D&D 5E Classes, and the structure of DPR

clearstream

(He, Him)
In another thread (on monks) I mentioned some analysis I'd recently done on class progression, relating to how they increase their DPR. It may be of general interest so I will post it here.
  1. All classes have step increases in DPR at 5th and 11th level. There are no exceptions.
  2. The design intent appears to be roughly they do 1x their base attack in tier 1, 2x in tier 2, and 3x in tier 3.
The interesting part is how different classes achieve this. There are three kinds of scaling
  • Some classes get another attack (fighters, warlocks using EB)
  • Other classes get another die per attack (paladins, most casters in their cantrips, rogue sneak attack)
  • A few get a sub-class feature that boosts their DPR expectation (monk, ranger)
That can be simplified to two approaches - either A) get another attack, or B) get more damage on your attacks. Fighters are A, monks are B. That gives the designers the ability to create some fundamental diversity in approaches to combat.

My view is that the use of sub-classes to boost DPR expectation is on the one hand quite interesting, while on the other hand noticeably problematic. As an example of how it can be interesting, compare the ranger Volley, Stalker's Flurry, Distant Strike, and Fey Reinforcements. These are all imaginative ways to add to DPR expectation. They're also a good example of how sub-class features can be underwhelming, or at least uneven. For further comparison, monks gain part of their 11th level DPR step in their martial die, and the other part in their sub-class feature.

What makes this even further interesting is that I noticed that at certain steps where other classes scaled offense, one or two - like ranger - gained something defensive. I suspect that the consistent undervaluing of defenses in theorycrafting probably leads to some underestimation of these classes' effectiveness in play.

In the end, I found that calculating the value of the barbarian hit die in ASIs, and then using that to assess the total value of features in each tier, showed a very intentional approach to the design. Some classes stand out as front loaded, others gain more as they go along. Diversity seemed consciously designed in. One can also see how - for the usual campaign (capping out at low-tier-3) - multiclassing is imbalanced. No real surprise. If there is enough interest I can tidy up my spreadsheet and post the progression structure as I analysed it.


[EDIT I'd like to draw attention to my posts from the bottom of page 4, which concretely develop this investigation.]
 
Last edited:

log in or register to remove this ad



ECMO3

Hero
  1. All classes have step increases in DPR at 5th and 11th level. There are no exceptions
  • Some classes get another attack (fighters, warlocks using EB)
  • Other classes get another die per attack (paladins, most casters in their cantrips, rogue sneak attack)
  • A few get a sub-class feature that boosts their DPR expectation (monk, ranger)
That can be simplified to two approaches - either A) get another attack, or B) get more damage on your attacks. Fighters are A, monks are B. That gives the designers the ability to create some fundamental diversity in approaches to combat.
Your premise and some of your supporting argueents are incorrect.

First Rogues get many smaller steps, not steps at 5th and 11th.

Second Rangers and Monk subclasses do not all get a step increase in damage at 11th level. Some subclasses get a damage bonus, but others don't and those that do for the most part do not get a step that takes them from 2x base to 3xbase. In many cases it is about 1 point of extra damage per turn.

Finally Monks (and Paladins and Rangers) get extra attacks at 5th level, not a boost
 

Charlaquin

Goblin Queen (She/Her/Hers)
Your premise and some of your supporting argueents are incorrect.
Seems about right to me.
First Rogues get many smaller steps, not steps at 5th and 11th.
But they keep up with the step increases other classes get.
Second Rangers and Monk subclasses do not all get a step increase in damage at 11th level. Some subclasses get a damage bonus, but others don't and those that do for the most part do not get a step that takes them from 2x base to 3xbase. In many cases it is about 1 point of extra damage per turn.
Remember that monks are also getting a martial arts die size increase at 11th, which accounts for part of their step increase, the other part being from subclass. Some subclasses are still below par though. Similarly, most ranger subclasses don’t give enough of a damage increase at 11th.
Finally Monks (and Paladins and Rangers) get extra attacks at 5th level, not a boost
Right, but all three get boosts at 11th rather than an additional extra attack.
 

ECMO3

Hero
Remember that monks are also getting a martial arts die size increase at 11th, which accounts for part of their step increase, the other part being from subclass. Some subclasses are still below par though. Similarly, most ranger subclasses don’t give enough of a damage increase at 11th.
The extra martial arts die is worth 1 point of damage per turn as is the increase in many of the Ranger subclasses. That is nowhere near going from 2x base to 3x base.

The other thing to mention is the full casters and while their cantrips are scaling exactly to 2xDPR and 3xDPR, they are using them less, so their DPR will generally go down from tier 2 to tier 3 as they cast more spells and use less cantrips. Not necessarily true for a blaster mage or a bladesinger but it is true for most caster builds, it can be true for Rangers as well if your build is more of a Gish Ranger that uses a lot of spells as opposed to a primarily martial Ranger.
 

TheOneGargoyle

Explorer
In another thread (on monks) I mentioned some analysis I'd recently done on class progression, relating to how they increase their DPR. It may be of general interest so I will post it here.
  1. All classes have step increases in DPR at 5th and 11th level. There are no exceptions.
  2. The design intent appears to be roughly they do 1x their base attack in tier 1, 2x in tier 2, and 3x in tier 3.
I don't understand your basis for claiming this.

From 5th level to 11th level, almost all base classes get significant and similar damage boosts:
* fighters gain an extra attack boosting their dmg by approx 9.5 (1d8+5),
* rogues gain 3d6 SA dmg for an avg boost of 10.5,
* paladins gain approx 2d8 with Imp Divine Smite and higher spell slots and more of them for an avg boost of about 9.0,
* warlocks gain 1d10+5 for approx 10.5 dmg increase,
* all pure casters get 3 levels higher spell slots and more of them increasing dmg output by much more than 10 pts.
So this would suggest that the bar for these "Step Increases" is at around 10 pts per round.

Monks between 5th and 11th level gain +1 dmg per round on avg (d6->d8), or +2 if they flurry. This is not remotely close to +10 dmg. Do you classify such a small incremental increase as a "step increase" ? I would think not.

The interesting part is how different classes achieve this. There are three kinds of scaling
  • Some classes get another attack (fighters, warlocks using EB)
  • Other classes get another die per attack (paladins, most casters in their cantrips, rogue sneak attack)
  • A few get a sub-class feature that boosts their DPR expectation (monk, ranger)
So, for this theory to be right, two things would need to be true at these level ranges (say from L5 - L11):
1) monks would need to get the remainder of a similar sort of dps boost from all (or the very least, the vast majority) of their subclasses, and
2) other classes subclasses would need to give no (or extremely minor) dmg boosts

Let's fact-check to see if these 2 things hold true:
1) monk subclasses - 5 of them give no dmg boost whatsoever (Long Death, Mercy, Open Hand, Shadow, Ascendant Dragon), 5 of them give a small dmg boost (Astral Self extra MA die ~+4.5, Drunken Master occasionally avoid disadv, Four Elements spend 1 more ki on a spell, Kensei +1 or +2 more than existing magic weapon, Sun Soul situationally Sunburst might do more dmg), and precisely zero subclasses give a boost which would make up the 8-9 pts they would need to equal the other classes.

2) other classes subclasses
- fighter: 6 give dmg boosts (arcane archer, PDK, battlemaster, EK, psi-warrior, rune knight) and 4 don't (cavalier, champion, echo knight, samurai), so more subclasses give dmg boosts than for monk !
- rogue: 3 give dmg boosts (arcane trickster, phantom, soulknife) and 6 don't (assassin, inquisitive, mastermind, scout, swashbuckler, thief), so even with sneak attack boosts every 2 levels in the base class, a third of rogue subclasses give dmg boosts on top of this !
- paladin: 4 give dmg boosts (crown, glory, vengeance, oathbreaker) and 5 don't (ancients, conquest, devotion, redemption, watchers), so approx half give dmg boost !
- warlock: 3 give dmg boosts (celestial, hexblade, undead), 6 don't (archfey, fathomless, fiend, genie, GOO, undying) so a third of warlock subclasses give dmg boosts, and that's without factoring in 2 extra invocations which can potentially give damage boosts too (sometimes depending on pact) such as Eldritch Spear, Maddening Hex, Pact of the Talisman, Superior Pact Weapon, etc.

So, verdict:
1) False.
2) False.

My view is that the use of sub-classes to boost DPR expectation is on the one hand quite interesting, while on the other hand noticeably problematic. As an example of how it can be interesting, compare the ranger Volley, Stalker's Flurry, Distant Strike, and Fey Reinforcements. These are all imaginative ways to add to DPR expectation. They're also a good example of how sub-class features can be underwhelming, or at least uneven. For further comparison, monks gain part of their 11th level DPR step in their martial die, and the other part in their sub-class feature.
As shown above, other classes subclasses give DPR boosts too, so this is just .... false ?
 

clearstream

(He, Him)
I don't understand your basis for claiming this.

From 5th level to 11th level, almost all base classes get significant and similar damage boosts:
* fighters gain an extra attack boosting their dmg by approx 9.5 (1d8+5),
At 5th level they gain an extra attack, at 11th level they gain an extra attack. These are the step increases I am thinking of. Can you clarify what you don't understand?

Perhaps it will be clearer if I introduce the term reference-structure? So we desire diversity, but the diversity is cohered - balanced a-priori - against the reference structure.

* rogues gain 3d6 SA dmg for an avg boost of 10.5,
* paladins gain approx 2d8 with Imp Divine Smite and higher spell slots and more of them for an avg boost of about 9.0,
* warlocks gain 1d10+5 for approx 10.5 dmg increase,
* all pure casters get 3 levels higher spell slots and more of them increasing dmg output by much more than 10 pts.
So this would suggest that the bar for these "Step Increases" is at around 10 pts per round.
It's true that the picture in detail is more complicated, I am precising the general structure the designers are using. There is a progression in DPR, and it is steppy, with noticeable steps for for all classes at 5th and 11th level (and later of course, but I am focusing on the most played tiers.)

Your premise and some of your supporting argueents are incorrect.

First Rogues get many smaller steps, not steps at 5th and 11th.

Second Rangers and Monk subclasses do not all get a step increase in damage at 11th level. Some subclasses get a damage bonus, but others don't and those that do for the most part do not get a step that takes them from 2x base to 3xbase. In many cases it is about 1 point of extra damage per turn.

Finally Monks (and Paladins and Rangers) get extra attacks at 5th level, not a boost
An extra attack is a step. Rogues do spread their increases out more than other classes, but the overall transition (increase in DPR) is by design about the same.

Monks between 5th and 11th level gain +1 dmg per round on avg (d6->d8), or +2 if they flurry. This is not remotely close to +10 dmg. Do you classify such a small incremental increase as a "step increase" ? I would think not.
Take a look at this baseline versus the monk from a Treantmonk video. This is how a monk using their flurry of blows looks against a warlock using EB+AB+Hex (Treantmonk's baseline). When you graph expected DPR you will see something similar to this for every class. Look where the 'step increases' are occurring - 5th and 11th.

monk steps.png


So, for this theory to be right, two things would need to be true at these level ranges (say from L5 - L11):
1) monks would need to get the remainder of a similar sort of dps boost from all (or the very least, the vast majority) of their subclasses, and
2) other classes subclasses would need to give no (or extremely minor) dmg boosts

Let's fact-check to see if these 2 things hold true:
1) monk subclasses - 5 of them give no dmg boost whatsoever (Long Death, Mercy, Open Hand, Shadow, Ascendant Dragon), 5 of them give a small dmg boost (Astral Self extra MA die ~+4.5, Drunken Master occasionally avoid disadv, Four Elements spend 1 more ki on a spell, Kensei +1 or +2 more than existing magic weapon, Sun Soul situationally Sunburst might do more dmg), and precisely zero subclasses give a boost which would make up the 8-9 pts they would need to equal the other classes.
I think it is important to call out that the designers are attempting to create a varied game, with many possible approaches to play, so we do not see uniformity. I am giving a general description of the structure, to which there is variation that can be best understood with reference to the general structure. Let's look at the five you list

Long Death - not dying is expected to improve your DPR. You'll often see with monks the designers being tricksy with defences that are also somehow offences.

Mercy - Flurry of Healing and Harm. The free Hand of Harm means you get your martial die + Wis mod on every flurry. In play that is a damage boost as you now have 11 ki and can regularly flurry (it seems expected by the maths that you will do). However, the sub-class is intended to give monks a healing role and so some of the power that could have been given in damage has instead been given in healing.

Open Hand - this is one of my favourite monk sub-classes and shows a case where a class is given a defence in place of stronger attacks. The designers again have traded part of the DPR step expectation for something else. That is very common across class and sub-class features. You often see that at the point where the reference-structure would suggest a DPR increase, a defence is gained instead. The point is that there is consistency to this across classes.

Shadow - on effect of invisibility is advantage on attacks against creatures that cannot see you, so the 11th level feature for shadow monks is expected to increase their DPR. Not all designs are equally successful, and the designers did something here that you see across the monk class far more than others IMO, which is to inadequately consider the action economy. I think they hoped to provide a defence that would also be a damage boost, and prenerfed it by costing an action.

Ascendent Dragon - I don't see how the Aspect of Wyrm destructive waves and the 11th level scaling on Breath do not count as a damage boost. Remember that these are in theory only half the monk's step at 11th. The other half is in their martial die.


2) other classes subclasses
- fighter: 6 give dmg boosts (arcane archer, PDK, battlemaster, EK, psi-warrior, rune knight) and 4 don't (cavalier, champion, echo knight, samurai), so more subclasses give dmg boosts than for monk !
- rogue: 3 give dmg boosts (arcane trickster, phantom, soulknife) and 6 don't (assassin, inquisitive, mastermind, scout, swashbuckler, thief), so even with sneak attack boosts every 2 levels in the base class, a third of rogue subclasses give dmg boosts on top of this !
- paladin: 4 give dmg boosts (crown, glory, vengeance, oathbreaker) and 5 don't (ancients, conquest, devotion, redemption, watchers), so approx half give dmg boost !
- warlock: 3 give dmg boosts (celestial, hexblade, undead), 6 don't (archfey, fathomless, fiend, genie, GOO, undying) so a third of warlock subclasses give dmg boosts, and that's without factoring in 2 extra invocations which can potentially give damage boosts too (sometimes depending on pact) such as Eldritch Spear, Maddening Hex, Pact of the Talisman, Superior Pact Weapon, etc.

So, verdict:
1) False.
2) False.


As shown above, other classes subclasses give DPR boosts too, so this is just .... false ?
I believe you are taking what I am saying in very much the wrong way. Yes, 5e is diverse. How do designers manage such diversity? They work to rubrics or reference-structures. The variation that exists varies in reference to that structure.

To give just one case, think about how Treantmonk uses the warlock as his baseline for evaluating classes. The reason that is so effective is because EB gains more bolts (aka extra attacks) and adds the ability modifier, which is itself scaling (agonizing), and has an easy to predict boost (Hex). He could have used the fighter, but then you have to consider different weapons, fighting styles, action surges and so on.

Treantmonk's warlock baseline works because it makes visible the 'steppy' reference-structure. That provides the context for other features to be better understood.
 

cbwjm

Legend
In general, classes get a damage increase, some though get a defensive increase. Barbarians don't get any additional DPR at level 11, however, they have a chance to keep fighting if dropped to 0 at level 11 with relentless rage instead.

Artificers gain spell storing item, which could be used for offensive spells but doesn't have to be.

While many classes do gain an offensive boost, some gain a defensive one which do not provide a DPR increase.
 

clearstream

(He, Him)
I'd love to see it.
I will second that! Please post it if you have the time.
Okay, I will need a little time but fear not, I will get there.

Time, because I performed the deconstruction in order to reconstruct an E6 kind of scaling for 5E. In this mod, I wanted to stop class scaling at 6th level, and thereafter only increase variety in approaches, so I needed to understand what the backbone of each class was. Unfortunately, that also means the information is currently broken up into tables suited for my purposes, and need to be stitched back together.

Here is a slice of my breakdown at levels 6 to 11. The three columns - 1, 2, 4 - are my way of evaluating each feature against an ASI. So an ASI is 2, a half-ASI is 1, and stronger-than-an-ASI is 4. I used the Detect Balance work by other gamers to validate my assessments. I examined features from PHB, XGE and TCoE. What I will do is stitch back in my level 1-5 breakdown. I think then you will get a strong sense for the patterns. [EDIT I think this table dates from before I realised I need to cost in the HD for each class! Anyway, I will post more - links to sheets I suppose - soon.]

rs 6 to 11.png
 
Last edited:

clearstream

(He, Him)
In general, classes get a damage increase, some though get a defensive increase. Barbarians don't get any additional DPR at level 11, however, they have a chance to keep fighting if dropped to 0 at level 11 with relentless rage instead.

Artificers gain spell storing item, which could be used for offensive spells but doesn't have to be.

While many classes do gain an offensive boost, some gain a defensive one which do not provide a DPR increase.
Exactly, that's why I think it is best to think of this as the baseline or reference-structure. And - crediting other respondents - it is very important to call out that there is diversity. Think about it like this
  1. The tiers are meaningful in 5E. Tier 2 characters will see a step (or steppy) increase in mechanical power over tier 1 characters. And tier 3 over tier 2, etc.
  2. The simplest place that is visible is in fighter extra attack and warlock eldritch blast.
  3. Each class has been consciously designed to have its own approach, but the approaches can be simplified to - A) I attack many times, B) I attack fewer times but do more damage, C) I attack and also defend. Fighters are A. Rogues are B. Rangers are C. Note that 'defend' is where you see even more diversity because it often spills over into explore (e.g. ability to hide).
  4. This simplifies the task of design as it means a designer knows the approach a class is using, and they know how much power to build into its features at each level. You can observe this most easily I think in the designs for sub-classes such as fey wanderer, if you compare its features to other more straightforward classes.
The designers introduce enormous variation as offsets from those backbones, but the intents always ring through. Look at every ranger 11th level feature from PHB, XGE and TCoE and compare those to every fighter 11th level feature. Cost the barbarian d12 HD in ASIs, and consider that against the value of spell slots and spell levels. You will see underlying skeleton bringing cohesion everywhere!
 



Charlaquin

Goblin Queen (She/Her/Hers)
Treantmonk's baseline is very unrealistic.
Having hex up all the time is rather unrealistic.
I mean, that’s the difference between on-paper analysis and actual play. Most of the assumptions that go into DPR calculations don’t hold up in actual play. The point is to understand the underlying structures of the design, not to accurately model gameplay.
 

clearstream

(He, Him)
Treantmonk's baseline is very unrealistic.
Having hex up all the time is rather unrealistic.
Agreed. What I believe is the biggest gap is in evaluating defence. Whiteroom analyses generally assign zero value to defence. That can be partly justified by the argument that the best defence brings you no closer to winning, against which is that it can defer losing... and often has broader applicability outside of combat.
 

FrogReaver

As long as i get to be the frog
Agreed. What I believe is the biggest gap is in evaluating defence. Whiteroom analyses generally assign zero value to defence. That can be partly justified by the argument that the best defence brings you no closer to winning, against which is that it can defer losing... and often has broader applicability outside of combat.
IMO. The defensive options compared to investment cost generally aren't very good.
 

clearstream

(He, Him)
IMO. The defensive options compared to investment cost generally aren't very good.
A problem facing designers is that if defenses are really good, combat takes an undesirably large number of rounds to resolve. If attacks aren't better than defenses, combat never resolves!

To me the best examples are defenses that fit well with the action-economy, and are either narrow or conditional. I kind of dislike the defense fighting style because it is simply always good. I prefer the 11th level monster slayer ranger or open hand monk features.
 

ECMO3

Hero
IMO. The defensive options compared to investment cost generally aren't very good.
Some are, some are not. Armor of Agathys if you can upcast it to 4th level or higher is very good, as is the new Gem Dragon feat. The shield spell is also very good.

If you are going to be in melee often, I would say all three of those are generally better investments than GWM in tier 2+. Generally defensive options get batter at higher levels where offensive options do not.
 

Mort

Legend
Supporter
Some are, some are not. Armor of Agathys if you can upcast it to 4th level or higher is very good, as is the new Gem Dragon feat. The shield spell is also very good.
Gift of the Gem Dragon?

It doesn't actually prevent damage - 2d8 damage and push 10' isn't horrible, but I wouldn't call it great.

Shield is quite good and scales well with level.

Not much experience with Armor of Agathys.



If you are going to be in melee often, I would say all three of those are generally better investments than GWM in tier 2+. Generally defensive options get batter at higher levels where offensive options do not.

Not if the player knows how to build around and utilize GWM properly.

If not, sure, GWM can actually be detrimental to DPR.

But if properly built around and utilized - it's huge.
 

FrogReaver

As long as i get to be the frog
A problem facing designers is that if defenses are really good, combat takes an undesirably large number of rounds to resolve. If attacks aren't better than defenses, combat never resolves!
So you agree that the 'whiteroom analysis' is appropriately valuing defensive options - it's just that you wish defensive options were designed differently?
 

Level Up!

An Advertisement

Advertisement4

Top