D&D General Data from a million DnDBeyond character sheets?

[Clint_L]

Let's say our level 5 berserker barbarian has 55 HP, AC 14, Str 18, Con 16, Dex 12, which is pretty typical. They'll use frenzy, and reckless attacks for this test. Great sword. Up against an owl bear, they do 35.83 damage per round, while taking 11.31 DPR. 5 rounds of survivability, but they kill the owl bear in 2.

Level 5 champion fighter has similar stats but let's say they got chain+shield and dueling, long sword. AC 20, 49 HP. They'll do 15.75 DPR (doubling that for one round if they action surge) while taking 9.8 DPR. They also have 5 rounds of survivability, but will take twice as long to kill the owl bear.

My point is that you have to take offence into account when assessing tanking, because killing enemies faster means taking less damage. Barbarians can tank exceptionally well while putting out big numbers - that berserker barbarian basically threw defence to the wind and still had the survivability of a fairly typical tanky fighter. You can definitely create builds with other classes that are as good or better at defence - eldritch knight was a great example, but only by paying a high price in offence.

And you can build an extremely tanky barbarian, if you want - that barbarian could have AC 20 and not do reckless attacks, for example, and their survivability would be ridiculous (owl bear now needs 11 rounds to kill them, and the barbarian is still doing 26.3 DPR). Folks seldom do this because the baseline barbarian is already a good tank while putting up big numbers, which is fun.

 

[UngeheuerLich]

Not convinced about what specifically?

If you let me know what specifically you disagree with I will try and prove it out mathematically.

How 1 to 10 hp make all that difference for that specific class.

 

[UngeheuerLich]

Let's say our level 5 berserker barbarian has 55 HP, AC 14, Str 18, Con 16, Dex 12, which is pretty typical. They'll use frenzy, and reckless attacks for this test. Great sword. Up against an owl bear, they do 35.83 damage per round, while taking 11.31 DPR. 5 rounds of survivability, but they kill the owl bear in 2.

I'd take 14 con and 14 dex over that and just use scale mail for an AC of 16.

Better ini, 2 more AC and better ranged capabilities. Is it better? I did not do the math, so I don't know. But I probably would like that more.

It is basically trading 5hp (10% less hp) for 10% lower chance to be hit (which are a about 14% of the attack rolls from the owlbear that would have hit.)

The 16 con barbarian has EHP of 60.5, the Dex 14 barbarian has EHP of 59.5.

That is very close. And only unfavourable for the Dex barb, because the Owl Bear has a very high to hit bonus.
If you account the higher initiative and the 1 in 20 chance to have an extra round of attacks might make up for the difference.

 

[FrogReaver]

Let's say our level 5 berserker barbarian has 55 HP, AC 14, Str 18, Con 16, Dex 12, which is pretty typical. They'll use frenzy, and reckless attacks for this test. Great sword. Up against an owl bear, they do 35.83 damage per round, while taking 11.31 DPR. 5 rounds of survivability, but they kill the owl bear in 2.

Level 5 champion fighter has similar stats but let's say they got chain+shield and dueling, long sword. AC 20, 49 HP. They'll do 15.75 DPR (doubling that for one round if they action surge) while taking 9.8 DPR. They also have 5 rounds of survivability, but will take twice as long to kill the owl bear.

My point is that you have to take offence into account when assessing tanking, because killing enemies faster means taking less damage. Barbarians can tank exceptionally well while putting out big numbers - that berserker barbarian basically threw defence to the wind and still had the survivability of a fairly typical tanky fighter. You can definitely create builds with other classes that are as good or better at defence - eldritch knight was a great example, but only by paying a high price in offence.

And you can build an extremely tanky barbarian, if you want - that barbarian could have AC 20 and not do reckless attacks, for example, and their survivability would be ridiculous (owl bear now needs 11 rounds to kill them, and the barbarian is still doing 26.3 DPR). Folks seldom do this because the baseline barbarian is already a good tank while putting up big numbers, which is fun.

What a terrible comparison. Not only are you assuming Rage - something you only have 3 uses of per day, but you also are assuming you are frenzying something you are doing even fewer times per day without significant penalties. Then the match that up against one of the worst Fighter Subclasses out there?

Heck, keep the rage and let's redo the comparison.
Barbarian now does 24.375 DPR. Will take 3 rounds to kill Owlbear. He will take 33.93 Damage. Leaving him with 21 HP.
Fighter will also kill the Owlbear in 3 rounds - if you are assuming rage then assuming action surge seems fair. Fighter will be left with 19.6 hp. However Fighter can also 2nd wind to get up to 30 hp.

So unless you are assuming a very specific Barbarian subclass that is also using what's for all intents a very limited use ability then the Barbarian isn't typically outperforming the fighter on his own.

Note also, rage must be activated - if the owl bear wins initiative the Barbarian takes 17 Damage that round. If the barbarian ever faces anything that can disable him, even for a turn, that's rage down.

If the barbarian ever faces anything with non-BPS attack component (fairly common as you level) he also starts taking even more damage by comparison.

Not to mention some nasty effects occur on a hit (or only force the saving throw chance to avoid if hit by the attack).

I'm a huge fan of barbarians and rage in general, but a high AC tends to be a better defense.

 

[FrogReaver]

I'd take 14 con and 14 dex over that and just use scale mail for an AC of 16.

Better ini, 2 more AC and better ranged capabilities. Is it better? I did not do the math, so I don't know. But I probably would like that more.

It is basically trading 5hp (10% less hp) for 10% lower chance to be hit (which are a about 14% of the attack rolls from the owlbear that would have hit.)

The 16 con barbarian has EHP of 60.5, the Dex 14 barbarian has EHP of 59.5.

That is very close. And only unfavourable for the Dex barb, because the Owl Bear has a very high to hit bonus.
If you account the higher initiative and the 1 in 20 chance to have an extra round of attacks might make up for the difference.

Personally I think sword and shield barbarians focusing on AC are highly underrated. Change up the Barbarians stats to 14 dex and 14 con, give the barbarian good medium armor and a shield. Now we are looking at something like 50 HP, 19 AC, 15.75 DPR (no reckless) and will take 5.5 DPR.

He will kill the owlbear in 4 rounds. He will take 22 Hp in Damage, leaving him 28 hp.

 

[Hussar]

The point isn’t comparing barbarians to other classes is it? The point is how much difference does 1hp/level actually make.

 

[FrogReaver]

The point isn’t comparing barbarians to other classes is it? The point is how much difference does 1hp/level actually make.

Right. IMO. It's not just 1 hp per level though. It's +1 con mod (con saves can prevent nasty effects and high damage). It's also +1 hp healed per hit dice. So to give a direct answer, +2 Con doesn't make much of a difference but more than +1hp/level. But it's also important to point out, +2 Str doesn't make much of a difference either. Both are small potatoes in the grand scheme of things.

If I'm in the party as the only melee PC then I'm taking about as much buffs to my survivability as I can. It's the single most important thing for me at that point. I would even take +2 Con over +2 Str for my ASI.

If there's another melee pc or 2 then I'm thinking much harder about trading some of that durability for damage.

 

[FrogReaver]

Anyways, back to the data. I'm trying to decide how best to remove/compensate for characters that may be based solely on the SRD.

Initial thought was to compare without life cleric, champion fighter, thief rogue and evoker wizard but that doesn't seem great either.

 

[darjr]

Anyways, back to the data. I'm trying to decide how best to remove/compensate for characters that may be based solely on the SRD.

Initial thought was to compare without life cleric, champion fighter, thief rogue and evoker wizard but that doesn't seem great either.

I dunno if this helps but you could find characters that absolutely are not based in if they have a feature outside of it. At least narrow down the potentials.

 

[FrogReaver]

I dunno if this helps but you could find characters that absolutely are not based in if they have a feature outside of it. At least narrow down the potentials.

I think if I look at just characters with a feat but without grappler as their only feat and take that population it would be interesting.