What's the CR of a PC?

[Trandir]

Just as the title asks. What should be the CR of a monster with full class levels? This question should be valid only for monsters with CR<1 since a greater CR would undermine some of the class features

Exemple: an orc monster with 8 barbarian levels,
A noble with 5 sorcerer levels.


Bonus: what CR should have your average PC according to the rules to assign the CR?

[MaxWilson]

Just as the title asks. What should be the CR of a monster with full class levels and that is a playable race?

If you just run the mechanical numbers IME they tend to come out around CR (Level - 3) to (Level - 1), but IMO you might as well just call it CR (Level) because classes tend to bring benefits that are undervalued by the CR equation such as mobility and Athletics Expertise and summoning spells, and the DMG encourages you to tweak CR post-table to make it more reflect reality. So, I always just treat level and CR as synonymous and it works out pretty well. 4 CR N monsters vs. 4 level N PCs is a pretty "fair" fight, in the sense that it's hard to predict who will win without knowing what tactics will be employed.

[Yunru]

PCs don't have CR.
Either you're making a PC, or you're not.
If it's not a PC, work out its CR like any other monster.

[Trandir]

PCs don't have CR.
Either you're making a PC, or you're not.
If it's not a PC, work out its CR like any other monster.

I mean yes but that's no fun. I might have worded it quite badly.

If you want to by the rules then what should be tha CR of a noble with class levels?

[JackPhoenix]

I mean yes but that's no fun. I might have worded it quite badly.

If you want to by the rules then what should be tha CR of a noble with class levels?

Depends on what class and level it is. Also what weapon and armor he's wearing. Plate and greatsword using paladin would have higher CR than lore bard with shortsword and leather armor, due to better defensive and offensive values.

See the DMG for CR calculations.

[MoiMagnus]

A NPC with N hit point dices is a a monster of CR N*2/3.

This means that a PC of level N, if you ignore most of its class features and feats, would be a monster of CR N*2/3.
Taking in account class features and feats, and assuming the PC is optimized, Level = CR is not a bad approximation.

(But avoid PC vs PC fights, they usually work poorly as PCs have too much AC and saves, and not enough hitpoints.)

[JackPhoenix]

A NPC with N hit point dices is a a monster of CR N*2/3.

This means that a PC of level N, if you ignore most of its class features and feats, would be a monster of CR N*2/3.
Taking in account class features and feats, and assuming the PC is optimized, Level = CR is not a bad approximation.

(But avoid PC vs PC fights, they usually work poorly as PCs have too much AC and saves, and not enough hitpoints.)

HD has no direct correlation to its CR, besides being part of defensive CR calculations (and even then, it's not HD, but HP derived from the HD). As CR is average of offensive and defensive CR, and more than just HP go into defensive CR, you could have monster with high defenses (AC, saves, resistances...) and low HD, high HD and low defenses, high HD and defenses and low damage potential or low HD and defenses and high damage potential all at the same CR.

[stoutstien]

I mean yes but that's no fun. I might have worded it quite badly.

If you want to by the rules then what should be tha CR of a noble with class levels?

It's really hard to do any kind of conversion. Player classes tend to be more powerful in terms of damage output but less tough as far as staying power.

then you have the issue that players are designed around an adventuring day with a bunch of encounters possible where NPCs are design for mostly a single encounter.

when you start adding more than one feature to an NPC the math goes completely off the rails.

[Spiritchaser]

(But avoid PC vs PC fights, they usually work poorly as PCs have too much AC and saves, and not enough hitpoints.)

I would disagree with this.

PC vs evil N(PC) fights tend to be different, since the players face a relative glass cannon with lots of very powerful features and potentially explosive burst damage.

I’ve run a couple of campaigns with predominantly PC based major and moderate foes and they can work just fine... but they tend to run differently. The PCs quickly learn to try and get the drop on their foes wherever possible, and are far more likely to scout, study and fear their opponents.

The main argument I see against PC foes isn’t experience for the Players, It’s DM exhaustion. I find it more demanding to run foes with so many options.

It’s also pretty much mandatory to war game the encounters a bit. The interplay of so many options tends to make it hard to play more than one or two at a time without some forethought, and figuring a CR almost has to be done in the context of how everything the enemy has lines up with everything the PCs have

Running CR monsters is definitely easier. Don’t let that stop you from trying though.

If you want to give it a shot, find an in game way to give the players some warning, at least for the first pc they face

Run a few what if simulations to get a read on difficulty

If you do this a lot you can expect the players to evolve strong gank-squad tactics.

[BloodSnake'sCha]

PCs don't have CR.
Either you're making a PC, or you're not.
If it's not a PC, work out its CR like any other monster.

But you do have rules in the DMG for it.
And I think he is asking for help with those rules.


Using Classes and Levels
You can create an NPC just as you would a player character, using the rules in the Player’s Handbook. You can even use a character sheet to keep track of the NPC’s vital information.

Class Options. In addition to the class options in the Player’s Handbook, two additional class options are available for evil player characters and NPCs: the Death domain for clerics and the oathbreaker for paladins. Both options are detailed at the end of this chapter.

Equipment. Most NPCs don’t need an exhaustive list of equipment. An enemy meant to be faced in combat requires weapons and armor, plus any treasure the NPC carries (including magic items that might be used against the adventurers).

Challenge Rating. An NPC built for combat needs a challenge rating. Use the rules in chapter 9 to determine the NPC’s challenge rating, just as you would for a monster you designed.


Monsters with Classes
You can use the rules in chapter 3 of the Player’s Handbook to give class levels to a monster. For example, you can turn an ordinary werewolf into a werewolf with four levels of the barbarian class (such a monster would be expressed as “Werewolf, 4th-level barbarian”).

Start with the monster’s stat block. The monster gains all the class features for every class level you add, with the following exceptions:

The monster doesn’t gain the starting equipment of the added class.
For each class level you add, the monster gains one Hit Die of its normal type (based on its size), ignoring the class’s Hit Die progression.
The monster’s proficiency bonus is based on its challenge rating, not its class levels.
Once you finish adding class levels to a monster, feel free to tweak its ability scores as you see fit (for example, raising the monster’s Intelligence score so that the monster is a more effective wizard), and make whatever other adjustments are needed. You’ll need to recalculate its challenge rating as though you had designed the monster from scratch.

Depending on the monster and the number of class levels you add to it, its challenge rating might change very little or increase dramatically. For example, a werewolf that gains four barbarian levels is a much greater threat than it was before. In contrast, the hit points, spells, and other class features that an ancient red dragon gains from five levels of wizard don’t increase its challenge rating.

[NNescio]

If you just run the mechanical numbers IME they tend to come out around CR (Level - 3) to (Level - 1), but IMO you might as well just call it CR (Level) because classes tend to bring benefits that are undervalued by the CR equation such as mobility and Athletics Expertise and summoning spells, and the DMG encourages you to tweak CR post-table to make it more reflect reality. So, I always just treat level and CR as synonymous and it works out pretty well. 4 CR N monsters vs. 4 level N PCs is a pretty "fair" fight, in the sense that it's hard to predict who will win without knowing what tactics will be employed.

That's a 3~4x Deadly encounter (depending on level/CR), which is mostly likely to result in a TPK.

Imagine a 4-person Level 1 party up against 4x Dire Wolves or Bugbears.
A Level 2 party up against 4x Berserkers
A Level 3 party against 4x Basilisks or Bearded Devils
Level 5s against 4x Beholder Zombies
12s against 4x Archmages
13s against 4x Adult Dragons or Beholders

It's egregiously inaccurate.

[BloodSnake'sCha]

That's a 3~4x Deadly encounter (depending on level/CR), which is mostly likely to result in a TPK.

Imagine a 4-person Level 1 party up against 4x Dire Wolves or Bugbears.
A Level 2 party up against 4x Berserkers
A Level 3 party against 4x Basilisks or Bearded Devils
Level 5s against 4x Beholder Zombies
12s against 4x Archmages
13s against 4x Adult Dragons or Beholders

It's egregiously inaccurate.
It is fine for a tactical table that play a war game.

[Yunru]

But you do have rules in the DMG for it.
And I think he is asking for help with those rules.

You even quote the answer though?
You use the same metric to determine CR as every other monster.

[MaxWilson]

That's a 3~4x Deadly encounter (depending on level/CR), which is mostly likely to result in a TPK.

Imagine a 4-person Level 1 party up against 4x Dire Wolves or Bugbears.
A Level 2 party up against 4x Berserkers
A Level 3 party against 4x Basilisks or Bearded Devils
Level 5s against 4x Beholder Zombies
12s against 4x Archmages
13s against 4x Adult Dragons or Beholders

It's egregiously inaccurate.

I've actually run these types of combats. PCs are surprisingly robust. Who wins depends on which tactics are used. None of the fights you just listed are unwinnable, but they're hard enough that the PCs might not win... Which is pretty good evidence that the effective CRs are similar.

[GlenSmash!]

Just as the title asks. What should be the CR of a monster with full class levels? This question should be valid only for monsters with CR<1 since a greater CR would undermine some of the class features

Exemple: an orc monster with 8 barbarian levels,
A noble with 5 sorcerer levels.


Bonus: what CR should have your average PC according to the rules to assign the CR?

The only way to know for sure is to add the PC levels then recalculate CR.

There are a few online tools that would make that easier.

[Rukelnikov]

As many have said before, if you wanna get the approximate CR of a creature you need to run it thru the guidelines in the DMG and see what comes up.

Consider though, that for some characters this just can't be done in a static manner.

Take a lvl 5 wizard for instance, if he memorized Fireball he would have a far higher ACR than if he didn't and memorized only utility 3rd lvl spells (Haste and Slow for instnace), add on to spell selection, that if said wizard with Fireball spent all of his 3rd level slots, then his CR would drop considerably, since ACR is the expected ammount of damage dealt in 3 three turns, and so PCs, casters in particular, have a variable CR.

However take into account that method is not always accurate, and requires some adjudication in part of the DM.

[BloodSnake'sCha]

As many have said before, if you wanna get the approximate CR of a creature you need to run it thru the guidelines in the DMG and see what comes up.

Consider though, that for some characters this just can't be done in a static manner.

Take a lvl 5 wizard for instance, if he memorized Fireball he would have a far higher ACR than if he didn't and memorized only utility 3rd lvl spells (Haste and Slow for instnace), add on to spell selection, that if said wizard with Fireball spent all of his 3rd level slots, then his CR would drop considerably, since ACR is the expected ammount of damage dealt in 3 three turns, and so PCs, casters in particular, have a variable CR.

However take into account that method is not always accurate, and requires some adjudication in part of the DM.

I just want to add that the wizard ACR will also change based on his allies, a barbarian friend can make the haste do a lot more damage.

[Rukelnikov]

I just want to add that the wizard ACR will also change based on his allies, a barbarian friend can make the haste do a lot more damage.

But the CR calculating system doesn't take that into account

[BloodSnake'sCha]

But the CR calculating system doesn't take that into account

And that is how a lower CR can be a bigger problem then a high one.

And I remember reading that tactics and environment should be accounted when calculating CR somewhere. Am I accidentally using an older edition rule(AFB)?

[Damon_Tor]

Just as the title asks. What should be the CR of a monster with full class levels? This question should be valid only for monsters with CR<1 since a greater CR would undermine some of the class features

Exemple: an orc monster with 8 barbarian levels,
A noble with 5 sorcerer levels.


Bonus: what CR should have your average PC according to the rules to assign the CR?

The CR of an NPC is the level at which it would, by itself, be a reasonable challenge for a party of 4. So a CR 2 NPC is a reasonable challenge for a party of 4 level 2 PCs. So the answer is "CR*4=PC level" Roughly.

[HappyDaze]

As many have said before, if you wanna get the approximate CR of a creature you need to run it thru the guidelines in the DMG and see what comes up.

Consider though, that for some characters this just can't be done in a static manner.

Take a lvl 5 wizard for instance, if he memorized Fireball he would have a far higher ACR than if he didn't and memorized only utility 3rd lvl spells (Haste and Slow for instnace), add on to spell selection, that if said wizard with Fireball spent all of his 3rd level slots, then his CR would drop considerably, since ACR is the expected ammount of damage dealt in 3 three turns, and so PCs, casters in particular, have a variable CR.

However take into account that method is not always accurate, and requires some adjudication in part of the DM.

The Monster Manual mentions that changing a spellcaster's prepared/known spell list does not alter it's Challenge. In play, it makes a big difference, but supposedly a Wizard 5 is a Wizard 5 regardless of spell choices made.

[Rukelnikov]

The Monster Manual mentions that changing a spellcaster's prepared/known spell list does not alter it's Challenge. In play, it makes a big difference, but supposedly a Wizard 5 is a Wizard 5 regardless of spell choices made.

The MM may say that, but the rules in the DMG on how to calc a monsters CR don't agree with that, since to calc the Offensive CR you need to make an average of the damage the creature can do in 3 turns.

A lvl 5 Wizard that has Fireball and Shatter prepared, could do Fireball, Fireball, Shatter, the guidelines for area attacks is that they affects 2 creatures which fail their ST, so this wizard would be doing 8d6(28) * 2 = 56 for each Fireball, plus 3d8(13.5) * 2 =27 for the Shatter, totalling 139 across 3 rounds, for an avg of 46 damage a round. This would put him at OCR 7, now, if the same wizard had only CC spells, and lets say Firebolt, he would be doing 11 damage a round, which would be OCR 1.

So, if the MM says that it is clearly in conflict with Create a Monster rules, and tbh, makes no sense at all.

[Grognerd]

Glancing at all of these calculations reminds me of why I wanted to find a shortcut to CR.

Generally, you don't need to know the CR of a PC, but what if you are using a high-level PC from one campaign as an NPC patron or antagonist for a different campaign? In that case, having a basic CR could conceivably be useful.

What I have anecdotally found, was that a quick and dirty conversion can be found by subtracting proficiency from level, with negatives equaling increments. So a Level 1 PC would be CR 1/2 (1-2=1 increment below CR 1). A level 12 PC would be CR 8 (12-4=8), and a level 20 PC would be CR 14 (20-6). It's not perfect, but it's quick and easy.

[MaxWilson]

The MM may say that, but the rules in the DMG on how to calc a monsters CR don't agree with that, since to calc the Offensive CR you need to make an average of the damage the creature can do in 3 turns.

A lvl 5 Wizard that has Fireball and Shatter prepared, could do Fireball, Fireball, Shatter, the guidelines for area attacks is that they affects 2 creatures which fail their ST, so this wizard would be doing 8d6(28) * 2 = 56 for each Fireball, plus 3d8(13.5) * 2 =27 for the Shatter, totalling 139 across 3 rounds, for an avg of 46 damage a round. This would put him at OCR 7, now, if the same wizard had only CC spells, and lets say Firebolt, he would be doing 11 damage a round, which would be OCR 1.

So, if the MM says that it is clearly in conflict with Create a Monster rules, and tbh, makes no sense at all.

FWIW, I understand that WotC's internal spreadsheet for calculating monster CR is a little bit more complex than the 5E rules, and that when it comes to special abilities that don't have an entry in the DMG table, the methodology is to account for the ability as if it were a spell of appropriate level, and then convert that spell into offensive DPR by looking at the expected-spell-damage-by-level chart, and then plug that into the offensive CPR. Ref: http://blogofholding.com/?p=7357

Emphasis mine:


Mike asked Jeremy about the table in the DMG for homebrewing monsters. It’s public record that I have my doubts about that table. Jeremy provided this interesting fact: apparently, the canonical formula for determining monster CR is encoded in an internally-used Excel spreadsheet. (We’ve seen this spreadsheet in action in the Mike Mearls Happy Fun Hour.) The table in the DMG was made after the spreadsheet, and was an attempt to reverse-engineer and simplify the spreadsheet’s formulas for DM’s home use: it’s not used as part of the process for creating for-publication monsters.

My research indicates that while D&D monster stats are internally consistent and carefully designed, their power levels don’t line up well with the DMG chart. This leads me to believe that the spreadsheet does what it intends to do, while the DMG chart does not.

I invite you to ponder this further mystery from the Deep Dive podcast:

In the podcast, Jeremy Crawford talked about how the D&D team approaches what he calls “action denial” attacks: paralysis, charm, etc. This is something I’ve wondered about. My initial research suggested that these attacks didn’t have much of an effect on CR, which I though was strange: taking out a combatant for a few turns seems like it should be a powerful ability.

Jeremy used paralysis as an example to illustrate the team’s approach. First, find the lowest-level spell which inflicts a condition. For paralysis, that’s Hold Person. Next, you translate that into damage by finding the damage output of the simplest pure-damage spell of that level. Hold Person is level 2, and its comparable damage spell is Scorching Ray, which does 6d6 (21) damage. Thus, the ability to paralyze is worth 21 “virtual” damage for Challenge Rating calculation purposes.

Now, this is obviously a bad method for computing CR, but it's bad in ways which explain the MM guidance: if you're basically just considering all spells to generically do the same damage that's listed on the DMG expected spell damage by level chart (which is itself not very predictive of actual spells), then it does make logical sense to let monsters swap spells out freely without changing the CR, since the details of what spells actually do isn't factored into CR anyway.

It would be easier to get upset about this if this were the only way in which CR was flawed, but by this point we should all be used to mostly ignoring CR anyway when it comes to actual adventure writing (as opposed to white room cage matches).

[GlenSmash!]

Glancing at all of these calculations reminds me of why I wanted to find a shortcut to CR.

Generally, you don't need to know the CR of a PC, but what if you are using a high-level PC from one campaign as an NPC patron or antagonist for a different campaign? In that case, having a basic CR could conceivably be useful.

What I have anecdotally found, was that a quick and dirty conversion can be found by subtracting proficiency from level, with negatives equaling increments. So a Level 1 PC would be CR 1/2 (1-2=1 increment below CR 1). A level 12 PC would be CR 8 (12-4=8), and a level 20 PC would be CR 14 (20-6). It's not perfect, but it's quick and easy.
I just grab an existing NPC block that would be close to the PC.

An Archmage from VGtM won't be exactly the same as Tim the Evocation wizard from the last campaign, but it will probably be close enough. And I don't have to do any calculations or conversions

[Rukelnikov]

FWIW, I understand that WotC's internal spreadsheet for calculating monster CR is a little bit more complex than the 5E rules, and that when it comes to special abilities that don't have an entry in the DMG table, the methodology is to account for the ability as if it were a spell of appropriate level, and then convert that spell into offensive DPR by looking at the expected-spell-damage-by-level chart, and then plug that into the offensive CPR. Ref: http://blogofholding.com/?p=7357

Emphasis mine:



Now, this is obviously a bad method for computing CR, but it's bad in ways which explain the MM guidance: if you're basically just considering all spells to generically do the same damage that's listed on the DMG expected spell damage by level chart (which is itself not very predictive of actual spells), then it does make logical sense to let monsters swap spells out freely without changing the CR, since the details of what spells actually do isn't factored into CR anyway.

That's pretty cool, I didn't know that, thx.


It would be easier to get upset about this if this were the only way in which CR was flawed, but by this point we should all be used to mostly ignoring CR anyway when it comes to actual adventure writing (as opposed to white room cage matches).

Yeah, I don't think I've followed 5e CR guidelines ever, or the encounter building ones for that matter.

[Nagog]

All things considered, there is no direct correlation. For example, a Noble with 5 Barbarian levels may be an imposing enemy, while a Noble with 5 Bard levels would be much less so. CR is calculated based on combat difficulty, whereas many classes have very well rounded stats for both combat and interaction. In a similar light, I have a Kobold enemy in the campaign I DM that has the abilities of an Illusionist Wizard. While direct combat with him alone may be lackluster for a party of 5 (no matter the level), he's quickly becoming the most hated enemy in the campaign to date due to how he's been able to mess with the party out of combat.

However, if you're intent on giving enemies class levels for combat, I'd recommend running a few test encounters, starting with an encounter that would be classified as Easy or Medium difficulty, then adding class levels to the enemies. With any luck (and a few tweaks here and there), the class levels will scale it up by a difficulty level or two, so it would still be difficult but not too challenging as to not be fun.

[SpawnOfMorbo]

Just as the title asks. What should be the CR of a monster with full class levels? This question should be valid only for monsters with CR<1 since a greater CR would undermine some of the class features

Exemple: an orc monster with 8 barbarian levels,
A noble with 5 sorcerer levels.


Bonus: what CR should have your average PC according to the rules to assign the CR?

Nothing.

CR is meaningless and doesn't actually have a value because CR is a broken system that relies on shrugs, maybes, and rules they forgot to update half way through the design process.

[Yunru]

Nothing.

CR is meaningless and doesn't actually have a value because CR is a broken system that relies on shrugs, maybes, and rules they forgot to update half way through the design process.

It's still a vague, rough guideline. Most creatures aren't more than 1 CR off.

[MaxWilson]

It's still a vague, rough guideline. Most creatures aren't more than 1 CR off.

Well, yes, but that's because most monsters are boring sacks of HP, which is exactly what CR is designed to model. The monsters that actually use their special abilities like high movement speed, constriction, paralysis, etc. tend to be deadlier than their CR makes them look, especially in combination.

I was rereading the DMG tables last night and it's rather shocking how many powerful abilities have negligible or zero impact on CR, when in reality they have a huge impact on tactical effectiveness. At the same time, minor variations in DPR have large impacts on CR. E.g. I recomputed Geryon's CR, and depending on whether or not you give him his extra constriction damage in all 3 rounds or only #2 and #3, and whether you round up or down, his CR can drop from CR 22 to CR 20--IIRC, dropping from 107 DPR to 91 DPR costs him something like -3 to offensive CR! In a real game that's just one unlucky miss on round one, and of course the real threat comes from synergies between e.g. Geryon's legendary Teleportation action, his large single-target DPR, and his regeneration: it's easy to imagine him teleporting in and disassembling + grappling the party healer, then flying off to heal and wait for party buffs to expire. (Bonus points if he casts Symbol of Pain beforehand so that any PCs chasing him walk right into it.)

All of the abilities and tactics that make a monster interesting don't really show up in CR. CR really is just a crude measure of essentially log(DPR * HP), and therefore metrics which rely on CR (like encounter difficulty) are not reliable.

[Rukelnikov]

Well, yes, but that's because most monsters are boring sacks of HP, which is exactly what CR is designed to model. The monsters that actually use their special abilities like high movement speed, constriction, paralysis, etc. tend to be deadlier than their CR makes them look, especially in combination.

I was rereading the DMG tables last night and it's rather shocking how many powerful abilities have negligible or zero impact on CR, when in reality they have a huge impact on tactical effectiveness. At the same time, minor variations in DPR have large impacts on CR. E.g. I recomputed Geryon's CR, and depending on whether or not you give him his extra constriction damage in all 3 rounds or only #2 and #3, and whether you round up or down, his CR can drop from CR 22 to CR 20--IIRC, dropping from 107 DPR to 91 DPR costs him something like -3 to offensive CR! In a real game that's just one unlucky miss on round one, and of course the real threat comes from synergies between e.g. Geryon's legendary Teleportation action, his large single-target DPR, and his regeneration: it's easy to imagine him teleporting in and disassembling + grappling the party healer, then flying off to heal and wait for party buffs to expire. (Bonus points if he casts Symbol of Pain beforehand so that any PCs chasing him walk right into it.)

All of the abilities and tactics that make a monster interesting don't really show up in CR. CR really is just a crude measure of essentially log(DPR * HP), and therefore metrics which rely on CR (like encounter difficulty) are not reliable.

Exactly, that's why I've never really used encounter builder system, and also why I strayed away from standard progression, since XP is also a factor of CR, and thus leveling system is busted too.

[Gignere]

That's pretty cool, I didn't know that, thx.



Yeah, I don't think I've followed 5e CR guidelines ever, or the encounter building ones for that matter.

I think you do, even if it is just minimally unless you roll totally randomly out of the MM. I’m sure you’ve never sent the tarrasque against a level 1 party.

[MaxWilson]

I think you do, even if it is just minimally unless you roll totally randomly out of the MM. I’m sure you’ve never sent the tarrasque against a level 1 party.

Outside of hack-and-slash one-shots, I never even compute encounter difficulty or total adventuring day "adjusted XP" until after the adventure is over. This is how I learned that I tend to prefer encounters somewhere between Deadly and Deadly x10, with an emphasis on Deadly x3-4. (At Deadly x5-6 the odds are usually tilted slightly against the PCs, maybe a 60% chance of losing unless they do something either creative or powergamey.) It would not be unusual for me to spend what turns out to have been 130% of the adventuring day XP budget in one encounter, then another 30% in another encounter, then 50% in a third. I.e. I don't usually violate total adventuring day XP guidelines as thoroughly as the individual encounter guidelines.

Does ignoring CR/difficulty until afterwards count as using the CR system or not?

[Gignere]

Outside of hack-and-slash one-shots, I never even compute encounter difficulty or total adventuring day "adjusted XP" until after the adventure is over. This is how I learned that I tend to prefer encounters somewhere between Deadly and Deadly x10, with an emphasis on Deadly x3-4. (At Deadly x5-6 the odds are usually tilted slightly against the PCs, maybe a 60% chance of losing unless they do something either creative or powergamey.) It would not be unusual for me to spend what turns out to have been 130% of the adventuring day XP budget in one encounter, then another 30% in another encounter, then 50% in a third. I.e. I don't usually violate total adventuring day XP guidelines as thoroughly as the individual encounter guidelines.

Does ignoring CR/difficulty until afterwards count as using the CR system or not?

In my experience it depends on the group. With a great group that knows what they are doing it seems like throwing a 3x - 6x deadly encounters is the right fit. But in a group of noobs even CR appropriate encounters may seem too much.

Or maybe I’m just a vicious DM when I send anything more than 3x deadly it’s time to fudge dice rolls unless I plan to kill or TPK.

I would never send something 10x deadly that’s no way it’s not a wipe unless the DM isn’t using the monsters or running the encounters appropriately. So I loosely base things off of CR like I would never send an Ancient Red Dragon against a level 1 party.

[MaxWilson]

In my experience it depends on the group. With a great group that knows what they are doing it seems like throwing a 3x - 6x deadly encounters is the right fit. But in a group of noobs even CR appropriate encounters may seem too much.

Or maybe I’m just a vicious DM when I send anything more than 3x deadly it’s time to fudge dice rolls unless I plan to kill or TPK.

I would never send something 10x deadly that’s no way it’s not a wipe unless the DM isn’t using the monsters or running the encounters appropriately. So I loosely base things off of CR like I would never send an Ancient Red Dragon against a level 1 party.

Well, one of my first experiences running 5E was setting a bunch of new players on a collision course with what turned out (when I did the math some time afterwards) to be a 10x Deadly threat, which they managed to beat with some luck and creativity and approximately zero powergaming. They did pretty okay in subsequent triple-Deadly-ish encounters as well, etc. They did TPK once that I can remember, but they also survived at least three situations where I outright told the players I was pretty sure they were all about to die... My takeaway from that is that 5E PCs are deceptively robust. When the DM thinks they're just about out of gas, they probably aren't really.

For example, if the party loses 90% of their HP in one fight... They've really only lost 45% of their total HP reserves. A short rest, spending almost all their Hit Dice, will have everybody back at full HP, even if there is no healer in the party.

I saw a first level Barbarian come within a few swings of taking down a CR 2 ogre, solo, in an arena fight, using completely straightforward tactics: trading attacks until somebody fell down. IIRC that was a Deadly x6 fight and he almost won anyway. The odds were against him but not THAT far against him.

My issue with non-Deadly fights is that knowing in advance that the PCs will definitely win makes the whole scene anticlimactic, the sort of thing I'd prefer to skip ("You kill them all, losing 3d6 HP each or two spell slots of your choice. What do you do next?") unless players want to play it out in detail.

[Rukelnikov]

I think you do, even if it is just minimally unless you roll totally randomly out of the MM. I’m sure you’ve never sent the tarrasque against a level 1 party.

Fun example since I sent a lvl 7 party against him lol

[Grognerd]

I think you do, even if it is just minimally unless you roll totally randomly out of the MM. I’m sure you’ve never sent the tarrasque against a level 1 party.

Not at all. Using common sense in assigning monster's against PC level is a practice that WELL precedes the development of CR and the CR encounter system. That CR reflects this does not mean that this reflects CR. Don't mistake old practice with new rules.

[MaxWilson]

Fun example since I sent a lvl 7 party against him lol

If a DM did this to me, I'd research everything I could about the Tarrasque's history and implied stats. Then I'd laugh and kill it with Call Lightning or something.

Tarrasque without unstoppable regen or long-range fear = weak.


Not at all. Using common sense in assigning monster's against PC level is a practice that WELL precedes the development of CR and the CR encounter system. That CR reflects this does not mean that this reflects CR. Don't mistake old practice with new rules.

+1, quite true.

Also, letting players select their own monster difficulty by putting more treasure and tougher monsters on deeper dungeon levels (or the equivalent).

[Rukelnikov]

If a DM did this to me, I'd research everything I could about the Tarrasque's history and implied stats. Then I'd laugh and kill it with Call Lightning or something.

Tarrasque without unstoppable regen or long-range fear = weak.

The legendary action movement is what got the best of the party basically, once a character dealt good damage ot it, the tarrasque would spend all his following legendaries (and in one casse also his round simply dashing) to close in on him, and kill him afterwards. I did buff the party considerably though, and while I certainly assumed they were gonna die, they did much better than expected.

They were basically fighting in an "endless war" between two factions that were kinda like ghosts in that they always come back even if destroyed, and the field they were fighting in had that effect, they didn't know they were gonna be fighting the tarrasque, though one player had his suspicions, and they chose to fight the supposedly very powerful creature over the promises of wealth that the supreme leader of the faction they were aiding promised them.

[Gignere]

Well, one of my first experiences running 5E was setting a bunch of new players on a collision course with what turned out (when I did the math some time afterwards) to be a 10x Deadly threat, which they managed to beat with some luck and creativity and approximately zero powergaming. They did pretty okay in subsequent triple-Deadly-ish encounters as well, etc. They did TPK once that I can remember, but they also survived at least three situations where I outright told the players I was pretty sure they were all about to die... My takeaway from that is that 5E PCs are deceptively robust. When the DM thinks they're just about out of gas, they probably aren't really.

For example, if the party loses 90% of their HP in one fight... They've really only lost 45% of their total HP reserves. A short rest, spending almost all their Hit Dice, will have everybody back at full HP, even if there is no healer in the party.

I saw a first level Barbarian come within a few swings of taking down a CR 2 ogre, solo, in an arena fight, using completely straightforward tactics: trading attacks until somebody fell down. IIRC that was a Deadly x6 fight and he almost won anyway. The odds were against him but not THAT far against him.

My issue with non-Deadly fights is that knowing in advance that the PCs will definitely win makes the whole scene anticlimactic, the sort of thing I'd prefer to skip ("You kill them all, losing 3d6 HP each or two spell slots of your choice. What do you do next?") unless players want to play it out in detail.

I don’t know it sounds like you are sending one big CR against the party and calculating deadly or not. But that’s not my style I like minion style fights. But that barbarian example is ridiculous it’s just luck of the die. The ogre slightly above average damage can one shot a barbarian if he goes first before the barbarian can rage. Even with rage that’s only two maybe three hits to kill the barbarian. The barbarian will need 6 hits to kill the ogre.

So what probably happened was the barbarian got lucky and instead of dropping in the expected 2.5 rounds made it to 4 rounds. Managed to hit all four times so it appears like he was only two shot from winning the ogre. But that’s just luck of the dice, run this fight 10 times and you’ll probably see a few where the ogre just one shots the barbarian. Will you than say the barbarian is weak?

[AdAstra]

I saw a first level Barbarian come within a few swings of taking down a CR 2 ogre, solo, in an arena fight, using completely straightforward tactics: trading attacks until somebody fell down. IIRC that was a Deadly x6 fight and he almost won anyway. The odds were against him but not THAT far against him.


Take no issue with the rest of your post, but I would point out that there are factors here that are heavily in favor of the Barbarian, thus significantly increasing his effective threat.

For one, the Barbarian is by definition a hard-to-kill melee beatstick. It makes sense that he'll punch above his weight (aka better than average) as a hard-to-kill beatstick, since the assumption is that an average adventuring party is not wholly composed of hard-to-kill beatsticks. Their abilities, above all else, revolve around getting into melee fights and winning through superior endurance, so him succeeding to this degree is not entirely surprising.

Ogres are not exactly the most powerful things in their weight class. A griffon does nearly 50% more damage while being comparable in most other respects (same health, +1 AC) for the purposes of a Barbarian.

He could also have been more lucky than average. Especially at level 1, luck matters a lot more than at other levels, though less so with Barbarian. Still potentially a major factor. You would probably have a good idea of roughly how lucky he was during the fight, though. Bad luck would have easily screwed him, though. If the Ogre had won initiative and hit with its first attack, it could have conceivably one-shot him.

[MaxWilson]

Take no issue with the rest of your post, but I would point out that there are factors here that are heavily in favor of the Barbarian, thus significantly increasing his effective threat.

For one, the Barbarian is by definition a hard-to-kill melee beatstick. It makes sense that he'll punch above his weight (aka better than average) as a hard-to-kill beatstick, since the assumption is that an average adventuring party is not wholly composed of hard-to-kill beatsticks. Their abilities, above all else, revolve around getting into melee fights and winning through superior endurance, so him succeeding to this degree is not entirely surprising.

Granted! Probably no other first level PC could have done as well against an ogre with those tactics, and I'd guesstimate that the Barbarian still had no better than a 1 in 3 chance of winning. But some people trust anecdotes more than analysis, so it seems relevant to point out that this is one example of a Deadly x6 match which is not impossible even if the player is just doing the obvious thing, and that if it had been a Hard or regular Deadly match the Barbarian would have been heavily favored to win, which is actually less exciting in play. 1 first level PC vs. 1 Orc is a Deadly x1.5 match, but if I'm going to offer 100 gp and +10 local Reputation to the winner of a gladiatorial combat you can bet the other guy isn't going to be a bog-standard 15 HP orc! It needs to be someone who's actually plausible as an intimidating Champion and who makes the player feel awesome if they manage to defeat them. An Ogre fit that bill, that time.


Ogres are not exactly the most powerful things in their weight class. A griffon does nearly 50% more damage while being comparable in most other respects (same health, +1 AC) for the purposes of a Barbarian.

Absolutely. A Peryton would have wrecked the Barbarian, or an Intellect Devourer, or a Griffin. There are plenty of CR 2ish monsters that would have done better than the Barbarian, and plenty of first-level PCs that would have done worse (and some that might have done better).


He could also have been more lucky than average. Especially at level 1, luck matters a lot more than at other levels, though less so with Barbarian. Still potentially a major factor. You would probably have a good idea of roughly how lucky he was during the fight, though. Bad luck would have easily screwed him, though. If the Ogre had won initiative and hit with its first attack, it could have conceivably one-shot him.

Absolutely. He was more lucky than average, and he still lost. The Ogre was a stronger champion than the Barb was, and it deserved to win. But it almost didn't win that time, despite being Deadly x6, and as you say that's unlucky but not entirely surprising.

My experience is that especially after 5th level or so, a "fair fight" (hard to predict which side will win, depending on tactics) usually, when I do the math afterwards, comes out to about Deadly x3 or Deadly x4. A fight which is definitely slanted against the PCs (i.e. they will probably lose, ~60%/40%, unless they use good tactics to improve their odds) seems to come out around Deadly x5 or Deadly x6. The Ogre/Barb fight is an example of the latter kind of fight, it just happened before 5th level.

Obviously it is possible for the DM to deliberately game the CR system to create incredibly deadly "Easy"-rated encounters, but I'm not talking about that. I'm just talking about setting up encounters without regard to CR, and then measuring the CR afterwards.

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


I don’t know it sounds like you are sending one big CR against the party and calculating deadly or not. But that’s not my style I like minion style fights. But that barbarian example is ridiculous it’s just luck of the die. The ogre slightly above average damage can one shot a barbarian if he goes first before the barbarian can rage. Even with rage that’s only two maybe three hits to kill the barbarian. The barbarian will need 6 hits to kill the ogre.

So what probably happened was the barbarian got lucky and instead of dropping in the expected 2.5 rounds made it to 4 rounds. Managed to hit all four times so it appears like he was only two shot from winning the ogre. But that’s just luck of the dice, run this fight 10 times and you’ll probably see a few where the ogre just one shots the barbarian. Will you than say the barbarian is weak?

I will definitely say that the Ogre is stronger than the Barbarian, which is the whole point. (I want to ensure that e.g. coming back at 3rd level and winning the arena fight feels like a Big Deal.) He was the local heavyweight champion. (Edit: I think I misremembered. He was actually the lowest-ranked of five or six champions, IIRC.) I also want to make sure that the player's choices matter, including build choices--some hypothetical PCs could have better or worse chances against the ogre than the barbarian who showed up in play. I don't know what you mean by "weak" here but yes, the odds were certainly against the barb, which as I've said repeatedly is what I've learned to expect from Deadly x6 fights: bad guys are favored to win unless PCs do something creative or powergamey. But they're not favored by all that much.

[mAc Chaos]

The closest I got was that a PC's CR is their level divided by 4.

So a level 8 PC is equal to a CR 2.

[MaxWilson]

The closest I got was that a PC's CR is their level divided by 4.

So a level 8 PC is equal to a CR 2.

That can't be accurate. If I take a typical Medium fight (say, four CR 2 Iron Shadows against four 8th level PCs), the PCs are definitely going to win that fight. I almost don't care what the PCs are. But if I swap those Iron Shadows out for four 8th level PCs, the difficulty goes way up, by definition: that's a 50/50 fight. My experience tells me 50/50 fights happen at approximately Deadly x3-4, which for an 8th level party of four means roughly 25,000 to 33,000 XP. Four CR 7 Giant Apes would be 23,200 XP; four CR 8 T-Rexes would be 31,200 XP, so it looks like CR = Level is approximately in the ballpark.

And yes, I would expect four 8th level PCs to have about the same chance of beating four T-Rexes as they would of beating four 8th level (N)PCs: roughly 50%.

[AdAstra]

I know one matchup I’ve actually thought about, and one that I think seems pretty reasonable, is a CR 3 Knight vs a level 5 melee fighter in splint. It’s not perfect, given that it has a fairly powerful support ability that’s useless when fighting solo, but that’s somewhat counterbalanced by the fact that the Knight is especially hardy to attack rolls due to high AC.

If I’m not mistaken, a stat-optimized fighter will usually come out on top, but the margin is fairly small if the fighter does not expend any resources and isn’t using one of the more powerful builds for such a fight, like PAM with spear and Dueling (GWM isn’t a good option against such a high AC). If the Fighter takes advantage of Action Surge, Second Wind, and whatever subclass features they have it should be no contest, but the Knight should still be able to get in some hits, except perhaps against an Eldritch Knight using Shield for the first three rounds.

So in this case, a CR 3 creature ranges from a close but winnable fight for a robotically played PC, to a serious but not deadly threat to a average player and character using a decent amount of resources, to a pushover who’ll have trouble even inflicting damage on builds/strategies that are specifically effective on it that are going all out. But importantly, the fighter should generally lose a meaningful chunk of HP or other resources.

[stoutstien]

I think one major flaw of CR in general is that it's only a challenge factor based on how deadly it is in terms of dealing damage to the party.
PCs have a huge list of ways to interact with problems other than hitting them where most NPCs are just piles of HP that hit back.

What would the CR of 10 boggles trying to stealing the relic from a keep with 3 floors and over 20 rooms? Now factor in that this kingdom considers all fey as sacred beings so you cannot kill any of them.

[JNAProductions]

I think one major flaw of CR in general is that it's only a challenge factor based on how deadly it is in terms of dealing damage to the party.
PCs have a huge list of ways to interact with problems other than hitting them where most NPCs are just piles of HP that hit back.

What would the CR of 10 boggles trying to stealing the relic from a keep with 3 floors and over 20 rooms? Now factor in that this kingdom considers all fey as sacred beings so you cannot kill any of them.

I don't consider that a flaw, I'd consider that a limitation.

CR is not meant to model every possible variety of challenge. It's meant to measure Combat challenge, and for that, it does so decently.

[stoutstien]

I don't consider that a flaw, I'd consider that a limitation.

CR is not meant to model every possible variety of challenge. It's meant to measure Combat challenge, and for that, it does so decently.

That's the problem. Combat may have special rules for action economy but combat it self isn't special in the terms of challenge. If CR was short for combat rating I'd agree but it being labeled as challenge rating is a half truth at best. If anything it should be a deadly rating.

The example above where the barbarian is trading hits with an ogre would be considered a pretty high challenge but is it? Deadly yes but challenging? What is the challenge in this scenario? Should not dying due to RNG be a basis of challenge?

[MaxWilson]

The example above where the barbarian is trading hits with an ogre would be considered a pretty high challenge but is it? Deadly yes but challenging? What is the challenge in this scenario? Should not dying due to RNG be a basis of challenge?

Nope! It was all of five minutes of play, a fun optional minigame, not a serious challenge. If the player had made it a goal to become regional champion (defeating all the other gladiators I had already made, of various Reputations and levels and styles), that would have required serious table time and probably lots of talking to NPCs and searching for magic artifacts. It would have been a real challenge. But it turned out to just be a brief, fun, unsuccessful diversion, and the ten minutes I had spent sketching out gladiators didn't get used. C'est la vie la DM. :)

(Knowing what I know now, I could have offered a better hook after his defeat by having the victorious Ogre come mock him as a weak little human afterward. "You don't have what it takes, weakling! [sneer] Why don't you make like a tree, and get out of town." Then someone quietly tips them off to a coach who has a grudge against the ogre and might give the barbarian tips on how to beat him, and then later on they see a poster advertising a different fight between two other gladiators. Players respond to cues about where they can find interesting stuff to do, and back then I knew less about setting up those cues.)



So in this case, a CR 3 creature ranges from a close but winnable fight for a robotically played PC, to a serious but not deadly threat to a average player and character using a decent amount of resources, to a pushover who’ll have trouble even inflicting damage on builds/strategies that are specifically effective on it that are going all out. But importantly, the fighter should generally lose a meaningful chunk of HP or other resources.

Therefore a fifth level Fighter should probably be rated somewhere around CR 4-5, since the first Fighter can beat the CR 3 Knight but against another Fighter 5 it's a toss up, by definition. Agreed?

[Nidgit]

CR as a scale might be a bit off, but it's fairly internally consistent. The MM also provides several examples of NPCs resembling PC classes to use as a point of reference.

For example, the Mage is a 9th-level spellcaster with 9 hit dice and is CR6. The Archmage is an 18th-level spellcaster, has 18 hit dice, and is CR12. Both CRs are 2/3 the level they are.

Druid is a 4th-level caster and Priest is a 5th-level caster, but both lack additional abilities or offensive spells so they're CR2. The Veteran is CR3 with two attacks per turn (suggesting at least Level 5) but lacks additional abilities. The Knight also fits this mold, but both the Knight and Veteran have additional hit dice because all medium creatures are locked at a d8 hit die, instead of a d10 or d12 for a Fighter or Barbarian.

Using this pattern, I'd suggest using 2/3 PC Level = CR as a baseline and then moving it up or down a little bit depending on how strong the character's abilities/spells are.

[Xetheral]

I would disagree with this.

PC vs evil N(PC) fights tend to be different, since the players face a relative glass cannon with lots of very powerful features and potentially explosive burst damage.

I’ve run a couple of campaigns with predominantly PC based major and moderate foes and they can work just fine... but they tend to run differently. The PCs quickly learn to try and get the drop on their foes wherever possible, and are far more likely to scout, study and fear their opponents.

The main argument I see against PC foes isn’t experience for the Players, It’s DM exhaustion. I find it more demanding to run foes with so many options.

It’s also pretty much mandatory to war game the encounters a bit. The interplay of so many options tends to make it hard to play more than one or two at a time without some forethought, and figuring a CR almost has to be done in the context of how everything the enemy has lines up with everything the PCs have

Running CR monsters is definitely easier. Don’t let that stop you from trying though.

If you want to give it a shot, find an in game way to give the players some warning, at least for the first pc they face

Run a few what if simulations to get a read on difficulty

If you do this a lot you can expect the players to evolve strong gank-squad tactics.

Following up on this post, it's also easier to telegraph difficulty when using PC-built antagonists than it is when using anything-goes monsters. If the PCs observe a PC-built NPC Wizard cast Fabricate, that gives them a ton of information about the Wizard's potential capabilities were it to come to a conflict later. By contrast, if the NPC wizard was just a monster statblock, observing it cast Fabricate provides almost no information about what else the NPC can do.

This is particularly useful for player-driven sandbox campaigns where accurately telegraphing potential encounter difficulty provides critical information that the PCs need when evaluating how much risk they want to take.

[MaxWilson]

If a bad guy is a 11th level wizard or 9th level fighter, it's still an NPC, not a PC. It's right in the name: non-player character. A character which is not attached to a player is an NPC, period.

Yes, NPCs can have classes and levels. This is why NPC-only classes like the Death Cleric and the Oathbreaker are called "NPC classes."

If you want a word to distinguish between NPCs with levels and NPCs that are built like monsters, you could say "NPC with class levels" or "PHB-style NPCs" vs. "MM-style NPCs", but don't call NPCs with class levels "PCs."



Following up on this post, it's also easier to telegraph difficulty when using PC-built antagonists than it is when using anything-goes monsters. If the PCs observe a PC-built NPC Wizard cast Fabricate, that gives them a ton of information about the Wizard's potential capabilities were it to come to a conflict later. By contrast, if the NPC wizard was just a monster statblock, observing it cast Fabricate provides almost no information about what else the NPC can do.

This is particularly useful for player-driven sandbox campaigns where accurately telegraphing potential encounter difficulty provides critical information that the PCs need when evaluating how much risk they want to take.

Xetheral makes a great point. I think my bias towards sandbox probably explains why I almost never use MM-style NPCs except for 0th level noncombatants.

Well, that plus the fact that using NPCs with classes is just simpler--I know exactly what a 11th level Diviner or 8th level Warrior (degenerate Fighter with no subclass) can do, and in principle so do the players. I don't have to look up a separate statblock in the MM to know what a "Knight" is, I just make someone a 4th level Cavalier or Samurai.

[Yakk]

I'm going to test my CR Formula (http://www.giantitp.com/forums/showthread.php?601554-CR-formula-math).

NPC Samurai, level 5:

18 Str, 14 Con, Plate, Shield, Longsword, Dueling style

Fighting Spirit grants 15 HP and advantage (worth about +3 ATK) for 3 rounds.

ATK: +7 "+3 from advantage"
AC: 20

21 DPR, but action surge grants another +33%, so 28 DPR.
44 HP + 15 temporary for 59 HP.

CR = AC/4 + ATK/4 + HP/36 + DPR/15 - 6
CR = 20/4 + 10/4 + 59/36 + 28/15 - 6
CR =~ 5.00555555556

Level 11

20 Str, 16 Con, Plate, Shield, Longsword + 1, Dueling style

30 HP fighting spirit and +3 to hit from advantage.

+10 ATK 20 AC

103 HP

37.5 DPR + 33% from action surge is 50 DPR

CR = AC/4 + ATK/4 + HP/36 + DPR/15 - 6
CR = 20/4 + 13/4 + 133/36 + 50/15 - 6
CR = 9.27777777778


Level 5 Samurai: CR 5
Level 11 Samurai: CR 9

Level 20 Samurai.

20 Str, 20 Con, Plate+1, Shield+1, Sunblade, Dueling style
Defensive Duelist, Shield Master

45 HP fighting spirit and +3 to hit from advantage.

224 HP
22 AC, ~+3 from Duelist (assume it helps about half of the time you are hit)
+5 attacks/round from rapid strike (advantage is free).

67.5 DPR base, 112.5 DPR after action surge.
ATK is +7 + Proficiency. Assuming +6 proficiency. +3 for constant advantage.

CR = AC/4 + ATK/4 + HP/36 + DPR/15 - 6
CR = 25/4 + 16/4 + 269/36 + 112.5/15 - 6
CR =~ 19

Level 5 Samurai: CR 5
Level 11 Samurai: CR 9
Level 20 Samurai: CR 19

L20 sanity check:
269 HP so CR 14 defensive. Expected AC of 18; actually 23, for +2.5 or CR 16.5 defensive.
112 DPR so CR 18 offensive. Expected ATK of 10 but has +16 for +3 or 21 offensive CR.

(16.5+21)/2 is 18.75 or 19 CR.

[Nidgit]

Level 5 Samurai: CR 9
Level 11 Samurai: CR 14
Level 20 Samurai: CR 23
This is comical. You're telling me that a Level 5 Samurai by themselves is a Deadly challenge for that same Samurai and three other party members?

Why not follow the CR formula in the DMG? Using that, a Level 5 Samurai is CR 5 and a Level 11 is CR 8, which seems much more reasonable.

[Breccia]

I hate this question (I don't hate you for asking it) because there's no good answer.

Party vs solo, the party has a monstrous advantage that supercedes something like levels. The party just has too many actions and advantages. For the sake of examples, imagine a 5th level rogue against a 5th level party. The rogue has one round to live. Maybe he gets a lucky hit, but even with with an 18 Dex, +1 finesse weapon and a Sneak Attack critical, the damage will be sub-30 and that's not enough to kill the party's healer. Then he's hit by 3-4 melee and ranged attacks and some third level spells and becomes an ink smear.

There is nearly zero risk to the party, despite the levels being the same.

Oh, and it gets worse. Now let's use, say, a 10th level cleric against the same party. Guy probably leads with flame strike, right? That's about 36 damage, probably hitting multiple targets. Awesome. But at 5th level, that still might not be enough to drop the party's healer. 50/50 or so. Now he's in melee with the party, making spellcasting a chore, while the remaining standing range members can spread out to minimize spell damage. Even giving the cleric AC 20 and 70 hp, not bad at all for that level, at 5th level the party probably has 2-3 hits plus clerics suck at Dex saving throws. The cleric can heal himself, true, but that's probably a waste since the party will just attack again next round.

There is risk to the party, but it's based largely on the cleric getting a lucky one-shot on a member or two. The smart money's still on the party. And this guy is double their level!

Now, everything changes dramatically when the solo member gets something as minor as four orcs. In the first case, the orcs act as a distraction, giving the solo rogue a solid chance to drop two members before retaliation. In the cleric's case, the orcs buy enough time for another powerful spell. The PCs might live through the first flame strike, they are not living through the second.

That is why I hate this question. PC/NPCs using standard rules are just too fragile by themselves, even with a significant level advantage. Put them in a team, and they really shine. That's why the characters use them. Almost every encounter I've made where the villain is a solo character, the party seems to win easier than I planned on.

Simply put, CRs are supposed to be an objective measure of risk. PC/NPCs just don't fit that mold.

[Yakk]

Using DMG directly.

59 HP gives you defensive CR of 1/2. Expected AC is 13 -- you have 20. So that is +3.5 defensive CR steps, or a defensive CR of 4.5

28 DPR gives a base offensive CR of 4. Attack bonus should be +5; but it is actually +10. So that adds another 2.5 offensive CR, bumping it up to 6.5.

Average of 6.5 and 4.5 is 5.0.

Aha, figured out my error -- I used AC/2 and ATK/2 and forgot it takes 4 points of AC/ATK to give 1 CR. Hence the over-estimate here.

Sorry! Back to work on formula.

Ok fixed.


This is comical. You're telling me that a Level 5 Samurai by themselves is a Deadly challenge for that same Samurai and three other party members?

Why not follow the CR formula in the DMG? Using that, a Level 5 Samurai is CR 5 and a Level 11 is CR 8, which seems much more reasonable.

Yes, there was a bug in the formula. I used AC/2 instead of AC/4 and ATK/2 instead of ATK/4. I fixed it.

I'm getting CR 9 using my formula and 8 CR using strait DMG numbers for a Level 11 Samurai (with a +1 sword). 3 attacks/round and action surge are painful.

Crazily, I'm also getting CR 19 with both my formula and the DMG for a level 20 Samurai (with a Sunblade, +1 shield and +1 plate). 5 attacks/round hurts.

[MaxWilson]

I hate this question (I don't hate you for asking it) because there's no good answer.

Party vs solo, the party has a monstrous advantage that supercedes something like levels. The party just has too many actions and advantages. For the sake of examples, imagine a 5th level rogue against a 5th level party. The rogue has one round to live. Maybe he gets a lucky hit, but even with with an 18 Dex, +1 finesse weapon and a Sneak Attack critical, the damage will be sub-30 and that's not enough to kill the party's healer. Then he's hit by 3-4 melee and ranged attacks and some third level spells and becomes an ink smear.

There is nearly zero risk to the party, despite the levels being the same.

Oh, and it gets worse. Now let's use, say, a 10th level cleric against the same party. Guy probably leads with flame strike, right? That's about 36 damage, probably hitting multiple targets. Awesome. But at 5th level, that still might not be enough to drop the party's healer. 50/50 or so. Now he's in melee with the party, making spellcasting a chore, while the remaining standing range members can spread out to minimize spell damage. Even giving the cleric AC 20 and 70 hp, not bad at all for that level, at 5th level the party probably has 2-3 hits plus clerics suck at Dex saving throws. The cleric can heal himself, true, but that's probably a waste since the party will just attack again next round.

It's not about "action economy", it's about total offense vs. total defense, and some defenses (like Stealth) work better for individuals than for groups.

I remember 9th level Rogue/Monk (Rogue 2/Shadow Monk 7) who forced a whole party of ~5th level PCs into headlong flight by virtue of Pass Without Trace + Stealth Expertise + Cunning Action (Hide) + Deflect Missiles plus a shortbow. (There was a standoff period in the middle where the PCs were holding Readied Actions waiting for the NPC to reappear, but ultimately the PCs wound up running away.) But a whole group of four Rogue/Monks would not have been much more powerful than the lone Rogue/Monk, because if you can deal with that tactic from one you can deal with it from four.

Similarly, solo monsters who use hit-and-run tactics are terrifying effective, even if they only have a regular action + bonus action + reaction just like a PC. It's not the number of actions, it's how much you can do with those actions.
Anyway, that 10th level cleric should cast Hold Person V on round one instead of Flame Strike. He'll still probably die but it's better to degrade PC offense quickly than slowly.

[pming]

Hiya!!

I'm of the opinion that the CR stuff is utter hogwash...but just for fun...:)

Write up you "Monster with Levels", but do it without mentioning class, and rename all the class-based abilities (e.g., its a "Unfettered Acolyte of Bis" and it has AC 14 due to "expert training in defensive acrobatics"....it's not a "Human Level 3 Monk with AC14"; same thing with any offensive abilities or miscellaneous abilities). Now take that write up and give it to a friend who is also a DM; ask him/her to 'figure out the CR for you'.

I think the problem is going into it KNOWING that you just slapped a PC class/level onto something and, because of that, thinking that it somehow makes the CR calculation "different". It doesn't. The CR calculation, as others have tried to point out, doesn't care where something is coming from...only what it does or can do. Hence, by rewording everything and getting someone ELSE to do the CR, you remove your own bias from the calculation.

Either that, or you could just write it all down, think about what you THINK it's CR should be, and just slap it on there. Honestly, it won't matter. Because in the end, you, as DM, are likely to 'fudge' the CR/XP anyway based on how tough the encounter ACTUALLY was anyway. (...thus, my original comment about CR being utter hogwash.. ;) ).

[AdAstra]

Therefore a fifth level Fighter should probably be rated somewhere around CR 4-5, since the first Fighter can beat the CR 3 Knight but against another Fighter 5 it's a toss up, by definition. Agreed?

Seems about right? The estimations I did were really rough though. It would probably be a good idea to see what things have a solid chance of killing the Fighter 1 on 1.

For example, a CR 6 Mage has a solid chance of bringing down a Str based fighter in two turns using Fireball. Of course, it would have to survive at least one of the fighter's turns, more if Second Wind gives above average HP. Definitely a very chancy matchup, but the fighter could win with a bit of luck. Of course, this is a very asymmetrical fight, so it doesn't tell us as much, especially since Mages are going to have a hard time surviving without some kind of backup.

A CR 5 Gladiator pretty handily outclasses a Level 5 fighter. Three attacks in melee vs two, more than twice the health of most characters at that level, and solid AC when Parry is taken into account. A Barbarian would likely have better luck, probably has a chance due to Rage.

CR 4-5 is probably accurate.

[MaxWilson]

Seems about right? The estimations I did were really rough though. It would probably be a good idea to see what things have a solid chance of killing the Fighter 1 on 1.

For example, a CR 6 Mage has a solid chance of bringing down a Str based fighter in two turns using Fireball. Of course, it would have to survive at least one of the fighter's turns, more if Second Wind gives above average HP. Definitely a very chancy matchup, but the fighter could win with a bit of luck. Of course, this is a very asymmetrical fight, so it doesn't tell us as much, especially since Mages are going to have a hard time surviving without some kind of backup.

A CR 5 Gladiator pretty handily outclasses a Level 5 fighter. Three attacks in melee vs two, more than twice the health of most characters at that level, and solid AC when Parry is taken into account. A Barbarian would likely have better luck, probably has a chance due to Rage.

CR 4-5 is probably accurate.

That seems reasonable.

My Monte Carlo sim (https://repl.it/repls/FickleSlowScientists), using a standard Str 16 Dex 14 Con 14 PAM GWM fighter 5 w/ splint mail and Defense style, using the best tactics I know for that build, achieves a win rate of only 17% vs. a Gladiator using good-but-not-quite-optimal-tactics (no spear-chucking, no indefinite waiting games, because they're too hard to program and they aren't tactics I would use as a DM against a PC). A 6th level Fighter who picks up HAM wins about 36% of the time instead.

If the Fighter 6 takes Prodigy (Athletics) instead of HAM his win rate increases to 50%.

Edit: corrected numbers after a serious bug fix.

A typical fight goes something like this:

Fighter 5 wins initiative!
Fighter 5 has 44 HP, Gladiator has 112 HP
Action surge!
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Gladiator parries!
Fighter misses!
Fighter 5 has 44 HP, Gladiator has 112 HP
Fighter casts Shield!
Gladiator hits for 7 damage!
Gladiator misses!
Gladiator misses!
Fighter 5 has 37 HP, Gladiator has 112 HP
Fighter heals 6 HP with Second Wind
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter 5 has 43 HP, Gladiator has 112 HP
Fighter casts Shield!
Gladiator misses!
Gladiator misses!
Gladiator misses!
Fighter 5 has 43 HP, Gladiator has 112 HP
Fighter tries to shove gladiator prone but fails.
Gladiator is knocked down!
Fighter misses!
Gladiator's opportunity attack misses!
Fighter 5 has 43 HP, Gladiator has 112 HP
Gladiator gets up
Fighter's PAM reaction misses!
Fighter 5 has 43 HP, Gladiator has 112 HP
Gladiator is knocked down!
Fighter misses!
Fighter GWM crits for 17 damage!
Gladiator parries!
Fighter misses!
Fighter 5 has 43 HP, Gladiator has 95 HP
Gladiator gets up
Fighter's PAM reaction hits for 13 damage!
Fighter 5 has 43 HP, Gladiator has 82 HP
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter crits for 9 damage!
Fighter misses!
Fighter 5 has 43 HP, Gladiator has 73 HP
Gladiator misses!
Gladiator crits for 12 damage!
Gladiator misses!
Fighter 5 has 31 HP, Gladiator has 73 HP
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter misses!
Fighter 5 has 31 HP, Gladiator has 73 HP
Gladiator misses!
Gladiator misses!
Gladiator misses!
Fighter 5 has 31 HP, Gladiator has 73 HP
Gladiator is knocked down!
Fighter misses!
Fighter misses!
Fighter casts Shield!
Gladiator's opportunity attack hits for 9 damage!
Fighter 5 has 31 HP, Gladiator has 73 HP
Gladiator gets up
Fighter 5 has 31 HP, Gladiator has 73 HP
Gladiator is knocked down!
Fighter misses!
Fighter misses!
Gladiator's opportunity attack misses!
Fighter 5 has 31 HP, Gladiator has 73 HP
Gladiator gets up
Fighter's PAM reaction crits for 14 damage!
Fighter 5 has 31 HP, Gladiator has 59 HP
Fighter tries to shove gladiator prone but fails.
Gladiator is knocked down!
Fighter misses!
Gladiator's opportunity attack misses!
Fighter 5 has 31 HP, Gladiator has 59 HP
Gladiator gets up
Gladiator parries!
Fighter's PAM reaction misses!
Fighter 5 has 31 HP, Gladiator has 59 HP
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter misses!
Fighter 5 has 31 HP, Gladiator has 59 HP
Gladiator hits for 12 damage!
Gladiator misses!
Gladiator misses!
Fighter 5 has 19 HP, Gladiator has 59 HP
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Gladiator parries!
Fighter misses!
Fighter 5 has 19 HP, Gladiator has 59 HP
Gladiator hits for 9 damage!
Gladiator misses!
Gladiator hits for 8 damage!
Gladiator has knocked fighter down!
Fighter 5 has 2 HP, Gladiator has 59 HP
Fighter gets back up
Fighter tries to shove gladiator prone but fails.
Gladiator is knocked down!
Fighter GWM hits for 17 damage!
Fighter 5 has 2 HP, Gladiator has 42 HP
Gladiator gets up
Gladiator misses!
Gladiator misses!
Gladiator hits for 9 damage!
Gladiator wins!

A rare victory for the fighter (17%) looks like this:

Gladiator wins initiative!
Fighter 5 has 44 HP, Gladiator has 112 HP
Gladiator misses!
Gladiator misses!
Fighter casts Shield!
Gladiator misses!
Fighter 5 has 44 HP, Gladiator has 112 HP
Action surge!
Gladiator is knocked down!
Fighter misses!
Gladiator parries!
Fighter misses!
Fighter misses!
Fighter GWM hits for 14 damage!
Fighter 5 has 44 HP, Gladiator has 98 HP
Gladiator gets up
Fighter's PAM reaction misses!
Fighter 5 has 44 HP, Gladiator has 98 HP
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter misses!
Fighter 5 has 44 HP, Gladiator has 98 HP
Fighter casts Shield!
Gladiator misses!
Gladiator misses!
Gladiator misses!
Fighter 5 has 44 HP, Gladiator has 98 HP
Gladiator is knocked down!
Fighter misses!
Gladiator parries!
Fighter misses!
Fighter 5 has 44 HP, Gladiator has 98 HP
Gladiator gets up
Fighter's PAM reaction misses!
Fighter 5 has 44 HP, Gladiator has 98 HP
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter misses!
Fighter 5 has 44 HP, Gladiator has 98 HP
Gladiator misses!
Gladiator misses!
Gladiator misses!
Fighter 5 has 44 HP, Gladiator has 98 HP
Fighter tries to shove gladiator prone but fails.
Gladiator is knocked down!
Fighter GWM crits for 17 damage!
Fighter GWM crits for 23 damage!
Gladiator's opportunity attack misses!
Fighter 5 has 44 HP, Gladiator has 58 HP
Gladiator gets up
Fighter's PAM reaction misses!
Fighter 5 has 44 HP, Gladiator has 58 HP
Gladiator is knocked down!
Fighter GWM hits for 21 damage!
Gladiator parries!
Fighter misses!
Fighter 5 has 44 HP, Gladiator has 37 HP
Gladiator gets up
Fighter's PAM reaction crits for 14 damage!
Fighter 5 has 44 HP, Gladiator has 23 HP
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter hits for 7 damage!
Fighter 5 has 44 HP, Gladiator has 16 HP
Fighter casts Shield!
Gladiator hits for 9 damage!
Gladiator misses!
Gladiator misses!
Fighter 5 has 35 HP, Gladiator has 16 HP
Fighter heals 15 HP with Second Wind
Gladiator is knocked down!
Gladiator parries!
Fighter misses!
Fighter 5 has 50 HP, Gladiator has 16 HP
Gladiator gets up
Fighter's PAM reaction misses!
Fighter 5 has 50 HP, Gladiator has 16 HP
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter hits for 4 damage!
Fighter 5 has 50 HP, Gladiator has 12 HP
Gladiator misses!
Gladiator hits for 9 damage!
Gladiator hits for 10 damage!
Gladiator has knocked fighter down!
Fighter 5 has 31 HP, Gladiator has 12 HP
Fighter gets back up
Fighter tries to shove gladiator prone but fails.
Gladiator is knocked down!
Fighter misses!
Fighter 5 has 31 HP, Gladiator has 12 HP
Gladiator gets up
Gladiator crits for 12 damage!
Gladiator misses!
Gladiator misses!
Fighter 5 has 19 HP, Gladiator has 12 HP
Fighter tries to shove gladiator prone but fails.
Gladiator is knocked down!
Fighter GWM crits for 17 damage!
Fighter 5 wins!

[Yakk]

Ya; parry is super-strong one-on-one and PAM is super-strong against multiple foes, as you get a tap on everyone who approaches you.

So 17% WR is pretty even-steven.

[MaxWilson]

Ya; parry is super-strong one-on-one and PAM is super-strong against multiple foes, as you get a tap on everyone who approaches you.

No, because you only have one reaction (unless you're an 18th level Cavalier).

[AdAstra]

Ya; parry is super-strong one-on-one and PAM is super-strong against multiple foes, as you get a tap on everyone who approaches you.

So 17% WR is pretty even-steven.

Parry is stronger the fewer attacks you face, but PAM is only better against multiple targets if they’re gradually filing into your reach, since you only have the one reaction. If you take your turn, then eight enemies surround you before you get another one, you get the same number of extra attacks (1) as if a single large monster had charged you.

[Zalabim]

That seems reasonable.

My Monte Carlo sim (https://repl.it/repls/FickleSlowScientists), using a standard Str 16 Dex 14 Con 14 PAM GWM fighter 5 w/ splint mail and Defense style, using the best tactics I know for that build, achieves a win rate of only 17% vs. a Gladiator using good-but-not-quite-optimal-tactics (no spear-chucking, no indefinite waiting games, because they're too hard to program and they aren't tactics I would use as a DM against a PC). A 6th level Fighter who picks up HAM wins about 36% of the time instead.

If the Fighter 6 takes Prodigy (Athletics) instead of HAM his win rate increases to 50%.

Edit: corrected numbers after a serious bug fix.

A typical fight goes something like this:

Fighter 5 wins initiative!
Fighter 5 has 44 HP, Gladiator has 112 HP
Action surge!
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Gladiator parries!
Fighter misses!
Fighter 5 has 44 HP, Gladiator has 112 HP
Fighter casts Shield!
Gladiator hits for 7 damage!
Gladiator misses!
Gladiator misses!
Fighter 5 has 37 HP, Gladiator has 112 HP
Fighter heals 6 HP with Second Wind
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter 5 has 43 HP, Gladiator has 112 HP
Fighter casts Shield!
Gladiator misses!
Gladiator misses!
Gladiator misses!
Fighter 5 has 43 HP, Gladiator has 112 HP
Fighter tries to shove gladiator prone but fails.
Gladiator is knocked down!
Fighter misses!
Gladiator's opportunity attack misses!
Fighter 5 has 43 HP, Gladiator has 112 HP
Gladiator gets up
Fighter's PAM reaction misses!
Fighter 5 has 43 HP, Gladiator has 112 HP
Gladiator is knocked down!
Fighter misses!
Fighter GWM crits for 17 damage!
Gladiator parries!
Fighter misses!
Fighter 5 has 43 HP, Gladiator has 95 HP
Gladiator gets up
Fighter's PAM reaction hits for 13 damage!
Fighter 5 has 43 HP, Gladiator has 82 HP
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter crits for 9 damage!
Fighter misses!
Fighter 5 has 43 HP, Gladiator has 73 HP
Gladiator misses!
Gladiator crits for 12 damage!
Gladiator misses!
Fighter 5 has 31 HP, Gladiator has 73 HP
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter misses!
Fighter 5 has 31 HP, Gladiator has 73 HP
Gladiator misses!
Gladiator misses!
Gladiator misses!
Fighter 5 has 31 HP, Gladiator has 73 HP
Gladiator is knocked down!
Fighter misses!
Fighter misses!
Fighter casts Shield!
Gladiator's opportunity attack hits for 9 damage!
Fighter 5 has 31 HP, Gladiator has 73 HP
Gladiator gets up
Fighter 5 has 31 HP, Gladiator has 73 HP
Gladiator is knocked down!
Fighter misses!
Fighter misses!
Gladiator's opportunity attack misses!
Fighter 5 has 31 HP, Gladiator has 73 HP
Gladiator gets up
Fighter's PAM reaction crits for 14 damage!
Fighter 5 has 31 HP, Gladiator has 59 HP
Fighter tries to shove gladiator prone but fails.
Gladiator is knocked down!
Fighter misses!
Gladiator's opportunity attack misses!
Fighter 5 has 31 HP, Gladiator has 59 HP
Gladiator gets up
Gladiator parries!
Fighter's PAM reaction misses!
Fighter 5 has 31 HP, Gladiator has 59 HP
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter misses!
Fighter 5 has 31 HP, Gladiator has 59 HP
Gladiator hits for 12 damage!
Gladiator misses!
Gladiator misses!
Fighter 5 has 19 HP, Gladiator has 59 HP
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Gladiator parries!
Fighter misses!
Fighter 5 has 19 HP, Gladiator has 59 HP
Gladiator hits for 9 damage!
Gladiator misses!
Gladiator hits for 8 damage!
Gladiator has knocked fighter down!
Fighter 5 has 2 HP, Gladiator has 59 HP
Fighter gets back up
Fighter tries to shove gladiator prone but fails.
Gladiator is knocked down!
Fighter GWM hits for 17 damage!
Fighter 5 has 2 HP, Gladiator has 42 HP
Gladiator gets up
Gladiator misses!
Gladiator misses!
Gladiator hits for 9 damage!
Gladiator wins!

A rare victory for the fighter (17%) looks like this:

Gladiator wins initiative!
Fighter 5 has 44 HP, Gladiator has 112 HP
Gladiator misses!
Gladiator misses!
Fighter casts Shield!
Gladiator misses!
Fighter 5 has 44 HP, Gladiator has 112 HP
Action surge!
Gladiator is knocked down!
Fighter misses!
Gladiator parries!
Fighter misses!
Fighter misses!
Fighter GWM hits for 14 damage!
Fighter 5 has 44 HP, Gladiator has 98 HP
Gladiator gets up
Fighter's PAM reaction misses!
Fighter 5 has 44 HP, Gladiator has 98 HP
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter misses!
Fighter 5 has 44 HP, Gladiator has 98 HP
Fighter casts Shield!
Gladiator misses!
Gladiator misses!
Gladiator misses!
Fighter 5 has 44 HP, Gladiator has 98 HP
Gladiator is knocked down!
Fighter misses!
Gladiator parries!
Fighter misses!
Fighter 5 has 44 HP, Gladiator has 98 HP
Gladiator gets up
Fighter's PAM reaction misses!
Fighter 5 has 44 HP, Gladiator has 98 HP
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter misses!
Fighter 5 has 44 HP, Gladiator has 98 HP
Gladiator misses!
Gladiator misses!
Gladiator misses!
Fighter 5 has 44 HP, Gladiator has 98 HP
Fighter tries to shove gladiator prone but fails.
Gladiator is knocked down!
Fighter GWM crits for 17 damage!
Fighter GWM crits for 23 damage!
Gladiator's opportunity attack misses!
Fighter 5 has 44 HP, Gladiator has 58 HP
Gladiator gets up
Fighter's PAM reaction misses!
Fighter 5 has 44 HP, Gladiator has 58 HP
Gladiator is knocked down!
Fighter GWM hits for 21 damage!
Gladiator parries!
Fighter misses!
Fighter 5 has 44 HP, Gladiator has 37 HP
Gladiator gets up
Fighter's PAM reaction crits for 14 damage!
Fighter 5 has 44 HP, Gladiator has 23 HP
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter hits for 7 damage!
Fighter 5 has 44 HP, Gladiator has 16 HP
Fighter casts Shield!
Gladiator hits for 9 damage!
Gladiator misses!
Gladiator misses!
Fighter 5 has 35 HP, Gladiator has 16 HP
Fighter heals 15 HP with Second Wind
Gladiator is knocked down!
Gladiator parries!
Fighter misses!
Fighter 5 has 50 HP, Gladiator has 16 HP
Gladiator gets up
Fighter's PAM reaction misses!
Fighter 5 has 50 HP, Gladiator has 16 HP
Fighter tries to shove gladiator prone but fails.
Fighter tries to shove gladiator prone but fails.
Fighter hits for 4 damage!
Fighter 5 has 50 HP, Gladiator has 12 HP
Gladiator misses!
Gladiator hits for 9 damage!
Gladiator hits for 10 damage!
Gladiator has knocked fighter down!
Fighter 5 has 31 HP, Gladiator has 12 HP
Fighter gets back up
Fighter tries to shove gladiator prone but fails.
Gladiator is knocked down!
Fighter misses!
Fighter 5 has 31 HP, Gladiator has 12 HP
Gladiator gets up
Gladiator crits for 12 damage!
Gladiator misses!
Gladiator misses!
Fighter 5 has 19 HP, Gladiator has 12 HP
Fighter tries to shove gladiator prone but fails.
Gladiator is knocked down!
Fighter GWM crits for 17 damage!
Fighter 5 wins!
A few things that can help with this kind of sim. Both combatants should only try to knock prone on their first attack. The gladiator by [heh] shield bashing first, and the fighter by shoving. The fighter has to attack with their polearm to be allowed a bonus action attack. The fighter should wait to action surge until a shove does succeed. I still suspect that not shoving at all is better for the fighter, for a better chance to break through parry's reaction limit, but without going into conditional hit chances to account for the small chance of that (which will take more time than I have right now), it does look to marginally lower damage taken and increase damage done in the short term. If you want to account for throwing spears, remember that you can give creatures an average 2d4 throwable weapons and 2d10 ammunition, and the gladiator is only limited by how fast it can draw spears after the first round of making ranged attacks. I doubt either side can end the battle with purely ranged attacks, even if it's 20 arrows to 5 spears.

Both combatants can fall back after successfully knocking prone, but I think the gladiator benefits from the tactic only if the fighter must press the melee. Decisions about the fighter pulling out a bow if the gladiator retreats and whether the gladiator should dash to get back in melee is just too much depth to plumb. On the fighter's end, making the gladiator throw spears for one round is only a small improvement defensively (after adding in an OA with disadvantage), but with luck, the gladiator would run out of spears then it would be a good advantage. Offensively, the eventual extra OA from PM helps pay for the cost of spending an attack shoving. If the fighter isn't going to kite a little, then don't bother shoving, I think. Conversely, if the gladiator is going to hit and run, bash it until you make it. Even if it's not optimal (PM mitigates some of the cost to the fighter's attacks), it'd be memorable. The gladiator is a particularly bad shoving match, for being small enough to allow it but actually being good at avoiding it.

It's interesting to me that the battle master's superiority dice have almost the same average HP value as the three casts of shield here (parry stops 5.5 damage with +1 dex, shield blocks ~8 damage on average), though that's 1) undervaluing using shield to avoid a shield bash and 2) assuming the fighter uses shield knowing the attack hit but not knowing the attack's actual roll (other than "not a crit").


Only shove/bash once
Only action surge against prone
~5 spears can be thrown
Both sides can hit-and-run a prone opponent.

[MaxWilson]

Excellent points Zalabim! I'd love to get the simulator into a condition such that other people can easily tweak monster or PC tactics. In fact I've been working on a program like that for a while now, but it's not ready yet so I wrote this simulation as a custom one-off script.

I'm sure you're right, all of those things would improve tactics. And IMO this kind of high-Deadly tactical puzzle solving to try to win is much, much more fun to play out than what should logically happen in non-Deadly fights: the weaker party (the NPC) recognizes its weakness and attempts to flee or surrender if possible, otherwise it either dies as hard as possible or has a nervous breakdown and dies in a panic. Non-Deadly fights still have *consequences* that are fun to roleplay, but in the combat-light games I prefer, I dislike rolling the actual fight unless the players really want to. But tough fights are always fun, even if death isn't actually on the line, because you don't know who's going to win, and tough fights are usually either rated high-Deadly or are against highly intelligent tool using opponents.

Obviously the Fighter's best strategy would be to be a Mobile Crossbow Expert, possibly a Sharpshooter too, so the Gladiator can't pin him in melee at all, and so the Gladiator has to win with only 2d4 spears and can't Parry. I didn't want to use that build for the sim though because it felt a little bit cheesy to exploit Mobile to that extent--at that point it's not really about computing effective CR at all, it's about disproving the utility of CR, since that same Fighter could easily kill a CR 16 Iron Golem as easily as the CR 5 Champion, if he had an adamantine weapon.

[Zalabim]

Sure, the fighter could be better equppied to counter the gladiator. Then I'd wonder what's a bad opponent for a mobile crossbow expert. It's interesting to me to see where it's a good idea to switch to secondary tactics. Of course that mobile archer can evade and punish the gladiator, but should the PM try their longbow here too? Or maybe they're in an arena surrounded by spears or bundles of javelins and they both won't run out. Should two strong melee warriors with a prone-centric combat routine actually hurl insults and spears when pitted against each other? I can calculate some by hand, figure out where relative advantages lie, but I don't have the coding knowledge to tinker with a simulator. There's so many options to consider too. We know the combatant's stats and abilities, but would they know each other? Even down to how much each knows about the to hit rolls, for shield and party, influences their value. I think the gladiator is my new "hardest to trip trippable in the MM." I imagine a lot of players wouldn't know that.

[MaxWilson]

I can calculate some by hand, figure out where relative advantages lie, but I don't have the coding knowledge to tinker with a simulator.

I would really, really love to remove this obstacle for you. It makes me want to cry when I see people computing PC build DPR or encounter difficulty with spreadsheets instead of Monte Carlo simulations of actual encounters they would like to run.

Doing a sim of an individual fight is pretty easy, but doing a program that can simulate any fight in a way that doesn't risk tripping over WotC's copyright (can't just enter all monsters and PC data up front--have to be able to handle new content and new rules/ad hoc rulings in the middle of the fight) AND yet is easy for non-programmers to use is a more complex problem that I haven't licked yet.

But solving this problem is the main thing that keeps me on 5E-related Internet forums despite my growing preference for AD&D at the table: 5E really is more interesting than AD&D as a combat simulation or CRPG, and thinking about weird builds and tactics really is fun.