiplier.

Fair point, but they don't remove it, yes?

Given 5e's typical lack of desire to write more rules than what they can get away with, I think it's logical they did things that way for more than waste time and paper.

I honestly don't see an explanation that's more coherent than the one I wrote, but if you have one I'd gladly read it.

MeeposFire

2019-02-03, 12:42 AM

It makes sense that the mod decreases at higher CRs because as the levels increase (and you are more likely to face such a threat) there are more ways for a group of PCs to overcome the challenge other than straight weapon damage (also while it does not mandate it they probably figure the higher level you are the more likely you will have access to one or more magical weapons or ways to make weapons magical decreasing the value of the resistance as well).

PhoenixPhyre

2019-02-03, 08:36 AM

Fair point, but they don't remove it, yes?

Given 5e's typical lack of desire to write more rules than what they can get away with, I think it's logical they did things that way for more than waste time and paper.

I honestly don't see an explanation that's more coherent than the one I wrote, but if you have one I'd gladly read it.

MeeposFire is correct. The only classes that do not get something that will pierce resistance naturally are:

Barbarians
Fighters (except EK)
Rogues (except AT)

It makes sense that the mod decreases at higher CRs because as the levels increase (and you are more likely to face such a threat) there are more ways for a group of PCs to overcome the challenge other than straight weapon damage (also while it does not mandate it they probably figure the higher level you are the more likely you will have access to one or more magical weapons or ways to make weapons magical decreasing the value of the resistance as well).

As a note: the modifier for resistances (and only for having multiple resistances or the common ones--a solitary resistance isn't worth anything) are (based on the tier of expected opponents, not CR, really):

Tier | eHP Multiplier
1 | 2
2 | 1.5
3 | 1.25
4 | 1

So they do go away by T4, and are minimal by T3. Immunities, on the other hand, scale like

Tier | eHP Multiplier
1 | 2
2 | 2
3 | 1.5
4 | 1.25

The system is set so that at low levels, resistances are important for balance (consider the Imp. Super squishy, except resistance to lots of things). At high level, they're situationally important. Note also that in a no-magic-item game, DMs should carefully consider what they're throwing at the party. There are lots of classic monsters that don't have such resistances (or have ones that are pierced by silver) even up to high levels. For example, use devils (silver works) not demons. Don't use golems unless you have a supply of adamantine weapons. Dragons are fine, though.

The "everything's immune/resistant" thing is overblown IMO. The numbers are skewed by the masses of demons in the books more than anything.

Unoriginal

2019-02-03, 08:45 AM

MeeposFire is correct. The only classes that do not get something that will pierce resistance naturally are:

Barbarians
Fighters (except EK)
Rogues (except AT)

As a note: the modifier for resistances (and only for having multiple resistances or the common ones--a solitary resistance isn't worth anything) are (based on the tier of expected opponents, not CR, really):

Tier | eHP Multiplier
1 | 2
2 | 1.5
3 | 1.25
4 | 1

So they do go away by T4, and are minimal by T3. Immunities, on the other hand, scale like

Tier | eHP Multiplier
1 | 2
2 | 2
3 | 1.5
4 | 1.25

The system is set so that at low levels, resistances are important for balance (consider the Imp. Super squishy, except resistance to lots of things). At high level, they're situationally important.

Fair point.

Note also that in a no-magic-item game, DMs should carefully consider what they're throwing at the party. There are lots of classic monsters that don't have such resistances (or have ones that are pierced by silver) even up to high levels. For example, use devils (silver works) not demons. Don't use golems unless you have a supply of adamantine weapons. Dragons are fine, though.

Eh, I'm against the idea that a DM shouldn't use a particular monster because the PCs don't have the tool to make it easier. General Zod wouldn't be an impressive villain if everyone had kryptonite.

Sometime the PCs will lack the tool, and it'll be hard, but it's not instant death either.

The "everything's immune/resistant" thing is overblown IMO.

True. It's more "enough things are resistant", IMO.

The numbers are skewed by the masses of demons in the books more than anything.

Damn demons, they keep spreading chaos and making numbers less reliable.

PhoenixPhyre

2019-02-03, 09:01 AM

Eh, I'm against the idea that a DM shouldn't use a particular monster because the PCs don't have the tool to make it easier. General Zod wouldn't be an impressive villain if everyone had kryptonite.

Sometime the PCs will lack the tool, and it'll be hard, but it's not instant death either.

It's less should not use but should use with caution. It goes for Rakshasa vs spell-caster-heavy parties without significant numbers of 6th level offensive spells. It goes for Mind Flayers or Intellect Devourers vs a party of knuckleheads. Foreshadowing should be involved (just like using werewolves vs a low-level party without warning), and a chance to find alternate means/tactics/etc. should also be available. That, and the ability to flee should also be available. And the effect (doubled effective health vs martials, etc) must be taken into consideration when planning encounter balance.

Not being able to make a meaningful contribution is frustrating. Being TPK'd by a monster you had no chance of preparing for is campaign ending. "Failing" against the boss because he just no-sells you unless you find his one weakness (exaggerating for effect here) is a form of railroading.

Resistance isn't as bad, because half is not zero. But it does alter the calculus heavily.

MaxWilson

2019-02-03, 12:33 PM

Barbarians
Fighters (except EK)
Rogues (except AT)

Zealot barbarians inflict radiant damage too. So every single class has options for damaging weapon-immune creatures.

PhoenixPhyre

2019-02-03, 12:35 PM

Zealot barbarians inflict radiant damage too. So every single class has options for damaging weapon-immune creatures.

So in the end it comes down to "let people know ahead of time that they'll need other strategies because magic weapons will be scarce or non-existent". Which I'm fine with. I'm all about having people communicate the important stuff ahead of time instead of trying to pull a plot twist after characters are made.

MaxWilson

2019-02-03, 03:19 PM

So in the end it comes down to "let people know ahead of time that they'll need other strategies because magic weapons will be scarce or non-existent". Which I'm fine with. I'm all about having people communicate the important stuff ahead of time instead of trying to pull a plot twist after characters are made.

Also in practice, anyone who plays repeated games with the same DM will figure out pretty quickly what their style is w/rt magic items and weapon-resistant monsters. It's not really any different than figuring out if the DM likes hordes vs. solos, or spellcasting humanoids vs. monstrous brutes.

Of course there nothing stopping you from asking up front either.

PhoenixPhyre

2019-02-03, 03:21 PM

Also in practice, anyone who plays repeated games with the same DM will figure out pretty quickly what their style is w/rt magic items and weapon-resistant monsters. It's not really any different than figuring out if the DM likes hordes vs. solos, or spellcasting humanoids vs. monstrous brutes.

True enough. But I'd rather be as upfront about it as possible--I'm not find of things that even smell lightly of Gotcha DM'ing. And if someone has a big problem with how I run things, I'll consider adjusting my style.

MaxWilson

2019-02-03, 05:05 PM

True enough. But I'd rather be as upfront about it as possible--I'm not find of things that even smell lightly of Gotcha DM'ing. And if someone has a big problem with how I run things, I'll consider adjusting my style.

Fair enough.

Lonely Tylenol

2019-02-03, 07:06 PM

You're right, I forgot about defensive fighting style. However we shouldn't assume most fighters take Shield & Defense. I'd wager that's ~20 of fighters.

I’m sorry, I see you being challenged on a lot of the stranger assumptions you’re making in this thread, but not this one, and this one needs to be challenged.

You say that an Ancient Red Dragon routinely hitting the plate-wearing Fighter undermines the Fighter’s identity. But then, when someone presents you an actual defense-oriented Fighter, you respond that that isn’t an assumption you can make of Fighters.

The problem with that is, wearing plate isn’t an identity. Having 18 AC isn’t a defensive identity. It’s a standard expectation nearly every PC who wades into melee should make, and the absolute bare minimum one should make to be considered an effective melee combatant. In fact, it’s one every single class of character that might get within striking distance of an enemy has ways to reach an 18 AC (the equivalent of plate) early:

- All Barbarians have medium armor and shield proficiency, and can reach 18 AC at level 1 with 14 Dexterity (which Barbarians are expected to have). A Barbarian without breastplate reaches 18 AC at 18 Con, or 20 Con + 16 Dex (which some Barbarians will get from racials). All of these things are achievable by level 8, with some sooner. They also have an in-build damage resistance mechanic.
- Valor Bards have medium armor and shield proficiency, and can reach 18 AC at level 3 with 14 Dexterity. They’re also proficient out of the gate with most finesse weapons, so Dexterity-based melee Bard is a pretty standard expectation.
- All Clerics have medium armor and shield proficiency, plus heavy armor proficiency if you are roughly half of all Cleric archetypes, and can reach 18 AC at level 1 with 14 Dexterity or 13 Strength and a shield. 20 AC is reachable by the heavy armor archetypes, with no special reason not to use a shield. This is not considering the extra AC afforded by class features (Forge Cleric’s always-on +1 to AC) or spells (Shield of Faith et al).
- All Fighters have heavy armor and shield proficiency, so 18 AC is attainable at level 1 with 14 Dexterity or 13 Strength and a shield. 20 AC is reachable with plate and a shield. This is not considering the extra AC afforded by the Defense fighting style, subclass features (such as Parry maneuver use), or spells.
- All Monks can reach 18 AC by level 8 with the Unarmored Defense feature simply by maxing their main stat. 20 AC is reachable by level 16 (Ancient Red Dragon territory) simply by maxing both main attributes, which is a standard expectation of the game. They also have an in-build disadvantage mechanic (bonus action Dodge). This is not considering the Kensai’s in-built AC mechanic, which makes 18 AC reachable by level 3.
- All Paladins have heavy armor and shield proficiency, so 18 AC is attainable at level 1 with 14 Dexterity or 13 Strength and a shield. 20 AC is reachable with plate and a shield. This is not considering the extra AC afforded by the Defense fighting style, subclass features, or spells (Shield of Faith et al).
- All Rangers have medium armor and shield proficiency, and can reach 18 AC at level 1 with 14 Dexterity and a shield (and at least that much Dexterity is the standard expectation of Rangers). Without a shield, a Ranger can still reach 18 AC by level 8 with the Defense fighting style (which, if they’re a melee Ranger, is encouraged, as Two-Weapon-Fighting kind of sucks).
- Rogue is perhaps the worst for this; they can reach 17 AC with studded leather armor, and 18 with the Dual Wielder feat by level 8 if Human, or 12 otherwise, by also maxing their main stat. But Rogues also have an in-built resistance mechanic and Evasion by that point, so they have defensive tools not reliant on their AC. This is not considering spells from Arcane Trickster.
- Hexblade Warlocks have medium armor and shield proficiency, and can reach 18 AC with 14 Dexterity and a shield, and also have an in-built miss mechanic (Armor of Hexes). This is not considering the AC and resistances afforded by spells (Shield et al).

All of this is single-classed; it doesn’t consider multiclasses or complex builds.

Feats decrease the barrier to entry for AC even more. Medium Armor Master gives shield-free 18 AC to all medium armor proficient characters (and if you’re melee, it’s heavily encouraged), and 20 AC with shield for truly defensive builds. Defensive Duelist adds your proficiency to AC against an attack as a reaction, making 18 trivial, but also 18 + proficiency attainable. Dual Wielder or Magic Initiate (Mage Armor) gets the Rogue to 18.

When literally half of all classes can reach 18 AC at around the same speed as the Fighter, and half the remaining classes can do so with their respective melee-oriented archetypes (or just a little effort), 18 AC isn’t special, much less an identity. It’s something you’re just expected to do, in order to achieve baseline competency.

So, please, don’t consider “wears armor” an identity that’s under attack here. It’s nothing worth mentioning. If you want to talk about an actual defensive identity at stake, talk about actual defensive Fighters—sword and board, maybe with Defense fighting style (trivialized by Champion, who gets two styles by mid-tier anyway, with [Offensive Style] + Defense being standard, but available to all Fighters), maybe with defensive maneuvers (Battlemaster Parry), maybe with defensive spells (Eldritch Knight is literally limited to abjuration as one of its two schools for spell selection), maybe with feats (Defensive Duelist). That is a defensive identity. “Wears armor” is something a majority of classes can reach; it certainly isn’t a Fighter identity.

And then, with 20 or 21 unbuffed AC plus elective features as the new baseline for “actually defensive identity”, it’s important to consider, for these CR 13+ creatures, that not only are they expected to be elite (they are just inherently stronger than any player, and thus should always be considered “more than one player” in terms of encounter balance) and thus probably better than average at hitting to account for turn disparity involved with including them, but by the level of play where it’s considered “acceptable”, you can make much more lenient assumptions about what buffs and spells are available to a group of players which adjusts the math: at 5th level, spells like Shield of Faith, Mage Armor, Shield, Warding Bond, Warding Wind, Mirror Image, Blur, Haste, and the like are your highest-level spells, or a significant expenditure of your limited daily resources. By 15th level, however, these spells are trivial enough that they (and others) can be cast every encounter worth noting, and these sorts of things do change how the math plays out. And this isn’t including the absolute game-changers you get later on.

If you are expecting unbuffed, unspecialized AC to provide meaningful defensive identity against the mathematically toughest creatures in the multiverse, you’re going to have a very, very bad time.