Does rolling AC benefit the underdog?

[Talakeal]

So there is an optional rule for the defender to add a d20 roll to AC instead of their base 10.

I would think this would make combat more random, thus favoring the underdog, but looking at the probability tables online doesn't quite seem to bare this out.

Anyone have any experience with using this rule? Any probability whizz able to break down how it works out?

Thanks!

[NomGarret]

Technically it's a .5 bonus to the defender, which is minor enough that it will affect about a roll an encounter, but not til higher levels. Realistically, though it will help a significant underdog, the additional swingy nature will wash out more than it may seem. You'll just notice when the underdog hitter rolls an 18 and the defender rolls a 2 more than the other way round.

[Jormengand]

Essentially, it favours the defender slightly as above, but it does also favour the underdog. For example, the probability of someone attacking someone with an AC bonus (their AC without the 10 or 1d20) 5 higher than the attacker's attack bonus hitting under the normal rule (http://anydice.com/program/3c49) is 30%, and under the variant rule (http://anydice.com/program/183) it's also 30% - the underdog benefit and the defender benefit cancel. If however we're talking someone with an AC bonus 10 higher, then the normal rule gives you a 5% chance to hit, and the variant gives you a 13.75% chance.

If their AC bonus is 10 lower than your attack bonus, however, you normally can only miss because 1s always miss, giving you a 95% hit rate, but under the variant you only have an 86.25% chance.

Essentially: it benefits the defender slightly and the underdog potentially a lot.

[Sneak Dog]

So there is an optional rule for the defender to add a d20 roll to AC instead of their base 10.

I would think this would make combat more random, thus favoring the underdog, but looking at the probability tables online doesn't quite seem to bare this out.

Anyone have any experience with using this rule? Any probability whizz able to break down how it works out?

Thanks!

First gut response: the average when rolling 1d20 is 10.5. This is higher than 10.

Now, math. For reference, odds calculated with Anydice. If their calculations are messed up, so is this post.
Rolling 1d20 with equals being success. For readability, the left column is 10+X and the chances are for the attacker to hit you.


10+X
Chance for 1d20+X
Chance for 10+X


16
26.25%
25%


15
30%
30%


14
34%
35%


13
38.25%
40%


12
42.75%
45%


11
47.50%
50%


10
52.50%
55%


9
57.25%
60%


8
61.75%
65%


7
66%
70%


6
70%
75%


5
73.75%
80%



In conclusion, the easier you are to hit, the greater the benefit of rolling your AC. The break-even point is at 15. So nearly always is rolling your AC better.

[animewatcha]

Due to the defense being a d20 now instead for base 10, would it not be affected by whatever affects d20 rolls in general? Like Alter Fortune, or destiny or whatever it is that is immedaite action that costs 200 xp. Would the d20 turn into a dex check now. Allowing for things like marshal Dex aura being added to it.

[Ualaa]

Average roll is (Minimum + Maximum) divided by 2.

So +10 is fixed.
But long term, a d20 is going to average 10.5.

If you're rolling the to-hit, that's a slight advantage over always having a 10.
If you're rolling to avoid being hit (rolling your AC against their fixed attack), that's again a slight advantage.

I run my combat on a laptop (using Hero Lab).
I'm happy to accept randomly rolled initiative for six monsters on the board.
I'll generally group them somewhat, just to keep the round moving quickly.
If I have initiatives of: 20, 18, 15, 15, 08 and 03; that means 3x @18 and 3x @08.

Players, in my experience, always want to roll.
Although the 'Tactical Console' rolls for them, when combat begins, even the guy whose 'dice hate him' wants to roll their own initiative.



So you can give all of the rolls to the player if you like.

It isn't any faster, if you roll to hit the player.
Than if they player rolls to miss you hit.

Sure, it might be faster/slower initially, to get used to a new system.

The roll is essentially a skill check, against a DC.
The roll vs DC is either d20 plus bonuses or 10 plus bonuses.

If your mob normally has d20 +19 to hit.
You can call their hit a DC of 29 (10 on the d20).
And then give the player an avoidance roll, which will be d20 +(their AC minus 10).
They lose the -10 on any fixed value that wouldn't have been rolled but is instead.
Effectively, you can either take 10 and not roll, or roll but not take 10.

Likely the least interesting would be everyone takes 10.
So just compare AC (which already includes a take 10) vs attack roll (at +10 higher, but without the d20 roll).
That would be massively faster, but also less fun.

You wouldn't have to reveal attack bonuses.
Just have the player roll to avoid two claw and one bite attacks.
Your mob has +19 to hit, so the player needs to roll against DC 29.
They can call out 26, 34 and 25, without knowing that 29 is the DC.
They'll figure it out fairly quickly; my players catch that an AC has changed and know that the mob is now using/not using Combat Expertise.
The same would be true with their avoidance, and the mob sometimes using Power Attack.



It wouldn't be overly hard to have the players make all/most of the rolls.
Based on their character sheets mostly.
Or at least give them the d20 rolls.

If the mob has to make a saving throw...
If it has a +9 to the relevant save.
Rather than rolling d20 +9.
It can have a DC of 19 (this is what fourth edition did).
And then the Wizard (or whoever triggered the save) gets to roll to hit that target.

If the normal course of events is...
Wizard casts a sleep spell.
The DC is 10, + Spell Level, + INT modifier.
So say 10+1+4 = 15.
The wizard instead rolls d20 +5 (trading away the 'take 10' for the d20 roll).
The mob would have had to roll a Will save, say d20+4.
Now the wizard is rolling d20 +5 against a DC of 14.
The mob has lost it's d20 roll, but gets to take 10 on the target DC.



If you were to use this system for 2-3 sessions, it would be very second nature very quickly.

Players do like to roll the dice.

[exelsisxax]

Plenty of demonstrations about mathmatics already.

What hasn't been said: this will not benefit your players. When monsters eat x4 crit damage hits, the game doesn't break. They're supposed to die. If a PC gets hit with a x4 crit confirmation from any meaningful enemy, it absolutely sucks for the player. There's a reason that natural weapons have minimum crit ranges and most civilized enemies aren't print as wielding keen scythes or heavy picks and improved critical.

Similarly, rolling for AC significantly increases the odds for a lucky 2-3 hit string that kills a PC. An attack can roll high, OR the PC can roll low. AC is worthwhile in its reliability, if you randomize a significant amount of it the volatility will seriously increase the odds of your players getting killed without reason or significance. Unless the campaign is pretty easy, the odds of character death are higher. Players are only better off it they're going to be dying all the time anyway from mobs with way out of scale attack bonuses when 10 = always hit but a D20 might evade something.

[Talakeal]

Plenty of demonstrations about mathmatics already.

What hasn't been said: this will not benefit your players. When monsters eat x4 crit damage hits, the game doesn't break. They're supposed to die. If a PC gets hit with a x4 crit confirmation from any meaningful enemy, it absolutely sucks for the player. There's a reason that natural weapons have minimum crit ranges and most civilized enemies aren't print as wielding keen scythes or heavy picks and improved critical.

Similarly, rolling for AC significantly increases the odds for a lucky 2-3 hit string that kills a PC. An attack can roll high, OR the PC can roll low. AC is worthwhile in its reliability, if you randomize a significant amount of it the volatility will seriously increase the odds of your players getting killed without reason or significance. Unless the campaign is pretty easy, the odds of character death are higher. Players are only better off it they're going to be dying all the time anyway from mobs with way out of scale attack bonuses when 10 = always hit but a D20 might evade something.

Right now there is sucha discrepency in ACs amongst my party that some of them are hit on a two and others arent hit on a nineteen. I think a little more randomization would make the game better for both of them.

[exelsisxax]

Right now there is sucha discrepency in ACs amongst my party that some of them are hit on a two and others arent hit on a nineteen. I think a little more randomization would make the game better for both of them.

I believe that addressing the underlying reason for a vast disparity in the game's primary defensive attribute would work better.

How do they have a range of at least 18 for that stat? Are they working at wildly different optimization levels?

[Ualaa]

If you're rolling the reverse of an AC, it isn't going to change the odds dramatically.
Rolling the die is very slightly advantageous, compared to setting the DC at 10 plus bonuses.



Normal situation: The mob rolls to Hit.
The player has 20 AC (static 10, plus 10 in armor class bonuses).
The mob has a +6 attack bonus (random d20, plus 6 in attack bonuses)

The mob rolls d20 +6, and needs to get a total of 20 or higher to hit.
If the roll is:
-- 01-13 (65% chance), the player is missed.
-- 14-20 (35% chance), the player is hit.

A tie goes to the attacker; if they hit the AC exactly, it is still a hit.
Because the mob is rolling the d20, it has an extra 5% in its' favor.



Reversal: The PC rolls to Avoid Being Hit.
The mob has a 16 attack (static 10, plus 6 in attack bonuses).
The player has a +10 avoidance bonus (random d20, plus 10 avoidance).

The player rolls d20 +10, and needs to beat a DC 16 to avoid being hit.
If the roll is:
-- 01-06 (30% chance), the player is still hit.
-- 07-20 (70% chance), the player dodges.

A tie goes to the attacker; if the avoidance roll equals the attack, it is still a hit.
Because the player is rolling the d20, they have an extra 5% in their favor.

[Vaz]

It also depends what you mean by underdog.

If you mean the person with a drastically lower AC than the opponent, then sure, in the math bods above can work it out.

The trick comes when how you want to apply that in game. The game is about min-maxing. If before, a suit of Full Plate Armour provided a bonus of +8, whereas that would have essentially been a +80% bonus, and they're now sat at an AC of 20 wearing their full plate with a shield. In the new circumstance, they have an equal chance of rolling above 20, as they do a 20 or below, but they also have a chance of completely fluffing their armour "roll" and only getting an 11. And that is with one of the highest levels of nonmagical armour in the game. It also does bad things to Combat Expertise, and certain defensive actions. Why should people increase their AC overmuch, when it might just be tanked due to an unlucky roll?

It's not as though AC is overpowered at the moment.

It also has the additional "benefit" of making combats last a bit longer, as you roll yet another dice.

If you want to test it out, create some BBEG's, with low armour rating, and some that are similar, but with a high rating, and rather than adding +10 to the AC of the gear, you roll a d20 and add that. It's the simplest and least intrusive way of dealing with things.

[Ualaa]

It (slightly) favors whoever rolls the d20.
Being an underdog or not being the an underdog doesn't matter.

Normally, the math gives a 5% edge to the players when they attack the mobs.
And it gives a 5% edge to the mobs, when the players are being attacked.

If you have the players attack normally, they retain the slight edge for attacks.
If you then have the players also roll their avoidance, they get the slight edge (they roll the d20) there too.

There isn't really a point in having both sides roll to avoid; if that were the case, just leave the default system in place.

[Ualaa]

8 AC is still 8 AC.
Whether rolling to hit, or rolling to avoid being hit.

Normal.
10 Base AC + 8 extra from armor.
Mob has a +5 attack bonus.

Without the armor on...
If the mob rolls:
01-04 (20% chance) they miss.
05-20 (80% chance) they hit.

With the armor on...
If the mob rolls:
01-12 (60% chance) they miss.
13-20 (40% chance) they hit.

The armor reduces the chances of being hit by 40%, which is 8/20.
Because the attacker is rolling to hit, they get the 5% edge.



Reversing things.
Avoidance roll is d20 +8 from armor.
Mob has an attack roll is 15.

Without the armor on (d20 +0, vs DC 15)
If the player's avoidance roll is:
01-15 (75% chance) the mob hits.
16-20 (25% chance) the mob misses.

With the armor on (d20 +8, vs DC 15)
If the player's avoidance roll is:
01-07 (35% chance) the mob hits.
08-20 (65% chance) the mob misses.

The armor reduces the chances of being hit by 40%, which is 8/20.
Because the defender is rolling to avoid being hit, they get the 5% edge.

[Ualaa]

Plenty of demonstrations about mathmatics already.

What hasn't been said: this will not benefit your players. When monsters eat x4 crit damage hits, the game doesn't break. They're supposed to die. If a PC gets hit with a x4 crit confirmation from any meaningful enemy, it absolutely sucks for the player. There's a reason that natural weapons have minimum crit ranges and most civilized enemies aren't print as wielding keen scythes or heavy picks and improved critical.

Similarly, rolling for AC significantly increases the odds for a lucky 2-3 hit string that kills a PC. An attack can roll high, OR the PC can roll low. AC is worthwhile in its reliability, if you randomize a significant amount of it the volatility will seriously increase the odds of your players getting killed without reason or significance. Unless the campaign is pretty easy, the odds of character death are higher. Players are only better off it they're going to be dying all the time anyway from mobs with way out of scale attack bonuses when 10 = always hit but a D20 might evade something.

Crit chance, chance to confirm the critical, and critical multiplier does not change.

If the mob threatens a critical on 19 or 20, in the normal system.
And deals x4 damage on a critical.

In the default system, a confirmed 19-20 deals x4 damage.

In the avoidance system, a confirmed initial 01-02 deals x4 damage.



If a rogue with a rapier threatens 18-20 (three numbers) and x2 damage.
In the avoidance system, they threaten a critical if your roll is 01-03 (three numbers) and x2 damage.

In the default system, the attacker rerolls and just need to hit.
In the avoidance system, the defender rerolls and just needs to be hit.

[Jormengand]

Ualaa, the alternate system is that both roll, not just the defender. The d20 replaces the 10 base AC that all creatures get just for existing.

[Vaz]

It (slightly) favors whoever rolls the d20.
Being an underdog or not being the an underdog doesn't matter.

Normally, the math gives a 5% edge to the players when they attack the mobs.
And it gives a 5% edge to the mobs, when the players are being attacked.

If you have the players attack normally, they retain the slight edge for attacks.
If you then have the players also roll their avoidance, they get the slight edge (they roll the d20) there too.

There isn't really a point in having both sides roll to avoid; if that were the case, just leave the default system in place.

8AC is 8AC. Until that 18AC that they would have had beforehand is obviated by that dice roll is nothing (and a coin toss to make it worse than what it is). It just means that characters won't invest in armour any more, especially given how expensive AC can be to increase. If a character with an AC of 18 would end up getting an AC of 9-28, then why bother with the 8AC. Sure, it gives it a higher median, it's just another dice roll which can gimp your character that you've invested a lot of resources in. Why not just ignore it, and mitigate it in other ways like Miss chances, and rely on having an AC of (Dex+1) to (Dex+20)?

It hurts more heavily armoured targets, and if the party is the more heavily armoured, it hurts them, in which case, I'd suggest just giving the BBEG (or whoever you intend the underdog) the +d20 to AC in place of the static +10. If the party is struggling with getting hit too easily, well, that's their fault for building their characters that way. Not having rope when climbing a mountain can be a severe problem. If the party are getting severely battered, and really can't cope, then you're giving them too many difficult encounters, and giving them a random chance to have a higher AC isn't going to fix that.

There are plenty of resources out there to help with hitting high AC targets if that's what the party are struggling with also; up to the extends of a Wand of Wraithstrike and a Wand Chambered weapon of the beatsticks favourite weapon conveniently left for them.

[Ualaa]

We're obviously talking about different optional rules, for rolling your defense.

I am referring to 'Players Roll All The Dice', page 133 of Unearthed Arcana.

Only the player rolls to determine if their character hits.
Only the player rolls to determine if their character avoid being hit.

There is no opposed roll.

The initial post did not refer to one system or another.
My posts have consistently assumed only the player rolls.

There might be another system somewhere, with an opposed roll.
And sure, if you want twice the randomness in your game, you can use it.

8 AC is still +40% in your favor, whatever the system you use.
If the mob threatens a x4, on a confirmed critical, rolling defense or not does not impact how deadly the mob's attack is; that is fixed irregardless of whether the mob rolls to hit or you roll to avoid being hit, with a +/- 5% difference in the odds of being hit, but no change in the devastation or lack thereof of a given hit.

[weckar]

The one Obvious consequence I see (and for some reason my spellchecker thinks Obvious is a German Noun today), is that it makes attacks dependent on two die rolls - it makes them slower but also puts them on a curve.

[Eldariel]

One thing to remember is that this game has reroll mechanics, which make any kind of a player-controlled roll more powerful than just the paper values. Might be they'll just retry on a critical bad roll.

[Gemini476]

I believe that addressing the underlying reason for a vast disparity in the game's primary defensive attribute would work better.

How do they have a range of at least 18 for that stat? Are they working at wildly different optimization levels?

While I obviously don't know the particulars, it's not too hard to have that kind of disparity in ACs - especially not when there's some classes that get up to +12 AC from nonmagical armor/shield alone. An unoptimized Fighter with 12 Dex, +2 full plate and a +1 tower shield has AC 26 for the cost of 6530gp - let's say that the enemy in question has an attack bonus of +8, so they hit the Fighter on an 18.

This means that the other character in question has AC10, meaning that they're unarmored. An unoptimized Wizard whose magic item budget and spell allocation is in non-AC stuff, for example.

Or, for a dumbish unoptimized high-level scenario, a Fighter with +23 AC over the Wizard from having +5 mithril armor/shield and all the magic items the Wizard also has.

Now, in an actually optimized scenario the Wizard is going to have all kinds of tricks to bring their AC up to respectable levels (just for starters, Bracers of Armor being that lead down to +15 and Shield brings it down to +11), but it's not that hard to have two characters with wildly different ACs.


Remember, the Bracers of Armor and Amulet of Natural Armor are in theory supposed to compete with, say, the Bracelet of Friends and Necklace of Fireballs. And the Wizard probably isn't intended to keep Mage Armor up 24/7.
It just didn't really work out that way, hence the Magic Item Compendium's rebates for those effects.

[Fizban]

Randomness always favors the underdog. The underdog only cares about this one fight, not the averages, because they're probably going to die anyway. More randomness means more chances for them to get a lucky break.

The players are not the underdog except maybe during the occasional super boss fight. Randomized AC is not as lethal to the PCs as giving all your antagonists x4 crit weapons, but it is extremely annoying and doubles all attack rolls.

Note the prevalence of Power Attack and it's interactions with this where no one knows what's going on.

[exelsisxax]

While I obviously don't know the particulars, it's not too hard to have that kind of disparity in ACs - especially not when there's some classes that get up to +12 AC from nonmagical armor/shield alone. An unoptimized Fighter with 12 Dex, +2 full plate and a +1 tower shield has AC 26 for the cost of 6530gp - let's say that the enemy in question has an attack bonus of +8, so they hit the Fighter on an 18.

This means that the other character in question has AC10, meaning that they're unarmored. An unoptimized Wizard whose magic item budget and spell allocation is in non-AC stuff, for example.

Or, for a dumbish unoptimized high-level scenario, a Fighter with +23 AC over the Wizard from having +5 mithril armor/shield and all the magic items the Wizard also has.

Now, in an actually optimized scenario the Wizard is going to have all kinds of tricks to bring their AC up to respectable levels (just for starters, Bracers of Armor being that lead down to +15 and Shield brings it down to +11), but it's not that hard to have two characters with wildly different ACs.


Remember, the Bracers of Armor and Amulet of Natural Armor are in theory supposed to compete with, say, the Bracelet of Friends and Necklace of Fireballs. And the Wizard probably isn't intended to keep Mage Armor up 24/7.
It just didn't really work out that way, hence the Magic Item Compendium's rebates for those effects.

All your examples are of wildly different optimization levels. If that is the issue, rolling for AC is only going to increase the disparity.

[sleepyphoenixx]

I wouldn't do it simply because it adds even more rolling to combat. The difference isn't great enough to warrant the extra effort needed.


I believe that addressing the underlying reason for a vast disparity in the game's primary defensive attribute would work better.

How do they have a range of at least 18 for that stat? Are they working at wildly different optimization levels?

It's pretty easy getting a range like that when some people don't care about AC and don't invest in it at all.
There's no reason to get just a few points after all - if your enemy can't miss on anything but a 1 either way you're getting nothing unless you plan on tanking full attacks (iterative, non-natural ones at that).

AC gets expensive fast unless you're getting a good base from mundane armor/shield/dex bonus, long-duration spells or some kind of stat-to-AC feature like the Monk's.
Relying purely on magic items for that breaks your WBL something fierce.

A spellcaster can usually get by just fine on miss chance and Mirror Image until True Seeing becomes common. There's the occassional enemy with Blindsight, but there's ways around that too.
That option is a lot cheaper, either on gold or spell slots, than getting effective AC so you can spend your gold on other things when WBL is still tight.

[Jay R]

At first glance, it's just a +.5 to AC. But that ignores the fact that a 1 is always a miss, and a 20 is always a miss.

So ignoring Keen blades, improved crit, etc., and assuming that the only automatic hits are 20s, it is the equivalent of the following change to the number needed to hit somebody. So if a character is usually hit on a 2 or better, this change is equivalent to being hit on a 4.75 or better. (Because there are now more than 400 different roll results instead of 20, the results don't come out evenly.)


2 --> 4.75
3 --> 5.3
4 --> 5.9
5 --> 6.55
6 --> 7.25
7 --> 8
8 --> 8.8
9 --> 9.65
10 --> 10.55
11 --> 11.45
12 --> 12.35
13 --> 13.2
14 --> 14
15 --> 14.75
16 --> 15.45
17 --> 16.1
18 --> 16.7
19 --> 17.25
20 --> 17.75

Note that the rules change helps everybody who can be hit on a roll less than 14, and hurts everyone who can't be hit on a 14.

[Ualaa]

In our last campaign, Rappan Athuk (FGG, Pathfinder version), there were vast AC discrepancies at times.
Way beyond a range of 18.

At the time, a Sorcerer/Summoner (Gestalt) figured his AC would suck regardless so he didn't even bother with Mage Armor, Shield, or anything like that.

A front-line melee was, while defensively fighting, sitting at 56 AC.
The range was 56-13 = 43...
The party average might have been around 35, at that point.

They both died with several iterations of those builds, before moving onto other builds.

The no AC guy was hit by anything.
A real glass cannon... plus an eidolon that was used to discover by triggering traps (and usually die), while the scout was quietly checking another direction.

The no armor, all dodge/defensive guy would scout without trapfinding skills (noisily getting the scout ambushed/killed a few times too) and/or get surprised; defensive fighting requires him to swing, so while flat-footed his AC was around 20.
If he had stayed with the group, near the middle, opening action charge/engage including defensive fighting for the AC boost, he would have been virtually immune to AC targeting attacks (with Improved Evasion) and great saves too.

[Talakeal]

I wasn't actually looking for a solution to my current campaign, we aren't even playing right now, but I am getting ready to form a new group and was thinking about which optional / house rules to use.

Honestly discrepancies in AC have been a big problem in every D&D game I have ever played in past a certain level. There is always someone who can't be hit unless the DM throws a monster at the party which then auto hits everyone else. The idea is that I could use more normal monsters who would still hit the high AC guy roughly the same percentage of the time without auto hitting the low AC people all the time, it is an overall buff to player survivability.

However, it would indeed bring in a bit more randomness and cause AC to have "diminishing returns" at high levels; I don't see this as a bad thing, but apparently a lot of people do. I can't help but feel that some of the reactions border on a little childish, but that is the nature of human psychology and I can't ignore it.

Also, it doesn't just work for the PCs, it applies both ways. I remember one game where one of the players found out she needed a natural 20 to hit the big boss monster and she literally got up from the table and went in the other room to play video games and told us to call her back when we got to a fight in which we could contribute. Ideally such a rule would help in situations like that as well.

But, the reaction seems overwhelmingly negative, and my players tend to be whinier than most imo, so I guess its back to the drawing board to look for a more heavy handed solution.


We're obviously talking about different optional rules, for rolling your defense.

I am referring to 'Players Roll All The Dice', page 133 of Unearthed Arcana.

Only the player rolls to determine if their character hits.
Only the player rolls to determine if their character avoid being hit.

There is no opposed roll.

The initial post did not refer to one system or another.
My posts have consistently assumed only the player rolls.

There might be another system somewhere, with an opposed roll.
And sure, if you want twice the randomness in your game, you can use it.

8 AC is still +40% in your favor, whatever the system you use.
If the mob threatens a x4, on a confirmed critical, rolling defense or not does not impact how deadly the mob's attack is; that is fixed irregardless of whether the mob rolls to hit or you roll to avoid being hit, with a +/- 5% difference in the odds of being hit, but no change in the devastation or lack thereof of a given hit.

Page 25 of the DMG under defensive rolls.

[exelsisxax]

Why do your players have this variation in AC? Is it optimization, specialization, or do some people just make no attempt to not get hit?

[Talakeal]

Why do your players have this variation in AC? Is it optimization, specialization, or do some people just make no attempt to not get hit?


People tend to focus on certain aspects of their character and go all out improving those aspects. Some people choose AC as one of those aspects, some people don't. Simple as that.

[Vaz]

Someone who goes to the depths of optimizing their AC should be allowed to reap the benefits of that, in as much as someone who specializes in longbows should be able to once in a while get to use their range.

That said, you've also got to be able to teach the party that they can't just rely on one shtick. A Sorcerer lobbing fireballs might be fine for killing hordes of tightly packed orcs in a pitched battle, but when they're up against a lone target, that damage quickly peters out now that they can't drop 20 foes at once, even more so when that target has a resistance, or even immunity to fire. A Sorcerer in that instance would either need to use a different type of elemental damage, or pick up Searing Spell metamagic.

Similarly, a tank fighter is going to actually have to earn the right to have his AC be so useful. If he's invulnerable, then even an animal intelligence creature will attack the weaker members of the group, and he needs to "pull" or "do aggro". No point being a tank if nothing's hitting you.

If someone wants stupid high AC, let them. There are plenty of ways of getting around that; touch attacks, for example leave the fighter in the same, if not worse off than other party members, as they pump their unemcumbered Dex higher. Alternatively, an AoE attacker (again, putting strain on that reflex/dex).

I'm not entirely sure as to what you're referring when you're saying things are childish.

When you've got players walking out of games you're playing (as a DM, I assume), and you're making someone who is attacking in melee with no other recourse only attack with a natural 20, you need to get your priorities straight as a DM. The first one is DON'T PLAY MONSTERS WITH SUCH HIGH DIFFICULTY AC. This is capsed, because hopefully, it makes sense, and it is ever more obvious. Seriously. If you're a DM, and your melee dudes can't hit things, and your players aren't having fun, you are a terrible DM. It is not YOUR game, and neither is it the players game. It is EVERYONE'S game, and EVERYONE has to have fun.

If the player flounced off, but could have done other things, then that player needs a stern talking to about sucking one up, and that things cannot always tailor themselves to the lowest common denominator of a team. If you're a DM, and you're forcing opponents with a AC's of 30+ against a party of Sub-level 10's, then you're doing something wrong.

[sleepyphoenixx]

Honestly discrepancies in AC have been a big problem in every D&D game I have ever played in past a certain level. There is always someone who can't be hit unless the DM throws a monster at the party which then auto hits everyone else. The idea is that I could use more normal monsters who would still hit the high AC guy roughly the same percentage of the time without auto hitting the low AC people all the time, it is an overall buff to player survivability.

The way to deal with that is to target non-AC defenses too. Or touch AC if you want.
Spellcasting enemies generally do both, as do many monster abilities.
And if your players don't get the idea that they should find a functioning balance between their various stats you can mention the idea to them.
The other way would be to simply replace quality with quantity. A 20 is always a hit, no matter how much AC they have.

If some players choose to disregard AC entirely that's their prerogative. But they have no room to complain later if they get killed because they didn't manage to avoid getting hit.
The same applies to those players who think "tank" is a valid archetype in D&D and focus on AC at the cost of everything else - let him enjoy being the unhittable melee monster. But he'd better have a plan in place to deal with all the other attacks too, or he won't last long beyond level 3 or so, if that.

There's also the fact to consider that after some point more AC is just a number.
If you get enough AC that even a CR +4 dragon can only hit you on a 20 that simply means that in most encounters a big chunk of your AC is wasted, because most monsters don't have that many full-BAB HD for their CR.

[exelsisxax]

People tend to focus on certain aspects of their character and go all out improving those aspects. Some people choose AC as one of those aspects, some people don't. Simple as that.

Then you are helpless to do anything about this (short of doing some sort of hard bounding on things) and should not try. They are intentionally creating this discrepancy.

[Talakeal]

Then you are helpless to do anything about this (short of doing some sort of hard bounding on things) and should not try. They are intentionally creating this discrepancy.

That's an awfully defeatist attitude. Its not like this is an isolated problem, I have seen it occur in every group I have ever been in and in every game we have ever played. Shadowrun was actually the worst, far worse than D&D ever was.

If the game has an optional rule that might solve this problem, I don't see why it is unreasonable to look into it.


Someone who goes to the depths of optimizing their AC should be allowed to reap the benefits of that, in as much as someone who specializes in longbows should be able to once in a while get to use their range.

That said, you've also got to be able to teach the party that they can't just rely on one shtick. A Sorcerer lobbing fireballs might be fine for killing hordes of tightly packed orcs in a pitched battle, but when they're up against a lone target, that damage quickly peters out now that they can't drop 20 foes at once, even more so when that target has a resistance, or even immunity to fire. A Sorcerer in that instance would either need to use a different type of elemental damage, or pick up Searing Spell metamagic.

Similarly, a tank fighter is going to actually have to earn the right to have his AC be so useful. If he's invulnerable, then even an animal intelligence creature will attack the weaker members of the group, and he needs to "pull" or "do aggro". No point being a tank if nothing's hitting you.

If someone wants stupid high AC, let them. There are plenty of ways of getting around that; touch attacks, for example leave the fighter in the same, if not worse off than other party members, as they pump their unemcumbered Dex higher. Alternatively, an AoE attacker (again, putting strain on that reflex/dex).

I'm not entirely sure as to what you're referring when you're saying things are childish.

When you've got players walking out of games you're playing (as a DM, I assume), and you're making someone who is attacking in melee with no other recourse only attack with a natural 20, you need to get your priorities straight as a DM. The first one is DON'T PLAY MONSTERS WITH SUCH HIGH DIFFICULTY AC. This is capsed, because hopefully, it makes sense, and it is ever more obvious. Seriously. If you're a DM, and your melee dudes can't hit things, and your players aren't having fun, you are a terrible DM. It is not YOUR game, and neither is it the players game. It is EVERYONE'S game, and EVERYONE has to have fun.

If the player flounced off, but could have done other things, then that player needs a stern talking to about sucking one up, and that things cannot always tailor themselves to the lowest common denominator of a team. If you're a DM, and you're forcing opponents with a AC's of 30+ against a party of Sub-level 10's, then you're doing something wrong.

Again, the goal is not to nerf the high AC character, but to throw the low AC guy a bone. Ideally the high AC guy would be hit roughly the same as they already are, but the low AC guys would have a chance of avoiding a hit.

The childish attitude's are people cutting off their noses to spite their face (a few posters seem to have had the attitude that if AC has diminishing returns past a certain point they would just forget about it at all and go into battle naked) or being a bit hypocritical.

For example, you seem to be saying that it is entirely the DM's fault no matter what the problem is. If a PC comes into the game with a 50 AC and a +0 Attack Bonus the player has done nothing wrong, but it is the DM's responsibility to rewrite the entire campaign world so that they never encounter anything that has a 21+ AC or that needs to roll to hit to challenge the PCs.

Obviously in real life there are going to be fights where certain tactics don't work (I think we both agree on this), but I don't think you need an asymmetrical system where the DM using non-standard tactics is a given but the players are rarely if ever forced to. I would personally like to be using a system where you can use most of the monster manual against most parties of PCs and have a variety of acceptable tactics for most matchups, and where "we attack!" is a viable strategy for one side or the other more often than not.


So, some more details about the player walking out:
I was not the DM.
The player who walked out was playing a "tank" who had put all of her stats into Con and Dex and all of her money into better armor and defensive magic items.
The DM was new, and he came to me (an experienced DM) before the game and asked if it was ok to use the monster with such a high AC. I said yes, because we can hit him fairly easily with non-standard tactics and he doesn't have enough HP or damage output to make it an unfair fight if we use our brains.
The knight player rolled to hit, the DM told her the AC she needed, and she left.
Myself (a barbarian) and the rogue flanked the monster, the wizard and the cleric used the aid another action on us, and we made short work of the monster.
The DM never ran another game.

[exelsisxax]

It's not defeatist, it's being logical. If some players don't care about not getting hit, they will have terrible AC and you can't do anything about it.

Unless you have miscommunicated the issue, the players the players intentionally did something in building their characters to create this situation, and they are fine with it.

Are they NOT fine with it? Have your players complained that AC is all over the place in the party(even if they didn't realize it's because they made the PCs wrong)? The only player you've said made a fuss about anything wasn't in your game and had no right to kick up a fuss anyway.

[Ualaa]

I first played with a group, where the DM would do their own thing.

If three players had varying levels of armor/defenses.
The no defense/AC guy would be hit on a 06+.
The medium armor guy would be hit on a 10+.
The high armor guy, would only be hit on a 14.
I'm not sure the exact numbers, but it was close enough to that.

Challenges were what the DM figured was level appropriate.
In Advanced 2nd edition, it was easy to approximate but some things were easy and some hard.
Two equally rewarding (experience) fights could be vastly easier/harder.

Similarly, our damage was relative.
The minion equivalents (like a 4th edition minion) might die on a second hit.
The main guys might not die for a few rounds.
And the boss would last longer.
He didn't use hit points all the time, even if the mob always had a hit point total.
A boss mob, would not fall in round two or three.

His stories were interesting and the campaigns rocked.
And this stuff was going on behind the screen, so we weren't aware of it at the time.
Or even knowledgeable on the DM aspect of the rules.
I probably wouldn't want to play this style, now that I know how things are supposed to work, and have more than 25 years of experience as a DM, but at the time it was fun.

[apocryphaGnosis]

I was running a short campaign with a bunch of variant rules a while ago. My group tends to get pretty distracted and uninvolved when its not their turn, so I liked the idea of the players having some sort of interaction, even on enemy turns. Obviously though, as has already been said adding a d20 to AC is really swingy and favors enemies more than players, so what I ended up doing was having enemies have 10 + AC mods as usual, but players base AC is 5 + x, and then they add a d10 to that, so they have something to do which adds a little randomness, but isn't out of control. They all liked the system a lot, I recommend it highly.

[Talakeal]

Its kind of weird, over here everyone is saying that more randomness always hurts the players and invalidates their choices, yet on the 5E thread everyone is constantly extolling the virtues of bounded accuracy, which is just a (far far more extreme) way of doing the same thing.

[exelsisxax]

Its kind of weird, over here everyone is saying that more randomness always hurts the players and invalidates their choices, yet on the 5E thread everyone is constantly extolling the virtues of bounded accuracy, which is just a (far far more extreme) way of doing the same thing.

Bounded accuracy doesn't make you roll to get your STR bonus. It just puts a cap on it. Doesn't work the same way at all.

[Talakeal]

Bounded accuracy doesn't make you roll to get your STR bonus. It just puts a cap on it. Doesn't work the same way at all.

That's an extremely weird way of looking at it. You still "get" your armor class no matter what, you just have to roll a certain thing for it to matter.

The only time you don't "get" your score is when it is so high or so low that it can't fail on a 1 or succeed on a 20, which is exactly the sort of situation that bounded accuracy removes.

For example, in 3.X I could easily make a character with an Armor Class so high nothing in the monster manual can hit me except on a 20. In 5e that is impossible, my armor class will always be at a point where level appropriate enemies can hit me with a decent roll, so using your logic that means I have to "roll to get my AC".

Edit: In short, the idea of bounded accuracy is making sure that the numbers don't overwhelm the RNG element. In 5E they do this by making the numbers smaller but keeping the d20 range the same. The 3.5 defensive role variant does this by keeping the numbers the same but increasing the range of the RNG. They do the same thing for the same reason, they just come at it from opposite directions.

[sleepyphoenixx]

Again, the goal is not to nerf the high AC character, but to throw the low AC guy a bone. Ideally the high AC guy would be hit roughly the same as they already are, but the low AC guys would have a chance of avoiding a hit.
If you want to throw the low AC guy a bone drop AC boosting items as loot. Preferably cheap ones that don't have much impact on WBL. Personally i like dropping my players a Shadow Cloak (DotU) if they don't buy it themselves.
People can get AC if they want, relatively easily. Some just choose not to.


The childish attitude's are people cutting off their noses to spite their face (a few posters seem to have had the attitude that if AC has diminishing returns past a certain point they would just forget about it at all and go into battle naked) or being a bit hypocritical.
The issue is that getting just a few points of AC generally does nothing. If your AC does not at least match the enemies hit chance plus two you may as well have AC 0.
And getting enough AC to decrease the chance of an enemy that actually fights in melee hitting gets progressively more expensive as levels get higher.

Let's go with an example: An adult black dragon is CR 11, which means you'll likely face him at ECL 10 or even lower unless your DM uses dragons as mooks.
Said dragon has a +24 to hit. so you need to have at least AC 26 before AC does anything at all for you. Assuming the dragon doesn't buff.
That's a pretty tall order for most level 10 characters who don't wear heavy armor/shield and also wants to spend WBL or spell slots on other things if your group isn't much into optimization, and that's just the bare minimum.
Someone who intends to tank a full attack had better have AC of at least 37-38.

Anyone who doesn't expect to fight at melee range is generally better off just making sure he can survive at least one hit and staying away from full attack range instead of spending all his resources on trying to reduce the dragons chance to hit.

Now dragons are the extreme end of the range, having full BAB, generally high strength and high HD for their CR. But the basic principle holds.
Monsters who are actually dangerous in melee tend to have high to-hit bonuses.


So, some more details about the player walking out:
I was not the DM.
The player who walked out was playing a "tank" who had put all of her stats into Con and Dex and all of her money into better armor and defensive magic items.
The DM was new, and he came to me (an experienced DM) before the game and asked if it was ok to use the monster with such a high AC. I said yes, because we can hit him fairly easily with non-standard tactics and he doesn't have enough HP or damage output to make it an unfair fight if we use our brains.
The knight player rolled to hit, the DM told her the AC she needed, and she left.
Myself (a barbarian) and the rogue flanked the monster, the wizard and the cleric used the aid another action on us, and we made short work of the monster.
The DM never ran another game.
So the player never put any effort at all into being able to hit things and then complained that she wasn't able to hit things? :smalltongue:

[Ualaa]

A tank generally has full BAB.

If they're all DEX & CON, then I hope they took the equivalent of Weapon Finesse.
That's not saying that they did, but +0 from STR added onto +10 from BAB, that's not a high total.
Sure they should at least have a Masterwork Weapon.
And that's at the very least.
Hitting is generally rather important.

If I'm playing anything that is not at animal intelligence or lower.
And I have something that never hits me.
And that I have a very hard time hitting.
And I have something else that is hitting hard enough to hurt.
It's absolute idiocy for me to ignore the thing killing me, to hit the thing that cannot hurt me.

Sure the mindless thing might go for whatever is closest.
And the animal intelligence may go for closest, until something hits it a lot harder.
But a large portion of the mobs in the game are going to ignore the *zero* offense character, for the one that hits hard.

I'd expect:
16-18 attack stat, after racial adjustments at level 1.
A stat up at 4th and 8th, into that stat.
At least a +2 item, probably a +3 item for attacks.
Probably +2 maybe +4 to the attack stat from a belt (so another +1 to +2).

So:
BAB +10
Stat +5/+7
Item +2/+3
Flanking +2
Somewhere in the neighborhood of +19 to +22 to hit.
That from a fairly average tanking character.

If they're instead sitting at:
BAB +10
Stat +0
Item +1
Flanking +2
And have +13... that's their problem, if they cannot hit.
Most mobs would likely ignore them too... to kill the dangerous companions.

If the tank cannot hit, with the first set of bonuses, they might have cause to be a bit upset.
But not every fight is going to be as easy as another.
If the tank has the second set of bonuses to hit...
Well, they'd be crazy to assume they can hit anything approaching their level.
Walking out is an option, but it's not necessarily a good one.
If you bring up an issue, and it continues despite this, then decide if you're having fun or not.
But without at least letting the other party know you have an issue...
Whatever... they're probably not worth having at the table.

I don't blame the rookie DM in that situation.
Nor the veteran DM who gave the advice.
The rest of the party showed the fight was balanced enough... they won, despite being a character down.

[Fizban]

Again, the goal is not to nerf the high AC character, but to throw the low AC guy a bone. Ideally the high AC guy would be hit roughly the same as they already are, but the low AC guys would have a chance of avoiding a hit.
The low AC guy doesn't need/deserve a bone. He tanked AC, he didn't stay out of range, he pays the consequences. Same way someone who tanks their reflex save gets wrecked by fireballs or the guy who tanks his will save gets mind controlled. Or, if he knows what he's doing, he never needs the AC. Making a massive, cumbersome change to game mechanics in order to maybe slightly benefit one person who doesn't need it hits about every point of bad design.

As for bounded accuracy, yeah I'm not sure what those people are talking about with attack rolls in specific. Might have something to do with the fact that while attacks may be more random in theory, there's no iterative attack penalties and the attack bonuses that ignore the bound are so much more powerful it easily makes up the difference, and being able to make all your attacks every round meaning you get more overall (compared to a similarly unoptimized 3.5 character). Could also be regarding various minions/ "NPCs" and summoned monsters, who are a lot more relevant with bounded accuracy, or the simple. Actually it's probably back to the unoptimized one: even if your character is completely backwards they're at most 5 points behind on their d20 rolls/defenses. Either way, the rest of the game is significantly less random so even if attack rolls are less reliably accurate at high levels than they used to be, it's compared to a much smaller number of spells.

I'd want to hear more before really weighing in on the tank-only PC vs tank monster, as-is I'd simply rule against the whole table for not communicating their expectations/combat options well enough to avoid the problem entirely/get through it after the fact.

[Jay R]

Honestly discrepancies in AC have been a big problem in every D&D game I have ever played in past a certain level. There is always someone who can't be hit unless the DM throws a monster at the party which then auto hits everyone else.

The various solutions include area effects which AC doesn't stop, or attacks stopped by saving throws, or traps, or rust monsters, etc.

The super-high AC doesn't help when you step in the quicksand. It doesn't stop a sleeping gas or a mind-control spell.

If most encounters are just slugfests, then the people with the best defenses against slugging have been given a huge advantage by the DM.

In my game, if a single character has been mostly unthreatened over a complete adventure, then in the next adventure I try to throw that player a bone by providing him a thrill, in the form of a threat that is more dangerous to him. The most obvious such threat is anything with a Will save, since the classes wearing the most powerful armor are usually weak at Will saves, but anything will do. Rust monsters are particularly good sometimes.

[exelsisxax]

The various solutions include area effects which AC doesn't stop, or attacks stopped by saving throws, or traps, or rust monsters, etc.

The super-high AC doesn't help when you step in the quicksand. It doesn't stop a sleeping gas or a mind-control spell.

If most encounters are just slugfests, then the people with the best defenses against slugging have been given a huge advantage by the DM.

In my game, if a single character has been mostly unthreatened over a complete adventure, then in the next adventure I try to throw that player a bone by providing him a thrill, in the form of a threat that is more dangerous to him. The most obvious such threat is anything with a Will save, since the classes wearing the most powerful armor are usually weak at Will saves, but anything will do. Rust monsters are particularly good sometimes.

But why rust monsters? "hit the reflex save" can give your players different challenges. Rust monsters is more of a "martials doing well?" TAKE AWAY THEIR NICE THINGS! If there was a monster that stole multiple spell slots per hit on a touch attack, it'd still be far less of a threat. Weapons and armor don't come back after 8 hours rest.

[sleepyphoenixx]

If the tank cannot hit, with the first set of bonuses, they might have cause to be a bit upset.
If the tank cannot hit he generally still has other options. If all the enemies have AC that requires a 20 to hit, then he has cause to complain.
Especially against humanoid opponents you can still Bull Rush (which doesn't need a hit roll, just a pure strength check), Grapple (which only needs to hit touch AC + strength check) or Trip (which only needs to hit touch AC + strength check).
All of these are doable even without the feats, and chances are that an enemy who focused so much on AC won't have the same defenses against these tactics.
And that's not counting any other modes of attack you may have picked up via your build. There's a skill trick that lets you hit touch AC, various methods to make an enemy flat-footed, flanking, martial maneuvers and so on.
And that's only counting what the character himself brings to the table. Getting buffed by your partymembers is always another option, or at least it should be. As are consumables.

None of that helps you if you've dumped your main stat at character creation and haven't made any provisions to make up for that fact though.

The various solutions include area effects which AC doesn't stop, or attacks stopped by saving throws, or traps, or rust monsters, etc.

The super-high AC doesn't help when you step in the quicksand. It doesn't stop a sleeping gas or a mind-control spell.

If most encounters are just slugfests, then the people with the best defenses against slugging have been given a huge advantage by the DM.

In my game, if a single character has been mostly unthreatened over a complete adventure, then in the next adventure I try to throw that player a bone by providing him a thrill, in the form of a threat that is more dangerous to him. The most obvious such threat is anything with a Will save, since the classes wearing the most powerful armor are usually weak at Will saves, but anything will do. Rust monsters are particularly good sometimes.
See the above. All those are valid tactics for NPCs too, and a lot of monsters are really good at grappling especially. Your AC won't help you there.
Giving your NPC Fighters Dungeoncrasher or martial adept levels/Martial Training feats is a nice way to spice up an encounter. Even just a few of the more obscure fighter feats can offer some variety.
Bottom line is, don't give your melee monsters Weapon Focus or something like that. Pick feats that do stuff, even if said stuff is usually suboptimal in the hands of a player. It makes for more memorable encounters.

Also if most of your encounters are just slugfests you should work on your encounter design. There's tons of different tactics to use even at low levels. Explore your options.
Something like Teamwork Benefits (PHB2/DMG2) can make even a group of mook-level bandit archers threatening. Give your goblin band some cheap poison or a few alchemical items.

At higher levels i feel that any group that's supposed to be a threat should contain at least one spellcaster, or at least a monster with some SLAs or special abilities that target something besides AC.

[Jay R]

When a character has something really cool - a special ability, super-high AC, whatever - there are two principles to keep in mind:

1. There should be some encounters that is successfully overpowers. The player paid for the ability; it needs to be worth it.
2. There should be some encounters that it doesn't work, and either she has to think of something else, or some other PC gets to shine. Otherwise, the cool ability has effectively ended the game.

[Talakeal]

Either way, the rest of the game is significantly less random so even if attack rolls are less reliably accurate at high levels than they used to be, it's compared to a much smaller number of spells.


Really? I find that unless you are a rogue / bard / paladin the 5E skills and saving throws are way more random than combat in 5E or than just about anything in 3E.


If you want to throw the low AC guy a bone drop AC boosting items as loot. Preferably cheap ones that don't have much impact on WBL. Personally i like dropping my players a Shadow Cloak (DotU) if they don't buy it themselves.
People can get AC if they want, relatively easily. Some just choose not to.

Yeah, it is a choice. Every game I have played has used either a the base RAW (with WBL and magic item buying / selling) or an optional attunement system, in either case the only people who actually make any serious investments in AC boosting items are the characters who see themselves as a sort of a tank.

Most people choose to focus on a few roles for their character, and if defense isn't one of those rolls they ignore it. I was just thinking that the defensive roll rules might be a subtle way to bring in a bit of "bounded accuracy" while still letting people have whatever numbers they choose for their characters.

[exelsisxax]

Really? I find that unless you are a rogue / bard / paladin the 5E skills and saving throws are way more random than combat in 5E or than just about anything in 3E.



Yeah, it is a choice. Every game I have played has used either a the base RAW (with WBL and magic item buying / selling) or an optional attunement system, in either case the only people who actually make any serious investments in AC boosting items are the characters who see themselves as a sort of a tank.

Most people choose to focus on a few roles for their character, and if defense isn't one of those rolls they ignore it. I was just thinking that the defensive roll rules might be a subtle way to bring in a bit of "bounded accuracy" while still letting people have whatever numbers they choose for their characters.

If you want soft bounds don't play above level 12 and don't play gestalt. Very efficiently resolves a lot of the discrepancy.

[Talakeal]

If you want soft bounds don't play above level 12 and don't play gestalt. Very efficiently resolves a lot of the discrepancy.

Problem comes up well before level 12, even in E6 it can rear its ugly head. My first 3E character (a monk / bladesinger) had an AC well into the 40s at level 10.

Out of curiosity, what's the problem with gestalt? In my experience that makes characters more broad and versatile, not less.

[exelsisxax]

Problem comes up well before level 12, even in E6 it can rear its ugly head. My first 3E character (a monk / bladesinger) had an AC well into the 40s at level 10.

Out of curiosity, what's the problem with gestalt? In my experience that makes characters more broad and versatile, not less.

Gestalt makes things even more swingy. More possible class features to improve AC to ever greater heights, which only some players will use while the rest stay at normal AC ranges.

[Talakeal]

Gestalt makes things even more swingy. More possible class features to improve AC to ever greater heights, which only some players will use while the rest stay at normal AC ranges.

Not really seeing it; the only significant AC boosting ability I can think of which can't be replicated with a feat or minor magic item is shape-changing into a form with a high natural AC, and that isn't really helped by gestalting. I am sure there is some obscure combination of prestige classes and non core base classes which allow for such a thing, but I can't think of any of the top of my head and certainly haven't seen any in play.