Introduction
I’ve calculated CR values for a lot of monsters from official source books; nearly all of them to be precise. In the process of doing so, one thing that’s stood out to me is that WotC has been getting more precise with their monster design over time, specifically when it comes to legendary monsters. What I mean by this is that legendary monsters published in more recent books are, on paper, more likely to be close to their listed CR than monsters from earlier book.
As I’ll show in the sections that follow, legendary monsters from early books were overpowered. In most cases, this resulted in the listed CR being 1-2 levels lower than it ought to be based on the monsters’ stat blocks. This sort of difference isn’t likely to make a huge difference in actual play, unless a DM is intending to push their PCs to their absolute limits, so it’s not surprising that they snuck through and have gone largely unnoticed until now.
I’ve broken up this post into two sections. The first looks at legendary monsters from adventures modules and how they’ve changed over the years. The second looks at legendary monster from source books. Specifically, it looks at how legendary monsters from the Monster Manual compares with those published in subsequent source books.
Adventure Modules
Every adventure module WotC has published so far has legendary monsters in it that can’t be found in any other books. To get a sense of how strong these monsters are for their respective CR values, I’ve calculated their XP values following the method I discussed in my previous post Calculating Monster XP.
This approach gives better resolution than comparing listed CRs with CR values calculated using chapter 9 of the DMG. And, since the difference between CRs is not uniformly defined, it also gives a much more consistent measure of how much stronger/weaker a monster is than expected across a wide range of CRs.
Figure \figref{fig:exp-ratio-vs-time-adventure-modules} (below) shows how these calculated XP values compare to the target XP value for both normal and legendary monsters from adventure modules.
For normal monsters, while the averages tends to be a bit on the weaker side of the spectrum, there is little change in the average over time. However, for legendary monsters there is a clear trend, where monsters from adventure modules published before 2018 were considerably stronger than expected compared to those published after.
It makes sense that WotC would get better at accurately assigning CR values as the system gets older and they gain more experience designing for it. However, it is puzzling that this trend only shows up for legendary monsters and not normal monsters as well.
To understand this better, Table 1 lists all of the legendary monsters from these adventure modules published prior to 2018.
Table 1
| Book | Monsters |
|---|---|
| HotDQ | Rezmir |
| RoT | Severin, Tiamat |
| PotA | Imix, Ogremoch, Olhydra, Yan-C-Bin |
| OotA | Baphomet, Demogorgon, Fraz-Urb’luu, Graz’zt, Juiblex, Orcus, Yeenoghu, Yestabrod, Zuggtmoy |
| CoS | Strahd von Zarovich |
| SKT | Klauth, Maegera the Dawn Titan, Slarkrethel |
| TftYP | Tarul Var |
| ToA | Acererak, Atropal |
For those who have run or read through these modules, you may immediately recognize part of the issue here. Most of these monsters are not fought directly in their respective adventures. They are fought indirectly, either in a weakened state, with allies, or intended to be avoided by the PCs entirely.
In fact, by my count only four of the legendary monsters listed above are suppose to be fought directly by the PCs under normal circumstances. These include Rezmir from Hoard of the Dragon Queen, Yestabrod from Out of the Abyss, Tarul Var from Tales from the Yawning Portal, and the atropal from Tomb of Annihilation.
As shown in Fig. \figref{fig:exp-ratio-vs-encounter-type} (above), when split into monsters who are fought directly by the PCs and those who aren’t, the legendary monsters who are driving this trend are clearly the later.
This result isn’t too surprising. The legendary monsters who are fought directly in these adventure modules are much more likely to have been encountered by playtesters and given feedback on than those who were intended to be avoided. And, for those legendary monsters encountered with support from allies or in weakened state, the playtest feedback would naturally be focused more on how the encounter felt as a whole, than how each monster felt individually.
Another interesting point to note about these monsters is that nearly all of the legendary monsters from Out of the Abyss (the demon lords) were republished in Mordenkainen’s Tome of Foes with adjustments made to their stat blocks, and again in Mordendainen Present: Monsters of the Multiverse. A comparison between the XP ratios from each of these books for each demon lord is shown in Fig. \figref{fig:exp-ratio-vs-monster-oota-mtof} (below).
For all of the monsters except Juiblex, who was the most on target of the original set, the updated stat blocks are changed to bring them more in line with the expected strength of their CR. In total, the average XP ratio for these monsters dropped from around 1.40 in Out of the Abyss, with a standard deviation of 0.31, to 1.04 in Mordenkainen’s Tome of Foes, with a standard deviation of 0.19. For their third publication in Mordenkainen’s Presents: Monsters of the Multiverse, these both improved further to an average ratio of 1.03 and a standard deviation of 0.15.
Source Books
For source books, the story is a bit more subtle. As Fig. \figref{fig:exp-ratio-vs-time-source-books} (below) shows, the average XP ratio for the published source books stays fairly consistent for both normal and legendary monsters.
The Monster Manual (MM) has a slightly higher average ratio, but the difference is small compared to what the trend showed for adventure modules.
However, when comparing the distribution of XP ratio values, as shown in Fig. \figref{fig:exp-ratio-cdf-mm-vs-source-books} (below), the MM stands out as a clear outlier from the rest of the source books, with a higher than normal portion of its monsters having a XP ratio between 1.3 and 1.7.
After looking through the legendary monsters from the MM in this range, this deviation was clearly the result of dragons. This is clearly illustrated in Fig. \figref{fig:exp-ratio-cdf-mm-vs-source-books-no-dragons} (below), which shows how the distribution of XP ratio values changes for the MM when dragons are removed from it.
Indeed, the legendary dragons in the MM have an average XP ratio of around 1.7 while legendary non-dragon monsters have an average XP ratio of around 1.2. In other words, the legendary dragons in the MM are around 40% tougher, on average, than the rest of the legendary monsters in the MM. This relationship is illustrated again in Fig. \figref{fig:exp-ratio-vs-cr-mm-type-dragon}, which shows XP ratios for individual monsters along with their target CR.
I first came across this observation around the start of 2020, and my initial theory was that this was probably intentional on the part of WotC. After all, dragons are suppose to be iconic creatures in Dungeons & Dragons. They’re in the name of the game! However, WotC recently released a source book dedicated specifically to dragons, Fizban’s Treasury of Dragons (FToD), which was included in the collection of non-MM source books shown in Fig. \figref{fig:exp-ratio-cdf-mm-vs-source-books} and it does not show this same behavior.
When compared with the legendary dragons from the MM, as Fig. \figref{fig:exp-ratio-vs-cr-mm-ftod-type-dragon}, shows, the dragons published in FToD are consistently weaker, and more in line with their target values. In terms of a numeric average, the legendary dragons from the MM, again, have an average XP ratio of around 1.7 while the legendary dragons from FToD have an average XP ratio of 1.2.
Conclusion
So, there you have it. The legendary monsters put out by WotC have gotten more in line with their listed CRs for both adventure modules and source books. I think this is generally a good thing, it shows that WotC has gotten better at working within their own system since the release of 5e. And, given their “next evolution” of D&D is suppose to work with monster stat blocks from 5e that means we will likely continue to get well balanced monsters going forward.