The DMG on page 274 has guidelines about damage per round for monsters of each Challenge Rating. However, most monsters deal that damage as a combination of multiple attacks (and sometimes other abilities like breath weapons).

Are there statistics of average damage per attack for published monsters by Challenge Rating, for the most damaging attack these monsters have? (just Monster Manual would be OK)

To adress the questions: for clarification

  • I currently have no specific problem I am trying to solve with this, I am just curious. I could see it being useful, for example to evaluate how much damage a feat would be that granted Uncanny Dodge would avoid.

  • I originally meant attacks with an attack roll (i.e. things that say attack, like a melee or ranged attack that has a to hit), not things like area damage spells or breath weapons. This typically would exclude damage sources that need to recharge (at least I am not aware of a normal attack roll one, that has a recharge).

  • I would include extra typed damage that is the direct result of the attack.

  • I'd exclude spells, summoning abilities and secondary non-damage effects that cause things like conditions.

  • I would consider attacks lowering the hit point maximum like normal damage attacks.

  • I'd consider legendary actions if one of them dealt a higher attack damage than the normal attack.

  • I'd ignore effects that deal damage when you hit the creature.

However, an answer that included breath weapons and other damage dealing attack forms, or for some reason made another call on these would be OK for me, als long as it explains what is or is not included.

  • 7
    \$\begingroup\$ Is there any use for this information beyond being trivia about a list of monsters? With questions like this you’ve asked before, you’ve always insisted “I think it’s useful”, but you’ve never explained how, that I can recall. \$\endgroup\$ Nov 1, 2022 at 10:17
  • 8
    \$\begingroup\$ @Cassie …but how does knowing the average change anything about your decision making as a DM? \$\endgroup\$ Nov 1, 2022 at 10:44
  • 9
    \$\begingroup\$ I didn't say it was an invalid question, and I didn't say it was opinion based. I just asked why you were looking for this information, because this type of question sets off the XY alarm. Is there something you are trying to do with this information, a problem you are trying to solve with this information, that you would be better served by asking about instead? \$\endgroup\$ Nov 1, 2022 at 12:29
  • 7
    \$\begingroup\$ Getting the "why" of the question would be a good addition, if expressible (and it's quite possible and ok for it to not be), but it is not necessary for answering the question. Doubly when something something appears to be idle curiosity, showing or explaining your own effort is useful (ie. you'll attract fewer downvotes). Specifically, what's preventing you from just counting over the source category you're aware of? \$\endgroup\$
    – Someone_Evil
    Nov 1, 2022 at 12:51
  • 4
    \$\begingroup\$ Per the Help Center: "You should only ask practical, answerable questions based on actual problems that you face." Given that "I currently have no specific problem I am trying to solve with this" this question shouldn't be here. \$\endgroup\$
    – Mołot
    Nov 1, 2022 at 20:14

1 Answer 1


Going by the SRD, average damage levels off to the 20-30 range

I'm limiting myself to the SRD primarily because I have the monster data on JSON form (source; credit to github user tkfu). That means we can do a spot of text parsing to get the damage values.

Fortunately monster actions are reasonably well templated, so we can parse those templates and ignore the rest as non-damage (which by manual inspection, they are). I've grabbed the listed number (ie. not the dice based damage number) for simplicity. For this analysis I did count actions which used attacks but caused no damage (say only grappling) as dealing 0 damage, and ignored on hit riders even if they included damage. This analysis also ignores spellcasting and other traits which causes damage. I can't guarantee there aren't a few odd edge cases which slipped through, but there are just below 600 counted cases, so I'm not gonna sift through them manually.

The result, including both the average of the average (=listed value) of attacks at that CR and the highest average (=listed value) for that CR, is plotted below (raw values in the markdown source):

plot of found average and max damage over CR

There's a couple of obvious things to point out, firstly the common thing at CR 12 which is an outlier as the only CR 12 monsters in the SRD are the Archmage and Erinyes. CRs 18 and 25 to 29 are simply empty. The CR 30 monster is the Tarrasque. The max values (especially at higher CRs) are going to be dominated by single use/Recharge actions (eg. Dragons' Breath Weapons).

For completeness, if you filter to solely attack rolls (which is a lot easier since you only need to look for the Hit keyword) you instead get:

plot of found average and max damage over CR

The spikeness would be in part due to the limited variety in the SRD, making the contrast between CR 11's Remorhaz's 50 damage bit and CR 12's spellcasters rather stark.

† There are different ways to average this, the usefulness will vary somewhat with what goals you have. This average of all attacks at the CR is (slightly) different to the average of attacks of creatures of that CR and certainly different to average of attacks made at certain threat levels/character levels. The latter requiring a lot more data or assumptions.

  • 1
    \$\begingroup\$ Thank you, this is already great! (also for the JSON link, nifty resource). Interesting that there are no monsters at all in the SRD at CR 18. In my experience from other such summaries, at CRs highter than 13 or so, there are so few entries that the results tend to be a swingy. \$\endgroup\$ Nov 1, 2022 at 17:24
  • \$\begingroup\$ @NobodytheHobgoblin CR 18 is the demilich, and it’s the only CR 18 in the MM. \$\endgroup\$ Dec 25, 2023 at 14:33

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .