Do statistics affect final CR of a monster?


Rules Questions


For example, I take a melee base monster and make it a CR 10 creature by adding class levels, templates, etc to that base.

According to the chart 1-1 in the Bestiary, the stats for a CR 10 should be: 130hp, AC 24, High Atk 18, Avg Dmg High 45.

I then use equipment, class features, spells and potions to increase the creatures statistics to: 160hp, AC 34, High Atk 24, Avg Dmg High 90.

Is this still a CR 10 monster? Do the final stats of the creature matter for determining CR or do the base stats before you add equipment, potions, spells and class abilities count in determining the creatures CR?


I think it is the final numbers that matter.


The idea is that an added class level should add one to the CR, but this depends on the type of class. Adding a sorcerer level to an ogre would probably not change it all that much and might keep its old CR. Same with templates and such. Giving a creature significantly more equipment than it should have for its modified CR adds to the CR as well.

Then, and only then, do you get to the test you mention. Does the creature's final data match up to where those data should be for its CR?


Your numbers are kind of all over the place.

CR Breakdown By Stat

HP - 12
AC - 19
At - 15
Dg - 17

Averaging these is 15.75. Only because of the very low HP compared to a CR 16, I would actually lower it a CR 14. However, the other stats are much too high for a CR 10 or even a 12, if you were to simply judge by HP. I would suggest increasing the HP to make a solid encounter in the CR 14 arena, so probably an extra 40 or so. What are its saves like?


its a discussion we are having with our current DM. Does it say any where in the rules that the final numbers are what counts and not the base CR?


CR is more of an art than a science, but you are explicitly told "compare your monster to the chart/other monsters of that CR and see if it fits" in at least one section that I know of.


Pathfinder Lost Omens, Rulebook, Starfinder Roleplaying Game Subscriber

I don't understand how you would consume magical items on the creature and not think to add CR. Also, adding equipment to non-PC race creatures should add quite a bit of CR.


If you're looking for cut-and-dry rules on what determines CR, you're gonna be disappointed. Even the chart of values given for Monster Creation completely ignores basic things like spells and monster abilities, which can completely change the capabilities of a monster.

CR is more an art than a science; the question isn't "do these numbers match up to the arbitrary CR chart?" but "does this monster/group, as a whole, pose an appropriate threat to a standard party at the level it's used". Given that, the final stats after buffs and equipment obviously mean a lot more than a 'baseline' that never actually sees play.


CR is a guideline, a tool for the DM, not a way to show how smart he can build the monsters and wipe the floor witht eh PCs. The final stats is what matters.


Not all monsters of the same CR are equal in potency, some are much stronger when grouped with a specific other monster, some are better or worse when the party possesses or lacks certain capabilities. There is no exact formula for CR, but in a perfect world there has been a bit of play testing to make a monsters rating more accurate.


Ciaran Barnes wrote:
Not all monsters of the same CR are equal in potency, some are much stronger when grouped with a specific other monster, some are better or worse when the party possesses or lacks certain capabilities. There is no exact formula for CR, but in a perfect world there has been a bit of play testing to make a monsters rating more accurate.

CR is even stated as such to be flawed at best.

There is a definite level of accuracy to it, but it's more on the +/- scale than can normally be recorded.

When WOTC designed the CR system, it was based on an Party of 4 cookie-cutter Characters with Average stats (8-14) and the prototypical Tanker-Fighter/Healer-Cleric/Controller-Wizard/Sneaker-Rogue setup.

The problem is that no party is EVER the archetypical build.

Firstly you can easily have 2 additional members of the Party - an Archer-Ranger and a Supporter-Bard.

Secondly, the game from the get-go is designed around building your own characters the way you want them to be. They can have ridiculously-high or -low stats, they can be crazy combinations of types or characters diverging from WOTC's intended roles for them (a melee Cleric, a Ranged Barbarian, etc).

Needless to say, CR was just awful and inaccurate as the day is long.

That's not to say it was COMPLETELY useless. By comparison to the utter garbage that was "monster difficulty = HD" that pervaded the game from OD&D through 2nd Edition and led to the demise of many a Party (because one Level 15 monster could be significantly weaker than a Level 10 monster and vice-versa), the CR system was a Godsend.

The main failure with the original CR system was that WOTC didn't realize that they completely misrepresented players, and thus you still had plenty of massive CR-to-actual-difficulty discrepancies, and parties could be eaten by a monster equal to their CR (which should rightly be a fairly mundane fight for them).

Paizo's CR system is surprisingly more accurate, taking more things into account, such as expected Party offenses and defenses at given levels, etc.

And, really, there is no way whatsoever for a book to accurately determine what the ACTUAL difficulty of any given fight for any given party will be.

Classes, stats, levels, feats, HP, skills, resources, gear... ALL of that plays into how hard a fight will be vs any given monster.

---

What would be REALLY, REALLY freakin' helpful is this:

If a Statistician or Actuary worked with a programmer to create a program that would create a "Real CR": the DM would load in all the data of a Party (ALL of it, basically copying their character sheets) into the program, and then bring up Monsters from a database (which is derived from the PRD), or added by hand (in the case of NPCs with Character Levels).

The program would then analyze the two sides, and determine what the chances of success for the party would be: ideally a "Normal" fight would be around 75%, a "Challenging" (CR=APL+1) fight would be 50%, Hard (CR=APL+2) would be 33%, and an "Epic" (CR=APL+3) fight would be 25%, etc.

The program would then assign an appropriate CR based on the general chances of success, in order to determine the XP granted to the party if they succeed.


chbgraphicarts wrote:

What would be REALLY, REALLY freakin' helpful is this:

If a Statistician or Actuary worked with a programmer to create a program that would create a "Real CR": the DM would load in all the data of a Party (ALL of it, basically copying their character sheets) into the program, and then bring up Monsters from a database (which is derived from the PRD), or added by hand (in the case of NPCs with Character Levels).

The program would then analyze the two sides, and determine what the chances of success for the party would be: ideally a "Normal" fight would be around 75%, a "Challenging" (CR=APL+1) fight would be 50%, Hard (CR=APL+2) would be 33%, and an "Epic" (CR=APL+3) fight would be 25%, etc.

The program would then assign an appropriate CR based on the general chances of success, in order to determine the XP granted to the party if they succeed.

The issue with this is as soon as you move past simplistic party play styles, actually coming up with a meaningful number is incredibly difficult.

For anything with a complexity beyond participants walking up to each other and wailing until one side dies (which is at least a somewhat simple number crunch) the best way of handling it would probably be a Monte Carlo simulation in which each side is played by a set of AI.

For those of you who don't know what a Monte Carlo simulation is, it is literally running the scenario in question a large number (e.g. thousands) of times and performing a statistical analysis on the results.

Yep, this would entail needing an AI that can understand the use of any arbitrary choice of spells, feats, abilities and equipment as well as devise and execute a close-to-optimal strategy using said equipment, feats, abilities and spells. This AI needs to also be fast enough for several of them to be able to run through an encounter several thousand times within a few seconds.

If this sounds incredibly difficult to the point of being impossible, it's probably because it is.

Oh, and FYI an APL+4 encounter should be a 50% chance of death. A 4 PC party has a CR of APL+4 and anything else of that CR should be of roughly equal power (not that it actually works out that way, but this is what the CR guidelines say). A CR=APL encounter should have a close to zero chance of losing.


Tiber11 wrote:
its a discussion we are having with our current DM. Does it say any where in the rules that the final numbers are what counts and not the base CR?
Target Statistics wrote:
CR: This is the approximate CR of the monster. This number might change as design progresses.
Other Statistics wrote:
When determining a creature's attack bonuses, refer to the guidelines from Table: Monster Statistics by CR based on the creature's CR. If the bonus is too low, consider increasing the creature's Strength or Dexterity, or increasing the amount of damage it deals to above the average. If the bonus is too high, consider decreasing the creature's Strength or Dexterity, or decrease the amount of damage it deals. If this value is significantly different, and the creature is intended to rely on melee or ranged attacks, consider adjusting the creature's CR.

The bolded advice is repeated several times throughout that chapter. If your values significantly differ, the CR should change to match.

Source


Thanks Buri, I was hoping for clearer rules, but that is what we got to argue with!

We're a party of six, all level 7 and we got to fight a Minotaur with north of 34 AC, attacking with 30's and hitting 30ish dmg with 5 attacks. He did 101 points in 3 of his 5 hits on the Paladin. His saves were at least 16's.

DM says he added 3 levels of War Priest and Full Plate among other things to the build and that it was a CR 11, but he doesn't take into account equipment, actual stat bumps, spells, etc...


Snowblind wrote:
If this sounds incredibly difficult to the point of being impossible, it's probably because it is.

It's not impossible in the least. It just requires a level of computational power that's still some years out, especially for your average laptop.

Give it time, and you'll be seeing it (though 7th Edition might be out by then, while Pathfinder'll still be going strong).


Tiber11 wrote:

Thanks Buri, I was hoping for clearer rules, but that is what we got to argue with!

We're a party of six, all level 7 and we got to fight a Minotaur with north of 34 AC, attacking with 30's and hitting 30ish dmg with 5 attacks. He did 101 points in 3 of his 5 hits on the Paladin. His saves were at least 16's.

DM says he added 3 levels of War Priest and Full Plate among other things to the build and that it was a CR 11, but he doesn't take into account equipment, actual stat bumps, spells, etc...

APL+4 is already a freakishly insane fight, and your DM is either an absolute sadist or an idiot. Or both.

Paizo recommends CR=APL+5 as the End Boss, so APL+4 shouldn't be a mid-boss fight at all; it should've been more like APL+2, or APL+3 at most.

Community / Forums / Pathfinder / Pathfinder First Edition / Rules Questions / Do statistics affect final CR of a monster? All Messageboards

Want to post a reply? Sign in.