In my post 2024 Monster Manual on a Business Card, where I tried to extrapolate WOTC monster building rules from the 2024 Monster Manual, I have a lot of chart and graphs slicing up the data in various ways, one of which had a key error.
When I tried to determine whether there was a big difference in damage output between legendary/dragon and non-legendary monsters, when I set up my graph I accidentally compared, not legendary vs non legendary, but legendary vs all monsters, including legendary. Since at high CR all monsters are legendary, at high levels I accidentally compared A vs A instead of A vs B. Not surprisingly, I found them pretty similar.
As I was poking around my charts and doublechecking formulas, I found and fixed this error. This time, I found much more interesting results.
Non-draconic legendary monsters (purple line) and dragons (green line, including both legendary and non legendary) do seem to deal a lot more damage than regular monsters. Like, a lot! In fact, that damage uptick that I identified around CR 20 seems to be entirely due to legendary monsters.
I had posited that 2024 damage could be described by two lines: one from CR 0-20, where damage increased by 7.5 per CR, and another from CR 21+, where damage increased by 12.5 per level. But doesn’t it actually look like the data can be described by two lines, 1 for normal and one for legendary/dragon?
I re-graphed. This time, I’m combining legendary monsters and non-legendary dragons into one lump, since they seem to be built the same way. And I’ve drawn two best-fit lines: one for normal monsters, one for legendary+. The normal monsters deal 6 dpr per CR, and the legendary+ deal 7.5 dpr per CR (which is the number I settled on for version 1 of my 2024 monster manual on a business card).
These best-fit lines work extraordinarily well up until that gap at CR 12 (there are no CR 12 dragons/legendaries). Then the legendary monsters “run hot” for a bit between CR 13 and 18, outperforming our best fit line.
Before making any further adjustments, let’s take a look at another graph: this one for monster hit points. Again, I’m breaking up monsters into normal and legendary. I’m adding a “best fit” line, which is the same one I used for version 1 of the business card: 15 hp per CR until CR 20, then 50 hp per CR.
If you take a look at the hp graph, you can see that the best-fit line works very well, except that it “runs cold” between CR 13 and 18 – exactly the same CR band where the damage graph runs hot – and to the same extent. The damage line’s outperformance of the legendary monster best-fit line is a mirror image of the hit point line’s underperformance of the hit point best-fit line.
It’s almost like our best-fit lines are right on target, except that, for some reason, the monsters between CR 13 and 18 have a universal skew towards having high damage and low hit points. Why could that be?
If we look at the data, we see something special about CR 13-17: that’s the range of adult dragons. Of the 19 legendary monsters in that CR range, 10 of them are dragons – and dragons all have very similar statistics and attack routines. Apart from their damage types, a few abilities, and a few numbers tweaks, they’re virtual palette swaps of each other. So this CR range is, essentially, dominated by approximately the same monster design, adjusted for CR, repeated 10 times. And adult dragons run hot. They have very powerful breath weapons for their CR, and their DPR is generally much higher compared to their hit points than is true for ancient dragons or young dragons. And the rest of CR 13-17 continues this trend, with lots of heavy hitters: revamped vampires and beholders, beefed up to silence all of us whiners who complained they were underwhelming in 2014; a real solid death knight; and so on.
And then, at CR 18, we have the demilich. From looking at the graphs above, you can see that the demilich stands out for its lack of hit points (180, on par with lots of creatures of CR 11 or 12 – but don’t worry, it has lots of resistances and immunities) and its very high damage output, standing out even over the murderer’s row of adult dragons. The demilich is designed as a glass cannon, not necessarily representative of what a stock CR 18 monster has to look like. The li’l guy has d4 hit dice and a 10 Constitution for Pete’s sake!
This result – that high-damage monsters tend to have low hit points – is not surprising, since my previous analysis showed a high inverse correlation between hit points and DPR: low-hit point monsters tend to hit hard. I think the trend we’re seeing from CR 13 to 18 is related to the specific monsters in the Monster Manual, not a general rule about the stats of monsters of that CR. Again, our biggest problem in this analysis is an extreme scarcity of data, but I hypothesize that when new CR 13-18 monsters are released, they’ll tend to have higher hit points and deal less damage than the ones we’ve seen so far.
changes to the MM on a business card
Based on our new understanding of the math, I think some changes are in order.
Hit points actually might not need a change: if my theory about glass-cannon mid-level legendaries is correct, it looks like we were right on target for hit points.
The big change is in damage output. I had a relatively low linear increase, 7.5 per CR, at low level, and a high increase, 12.5 per CR, at high level, but now I think that was an artifact of the fact that high level is all legendary monsters. I now think that monster damage can be expressed very simply, with a single algorithm and an extra modifier for legendary monsters.
“Normal” (non-dragon, non-legendary) monsters deal 6, not 7.5, damage per Challenge Rating – that’s pretty well attested in the damage graph above. That’s just 1 DPR per CR higher than it was in the 2014 MM.
Legendary (and baby dragon) monsters get a 25% damage bonus compared to normal monsters, bringing them up to the 7.5 hit points per CR that we observed on the general chart. I’m totally getting rid of the 12.5 DPR boost at high level, which I think now might have been me attempting to overfit the damage curve without taking into account legendary monsters’ low hit points at midlevel. I still might have been right the first time – new data might provide more evidence for the hypothetical big damage spike at high levels – in which case I’ll re-adjust. Right now, though, I think this understanding is the best I have.
Here’s an updated 2024 Monster Manual on a Business Card and on a single sheet:
I’ll adjust them on the main 2024 MM on a Business Card post too, since that’s probably the main place people will look for it.
This has been a cautionary tale in double checking the letter values in your google sheets formulas, but it also brought us to some interesting conclusions! I love the fact that 2024 legendaries get a damage boost, breaking away from the 2014 design philosophy that their statistics have to be in line with non-legendary monsters of the same CR. I don’t like the fact that their hit points are unchanged (and, in many cases, low) until CR 20 or so. Legendaries need a lot of hit points! If it was me, I’d give them, at minimum, 50% extra hit points – now that 2024 has jettisoned the 2014 idea that legendary CR = nonlegendary CR, it’s time to go that extra mile. But I have plenty of opportunities to design monsters the way I want: this project is strictly for figuring out how WOTC designs theirs.
