D&D 5E Know Your Enemy - A mini-guide to monster stats (by raleel)

MerricB

Eternal Optimist
Supporter
Originally posted by raleel:

Know Your Enemy
A mini-guide to monster stats​
 
It is said that if you know your enemies and know yourself, you will not be imperiled in a hundred battles; if you do not know your enemies but do know yourself, you will win one and lose one; if you do not know your enemies nor yourself, you will be imperiled in every single battle. -Sun Tzu​
 
 
D&D has been about numbers for a long time. Players have been optimizing since the earliest editions. Many guides exist to help in these endeavors, and I don’t intend to tread on their ground. Instead, I will take on the other side of the equation.
 
 
It can also be said that it is important for the DM to have a clear understanding of his monsters. Not only to counter and provide suitable challenges to the optimizers, but also to be able to finely tune his combats to allow particular players to shine, to prevent TPKs, and make the story much more engaging. After all, the victory that is won after hard battle is far sweeter and will garner you more kudos and engaged players.
 
 
One of the most valuable tools in optimization is knowing your enemy. That is, knowing when to fireball and when to lightning bolt. When to holy smite and when to hold person. When to bring undead and when to bring an aberration.
 
 
This guide will attempt to do extensive analysis of the monsters, as published, so that you may better make informed decisions and avoid any unpleasantness


Originally posted by raleel:

Section 1
Defenses by category​
If your enemy is secure at all points, be prepared for him. If he is in superior strength, evade him. If your opponent is temperamental, seek to irritate him. Pretend to be weak, that he may grow arrogant. If he is taking his ease, give him no rest. If his forces are united, separate them. If sovereign and subject are in accord, put division between them. Attack him where he is unprepared, appear where you are not expected.
 
You are probably going to interact with most things in the Monster Manual in one specific way - through their defenses. Whether via hand to hand combat, spells, or social manipulation, they are there to be the antagonists to the PC’s protagonists. Thus, knowing their defenses is incredibly important.
 
Methodology
Very simple - a spreadsheet with the AC, HP, and attributes of all the monsters in the monster manual and some basic statistics run on it - an average of each. AC is not compared with anything else. In all cases, the highest AC available for the monster is used.
Additionally, an average difference between the minimum of (Wisdom, Constitution, Dexterity) and the maximum of these three attributes. These three represent the most common saving throws by far. This measurement serves as a basic gauge of how important it is to be selective in your spell selection with a particular type of monster.
In some cases, multiple variants of a monster were available. They were computed in as well, which may lead to very slight weighting towards that monster’s attribute array. However, there are over 400 monsters - I am not concerned by the very slight impact that they may have.
In some cases, the groupings (ooze, plant, etc) are very small. They will show very slight deviations from the attributes of the creatures.
 
It is extremely
important to note - I do not factor in ANY immunities, resistances, or special abilities. this is straight up defenses.
 The Data
Let us start by looking at all the monsters as an aggregated whole
 
typeavg acavg stravg dexavg conavg intavg wisavg cha
all14.2545871614.9541284412.8532110114.954128448.51605504611.7958715610.11954023
 
As you can see, the averages are much lower than one might come to expect. As a whole, the monsters are not bright, but relatively strong and tough. 
 
Lets add some meta-analysis into the equation
Typeavg minavg maxavg diffmodifier impact
all10.5229357816.392201835.8692660553
 
As mentioned above, this is the average minimum of Con/Dex/Wis, and the average maximum. The average diff is the difference between these. The modifier impact is essentially how much of a bonus you can expect to get by switching from attacking the maximum to attacking the minimum. Over the whole of all the monsters - 15% on spell defenses.
 
Lets break this down a big more - by monster type. Other posts will detail other ways to break this down
 
typeavg acavg stravg dexavg conavg intavg wisavg cha
aberration14.5263157913.7894736813.3157894714.7894736811.5263157911.2631578911.05263158
beast12.0309278412.0515463913.0824742312.360824742.32989690710.907216495.175257732
celestial17.4285714321.4285714318.5714285720.8571428617.2857142920.1428571421.28571429
construct15.312514.93751214.1255.87510.68755.5625
dragon18.3111111121.6222222211.420.2444444413.911111111316.8
elemental14.4347826114.2608695713.52173913169.78260869611.5652173910.69565217
fey14.1428571410.4285714315.428571431212.142857141313.71428571
fiend15.6944444416.1944444414.4722222216.7222222211.9444444413.1666666713.83333333
giant15.0769230821.0769230811.2307692318.230769239.46153846210.5384615410.91666667
humanoid14.0704225413.352112681312.9014084510.8309859211.7323943710.85915493
monstrosity14.6538461517.7692307712.8461538516.173076926.98076923112.59.596153846
ooze7.414.45.2161.461.2
plant10.9285714310.857142867.71428571412.642857146.1428571438.8571428574.642857143
undead14.1290322612.0322580614.0645161314.5806451610.9354838711.9032258112.38709677
 
As you can see, most monster types don't deviant very far on AC. Indeed, half of the types don't even alter more than 1 point from the population. You can expect your to hit vs AC to remain relatively stable over the course of many levels. The really large noteable exceptions here are Dragons, Oozes, Plants, and Beasts. Dragons are always tough, but there are ways to deal with them. Oozes - having a low AC is not necessarily a bad thing for them, because several cause problems with weapons.
 
Beasts are important though - particularly for druids. Beat it in mind when you are wildshaping or summoning - they tend to be low AC and going to get beat up a bit.
 
The other interesting things here are that Dragons are very weak in the Dex. Big size, lots of HP, solid Wisdom, and very tough on Con. However, fireballs and lightning bolts will take them down. If you think you are going to meet a dragon, load up on these, and make sure you pay attention to color of the skin
smile.gif
 Fey are at the other end, strong in Dex, and weak in Con. Not unexpected. Neither is the Beast weakness to will. Celestials (and to a lesser extend fiends) don't really have a weak spot. Fortunately, you probably won't be fighting many of them.
 
It is not well reflected in this average, but my own viewing of Humanoids showed they tended towards the ends of the spectrum. Perhaps I will go back and add standard deviations to this to help.

Lets take a look at the advantage you may garner
Typeavg minavg maxavg diffmodifier impact
aberration9.89473684216.157894746.2631578953
beast9.91752577314.649484544.7319587632
celestial18.1428571421.857142863.7142857142
construct8.37516.81258.43754
dragon10.8666666720.466666679.65
elemental10.8695652217.347826096.478260873
fey10.7142857116.857142866.1428571433
fiend12.3611111117.555555565.1944444443
giant919.0769230810.076923085
humanoid10.6901408514.563380283.8732394372
monstrosity11.4230769216.653846155.2307692313
ooze5.21610.85
plant6.64285714312.928571436.2857142863
undead10.6129032316.709677426.0967741943
As you can see, picking you spells well with a giant can make a 25% difference in their ability to defend, and the same with dragons. In both of those cases, their low dexterities are really hurting them.
 
Interesting that Humanoids are pretty much the same across the board. You don't need to fine tune for them as much, which is not too surprising. 


Originally posted by raleel:

Section 2
Defenses by CR
So, change in format here, because the tables are very large. So, we'll use charts.
 
Methodology
Very much the same as the previous section - it uses the same data. However, splitting it out by CR you can see growth per level and how your proficiency bonus is going to impact. DMs, you can tell where there are weak stats and where there is some excellent design space to work with.
 
Again, it is important to emphasize - no consideration was made for immunities. That might be another section
smile.gif

 
Anomolies
You should probably just ignore the big spikes in CR18. There is only one CR 18 monster in the MM - the demilich. The stats are radically skewed compared to most other things - low strength, high dex, etc.I would probably just ignore most everything above CR20. Those things are going to be special set piece battles. THe stuff below CR10 though - you'll fight that for a long long time. 
 
You will notice that Dex is very flat. In fact, it is so flat that any Dex save spells are going to get more effective as they go up, because proficiency bonus will overcome the increase in dex. Dex is very flat because the monster design space makes things bigger and more lumbering as CR increases as a whole. DMs, I'd recommend making sure to put together an adequate set of higher Dex (16+) monsters to prevent fireball inflation.
 
 
 
Last edited:

log in or register to remove this ad

Image copy.
 

Attachments

  • image_1.png
    image_1.png
    22.4 KB · Views: 2,038
  • image 2.png
    image 2.png
    28.8 KB · Views: 1,985
  • image 3.png
    image 3.png
    26.6 KB · Views: 1,907
  • image 4.png
    image 4.png
    25 KB · Views: 1,865
  • image 6.png
    image 6.png
    24.5 KB · Views: 1,871
  • image 7.png
    image 7.png
    22.7 KB · Views: 1,956
  • image 8.png
    image 8.png
    23.1 KB · Views: 1,880

Originally posted by Clutchbone:

Wow! This is amazing work, thanks!
 
So it seems that Evokers are even better than they seem, based on the flat Dex scores across most monster types/CR levels.


Originally posted by GladiusLegis:

Great work, though I have one question. On saving throws, would it have been better to take just bonuses into account, rather than the ability score proper? Because many monsters have proficiency in saves, and just plotting ability scores doesn't account for that.


Originally posted by Sorxores:

and maybe add teh legendary resistance as well, since they are very important for both party and DM strategy. I won'T go cast disintegrate against a lich as my first spell, I'll wait for the lich to burn his legendary resistance on smaller spell like fireball, hold person and contagion.


Originally posted by raleel:

GladiusLegis wrote:Great work, though I have one question. On saving throws, would it have been better to take just bonuses into account, rather than the ability score proper? Because many monsters have proficiency in saves, and just plotting ability scores doesn't account for that.
 
I could have, yes. Perhaps another table, but it is quite a bit more work for that. Maybe with immunities as well
smile.gif

 
However, most creatures don't have specific ones, and this was averaged over type and CR. You are likely to be fighting stuff that has a similar Proficiency bonus to you, so... essentially they should come close to cancelling. Stat inflation won't, though. What this does say is that as you increase in level, certain stats will go up naturally, regardless of proficiency bonuses. it will only get worse if they do have proficiency bonuses.
 
Consider dragons. Many of them have proficiency with Dex, which means that they won't be that easy to hit with a fireball. But they also have proficiency with Con, which does increase with level. Thus, Con is going to be very bad because they are at least matching your stat increase and matching your proficiency.


Originally posted by raleel:

Clutchbone wrote:Wow! This is amazing work, thanks!
 
So it seems that Evokers are even better than they seem, based on the flat Dex scores across most monster types/CR levels.
 
yes, as a whole. Proficiency ends up effectively neutrallizing each other, and you end up with your stat increase against theirs. Since you are likely going to be dropping every last point into your casting stat, up to 20, you will increase in relative effectiveness.


Originally posted by raleel:

Sorxores wrote:and maybe add teh legendary resistance as well, since they are very important for both party and DM strategy. I won'T go cast disintegrate against a lich as my first spell, I'll wait for the lich to burn his legendary resistance on smaller spell like fireball, hold person and contagion.
 
I'll probably drop that in with immunities. That's quite a chunk more data to add in
smile.gif



Originally posted by TheBigHouse:

raleel wrote: 
Sorxores wrote:and maybe add teh legendary resistance as well, since they are very important for both party and DM strategy. I won'T go cast disintegrate against a lich as my first spell, I'll wait for the lich to burn his legendary resistance on smaller spell like fireball, hold person and contagion.
 
 
I'll probably drop that in with immunities. That's quite a chunk more data to add in
smile.gif
 
Awesome work on this raleel. I would also be intersted in seeing the numbers with prof bonuses added in. It would be intersting to see if certain monster types tend to have prof bonuses which affect which defense is optimal to target.


Originally posted by Bloodlust1997:

This is great PLAYER knowledge, not very useful in game if you're playing first level characters progressing upwards, otherwise you're using Meta Gaming.


Originally posted by Tfuture:

Awesome analysis.. my only suggestion would be to perhaps remove the Terrsaque from your analysis as IMO its meant to be a ridiculously overpowered monster and would come under a statisitical outlier and i dont think realistically a party would/should encounter without an extremely sadistic DM.. like i said just my opinion, feel free to disregard.


Originally posted by raleel:

I left it in there for the sake of completeness. I would personally ignore everything over about CR17. the are so specially geared and matter for only a small part of the game.
 
re: prof bonuses. Perhaps when I get some more time. I think it does matter. I looked at several monsters with high Cons and they didn't have prof bonuses there, but had bonuses to shore up their weak stats.


Originally posted by R123456:

Any chance you would be willing to share your data file used to do the calculations and visualizations?  I'd like to pull the data into R-Studio and do some analysis.  My MM will arrive next week and having the same data you used would save me a lot of time.


Originally posted by Clutchbone:

Bloodlust1997 wrote:This is great PLAYER knowledge, not very useful in game if you're playing first level characters progressing upwards, otherwise you're using Meta Gaming.
 
Many people tailor their character concept and progression to be "coincidently" optimal down the road. "Oh gosh, I sure am lucky that my evocation focus is great for fighting the low-dex-save dragons we're encountering on this adventure [Hoard of the Dragon Queen]!"
 
Others metagame as a whole party to make sure that multiple roles/proficiencies/spell saves/etc are covered. "Wow, isn't it great that a fighter, wizard, cleric and rogue all happened to be in this exact tavern at the exact moment the cultists attacked?"
 
Also some people just like seeing the data and design behind a gaming system.
 


Originally posted by borg285:

Please open your doc for public viewing.
Please add linear regression equations to scatter plots rather than averages. While showing the standard deviation is a nice simplification, oft-times just looking at the data gives us intuition on whether or not we have enough data to safely rely on a model we use. Please reduce on how many significant digits you use. As is, it's TMI.  I'd prefer 1-2 digits after the decimal point.
Rather than showing max - min, which biases towards outliers, please use the lower and upper quartile to show the spread. Instead of the average, consider the median which should lower the spikyness of your data.
Please mention what sources you are using to feed your data.


Originally posted by 7he_professor:

R123456 wrote:Any chance you would be willing to share your data file used to do the calculations and visualizations?  I'd like to pull the data into R-Studio and do some analysis.  My MM will arrive next week and having the same data you used would save me a lot of time.
 
See my Sig for links to data files on Creatures and Spells.  They do not contain Intellectual Property of WotC (Beholders, Slaad, etc) though, so they're not 100% complete.

Feel free to add tabs to the spreadsheet with graphs, charts, pivot tables or other data as well.


Originally posted by Havenward:


Clutchbone wrote:Wow! This is amazing work, thanks!
 
So it seems that Evokers are even better than they seem, based on the flat Dex scores across most monster types/CR levels.
Clutchbone wrote:Wow! This is amazing work, thanks!
 
So it seems that Evokers are even better than they seem, based on the flat Dex scores across most monster types/CR levels.
 
I would be very careful about makeing this leap.  At higher CRs Abilities mean less for saving throws than at earlier CRs because of the influence of proficiencies.  But also important, Dex spells tend to have a lot of other mitigation aimed at them than saves.  Resistance and temp hit points for example.  While I would agree looking at these graphs gives the impression that useing a damaging spell with a dex save against a creature with a 12 dex at cr 15 may be better than using a damageing spell with a con save against a creature with a 30 con at cr 20.  However compare that with AC that rises to about 18 at cr 15.  However, you can't  make the same comparison with wisdom or charisma for example, cause the effects of those spells are different than the effects of a dex based damage spell.  Their effects are also not subject to the same mitigation factors like resistance, Hit Points, or Temp Hitpoints.  
 
What I am not saying:  That Evocation and it's dex save spells aren't good.  You may be absolutely correct.  I'm just saying its hard to make that conclusion based on Dex growth.
 

Remove ads

Top