Offensive Challenge Rating: Calculating Damage/Round

[This be Richard]

Probably a dumb question, but here goes.

When you calculate the challenge rating (and the related xp quantity for encounter budgeting) of a custom monster -- or an NPC or whatever else -- part of the calculation is based on its expected damage per round. What seems strange to me, though, is that it appears to calculate the "expected damage per round" of attacks with chances of missing as though they always hit.

I get this from their example with the Young White Dragon on page 278 of the DMG under Overall Damage Output:
"For example, a young white dragon has a multiattack routine (one bite attack and two claw attacks) that deals an average of 37 damage each round...."
Looking that attack up in the MM, that's the average damage for the multiattack routine if and only if all three attacks hit.

Now, I'd normally just figure that it's telling you to account for the average damage in a worst case (for the players) scenario where every attack hits, but this convention feels strange when I look at other sources of damage. For instance...
"[A] balor's Fire Aura deals 10 fire damage to any creature that hits the balor with a melee attack. The aura also deals 10 fire damage to all creatures within 5 feet of the balor at the start of each of the balor's turns. If you assume that one character in the party is within 5 feet of the balor at all times, hitting it with a melee weapon every round, then the balor's damage output per round increases by 20."
The aura is (almost) guaranteed to inflict 10 damage on at least one character, but it doesn't get inflated at all in the calculations on account of it not having the same chance of missing as, say, the young white dragon's multiattack routine.

The fact that damage sources with and without miss chances are added together without either of them being adjusted to account for the difference in reliability makes me feel like these calculations aren't reliable. I feel like either attacks that can miss ought to be counted as though they did less damage or attacks that can't miss should be counted as though they did more.

So my question is basically, given all of this, what should I do?

Is the difference sufficiently negligible that I should just drop it? Do other people halve the "average damage" or something from something that can miss? Boost the damage from damage that's a sure thing?
Probably I just shouldn't worry about it. But I'm a hand-wringer, and when I calculate these things, I want to do it right.

[Kryx]

appears to calculate the "expected damage per round" of attacks with chances of missing as though they always hit..
Correct. You'd have to redo the whole chart to get it otherwise. It's unfortunate, but that's how they did it.

The difference is definitely important. Even for save vs attack - saves do a lot more, especially if they still take half.

[kaoskonfety]

From my reading the "damage" portion of the CR calculation is pretty much straight up "*can* it kill you in one round/two rounds/3 rounds".

I concluded it assumes worse case scenario (everything hits on the big nova) because most fights will be fairly short (under 5 rounds) and "can it drop a PC outright/ 2 rounds/ 3 rounds" is the main relevant question in the question "does its damage output seriously impact the difficulty of the fight".

My 2.7 cents (stupid inflation)

[This be Richard]

The difference is definitely important. Even for save vs attack - saves do a lot more, especially if they still take half.
That's the answer I was afraid of.
Has anyone developed a good way to reconcile the differences?


I concluded it assumes worse case scenario (everything hits on the big nova) because most fights will be fairly short
Hm. So if I'm reading you right, you're saying that the goal is to make sure you don't throw anything at the party that could conceivably wreck it too quickly, even if that has the chance of making encounters with attacks that can miss or be saved against significantly easier than the kind that do their damage more reliably.
That's sound reasoning, though it's kind of a shame because it makes for inequality between different monsters/combatants/etc that are meant to be of the same difficulty.

[VoxRationis]

What AC would you use to calibrate your chance of attacks hitting?

[This be Richard]

What AC would you use to calibrate your chance of attacks hitting?
That's a very difficult question, especially considering that the monster's attack bonus is already -- and separately -- accounted for in the offensive challenge rating.
This is the kind of thing that makes homebrewing intimidating to me. I get into the crunch of things just enough to see complications under the hood, but not enough that I feel equipped to tackle those systems.

[Kryx]

For AC you'd have to use an average and scale that. For instance at 1 a rogue is 14, a fighter 16, a caster 15.
At higher levels it goes up.


Honestly I wouldn't worry about this. CR is not a science. It's pretty good imo, but this isn't worth fixing.

[DivisibleByZero]

That's a very difficult question, especially considering that the monster's attack bonus is already -- and separately -- accounted for in the offensive challenge rating.
This is the kind of thing that makes homebrewing intimidating to me. I get into the crunch of things just enough to see complications under the hood, but not enough that I feel equipped to tackle those systems.

But your first point is exactly why it works as is.
The amount of damage that it is typically capable of doing is a separate entity from how big its attack bonus is.
Take, for example, a fictional creature with huge damage potential and low attack bonus. The CR calculation for the damage is high, the CR calculation for actually hitting with that attack is low, and as such it will be middling, and therefore balanced.

[Kryx]

But your first point is exactly why it works as is.
The amount of damage that it is typically capable of doing is a separate entity from how big its attack bonus is.
Take, for example, a fictional creature with huge damage potential and low attack bonus. The CR calculation for the damage is high, the CR calculation for actually hitting with that attack is low, and as such it will be middling, and therefore balanced.
It doesn't work for auto damage though. That is calculated the same as attacks and saves.

Nor does it work for save for half which is a large amount.

[DivisibleByZero]

It doesn't work for auto damage though. That is calculated the same as attacks and saves.

Nor does it work for save for half which is a large amount.

I've used the guidelines to create quite a few different creatures, and I haven't run into any issues. That's all I'm saying. The guidelines they gave us actually do work, even if there *appears* to be issues with portions of it.

[VoxRationis]

Honestly I wouldn't worry about this. CR is not a science. It's pretty good imo, but this isn't worth fixing.

Speaking of which... I have a question. I've done most of my DMing in 3.5 and that's the setup I'm used to (even though CR was often miscalculated for that edition). I've noticed in this edition that the CRs are massively skewed towards the bottom, and that creatures which once had mid-level CRs (e.g., the remorhaz, or most of the tougher animals) now have low-level CRs. But a lot of them don't actually look that much easier to fight than before. Is this because I'm not well-acquainted with the capabilities of 5e creatures, or does CR mean something slightly different now?

[This be Richard]

Alright, I suppose I'll just try not to worry about it too much.

Thanks, everyone!

[Kryx]

I've used the guidelines to create quite a few different creatures, and I haven't run into any issues. That's all I'm saying. The guidelines they gave us actually do work, even if there *appears* to be issues with portions of it.
Agreed that CR works as is for the most part.




CR mean something slightly different now?
CR in 3.5 is not the same as CR in 5e.

[kaoskonfety]

Alright, I suppose I'll just try not to worry about it too much.

Thanks, everyone!

I was scratching my head too when I was looking at the chart, till I remembered all of those times the big bad walked up and pulled a few 20's out of its butt and I had to fudge numbers down to "hits" to prevent a 2 round TPK because I gave the big bad a little too much *oomfh*.

Your party will fight a very large number of foes over their careers, sooner or later they are going to have a bad day. The chart is assuming this. The DM should be assuming this. It also means that if the bigbad is doing some consistent damage with no rolls it just means the party is having a bad day (in the example you ARE fighting at least one Balor - that's pretty close to the definition of a bad day). Also note the "auto-damage" cannot critical (I hope?) much of the rest of a creatures damage can and this sort of effect is generally a small amount for the potential damage.

I can see myself doing some small adjustment if most or all of the beasties damage is "no roll, no save" style with some adhoc "a CR or 2 more" - if the Balor Fire Aura was most of its damage for instance. But this turns the creature from a standard monster into more of "an environmental hazard you overcome by doing damage to it".

The alternative - forcing the DM to calculate the monsters expected DPR vs their party (or some imaginary "typical" party) case by case only gets you a tiny bit "better" a result for a significant slice of work.

It does the job "good enough"

[Demonic Spoon]

Correct. You'd have to redo the whole chart to get it otherwise. It's unfortunate, but that's how they did it.

The difference is definitely important. Even for save vs attack - saves do a lot more, especially if they still take half.

I don't see how else you would do this. It seems like it would be annoying and difficult to use the chart if you had to manually account for the additional DPR provided by an increased attack bonus rather than just using the heuristic of "Add one to offensive CR for every +2 attack bonus above expected"

With regards to save-based abilities - I'm not at my book, but isn't the "save DC" column purposely a couple points lower than expected compared to the attack bonus - For a given CR, it's easier to save against an effect with that DC than to avoid an attack with that attack bonus.

[SharkForce]

I would suggest that a simpler solution would be to just count unavoidable damage as being worth more for CR purposes. twice as much for totally unavoidable, half again as much for difficult to avoid (save for half, damaging fire radius that you can stand outside of, etc).

[pwykersotz]

Speaking of which... I have a question. I've done most of my DMing in 3.5 and that's the setup I'm used to (even though CR was often miscalculated for that edition). I've noticed in this edition that the CRs are massively skewed towards the bottom, and that creatures which once had mid-level CRs (e.g., the remorhaz, or most of the tougher animals) now have low-level CRs. But a lot of them don't actually look that much easier to fight than before. Is this because I'm not well-acquainted with the capabilities of 5e creatures, or does CR mean something slightly different now?

It's probably that you aren't well acquainted. The numbers looked high to me at first too.

[Kryx]

I would suggest that a simpler solution would be to just count unavoidable damage as being worth more for CR purposes. twice as much for totally unavoidable, half again as much for difficult to avoid (save for half, damaging fire radius that you can stand outside of, etc).
That's the method I'd use as well.

[broodax]

Speaking of which... I have a question. I've done most of my DMing in 3.5 and that's the setup I'm used to (even though CR was often miscalculated for that edition). I've noticed in this edition that the CRs are massively skewed towards the bottom, and that creatures which once had mid-level CRs (e.g., the remorhaz, or most of the tougher animals) now have low-level CRs. But a lot of them don't actually look that much easier to fight than before. Is this because I'm not well-acquainted with the capabilities of 5e creatures, or does CR mean something slightly different now?

It does mean something different. In 3.5 it was supposed to be calibrated to take away a certain amount of resources from a party. In 5e it is a measure of how likely it is that the party loses a member or even suffers a TPK. In some cases these mean the same thing - 1 death out of 4 is technically 1/4 of your resources, and a Medium-Hard encounter should take about half of your short rest abilities - but they're different in many cases by design.