Reposted from the Design Issues thread, as now is the given time for this post:
The Inherent Problems of Challenge Rating
-Challenge Rating is built for single-monster encounters. Which stop working very quickly, simply because the single monster has an inherent action disadvantage. Four standard actions (some of which become full-attacks) will most often overwhelm one.
-Challenge Rating is built for a very sub-par party of four. The rules offer no guidance on how to adjust encounters for different party sizes. Six or seven PCs can take on different classes of encounters than parties of two or three. Also remember that Challenge Rating is built for a party of the Four Iconics, the absolutely horrible party that gave Toughness to Mialee.
-Challenge Rating, and its daughter Encounter Level, assume the party will face four encounters in a day. The rules offer no guidance on how to tweak encounters when the party will face fewer or greater numbers of them, aside from the tiny table on Page 49 of the 3.5 DMG that uses terms like "Challenging," "Very Difficult," and "Overpowering" as its guidance. Also note that often, the party has a measure of control over the number of encounters it faces in a day, as the party can often choose to rest. Also, since different campaigns place a different emphasis on number of encounters per day, this fact means that Challenge Rating is, by definition, inconsistent among campaigns.
-Inconsistent circumstances behind encounters. The rules provide no guidance about how to tailor Challenge Rating and Encounter Levels to various circumstances. An encounter that takes place behind the great double doors, where the PCs can load up all their rounds/lvl buff spells, is a wildly different level of challenge than an encounter that takes place in the middle of the night, when two-thirds of the party is sleeping.
-Inconsistency among monsters with a given Challenge Rating. One can take two monsters of equal challenge rating, and often one will be more powerful, more Challenging, than the other. A quick look at the back of the 3.5 Monster Manual shows that an Ogre Mage and a Nine-Headed Hydra are both CR 8.
-Inconsistency within the monsters themselves. For several reasons, the first of which goes hand-in-hand with Inconsistent Circumstances: A Human Cleric15 is a wildly different challenge when he's pre-buffed than when he's unbuffed.
-Challenge Ratings do not take into account monster equipment, which a monster can have, but is most often is not factored into its Challenge Rating. For example, if I give a Babau a Chain Shirt, its AC goes up by +4, with no adjustment to Challenge Rating. If I give a Glabrezu a melee weapon, it suddenly can switch out its two Claw attacks (+18/+18 - 1d6+5) for three weapon attacks (+20/+15/+10 - weapondamage+15, reduces pincer attacks to +18), significantly increasing its ability to damage PCs just by picking up a weapon of negligible GP value.
-Monsters have varying levels of optimization of their stat blocks for their Challenge Rating. That same Glabrezu from the previous example has Persuasive as one of its five feats. If I switch Persuasive out for, say, Weapon Focus (glaive), its ability to hurt PCs, its Challenge, increses, with no adjustment to Challenge Rating.
-The monster advancement rules are busted. As I've posted previously, if your DM touches the monster advancement rules, his or her monsters will be significantly more powerful than their Challenge Rating shows. For example, did you know that if a monster is given one nonassociated class level, its CR increases by 0.5, effectively nothing due to rounding, yet more importantly, the monster receives the Elite Array (base stats of 15/14/13/12/10/8 instead of 11s and 10s) for free?
My previous example post on the matter showed how quickly an advanced monster scales for its Challenge Rating by using an Advanced Otyugh, and how, for +3 CR, it would go from 6HD to 18HD, and all the crazy amounts of benefits it would receive. Page 292 of the 3.5 Monster Manual uses a 15HD Otyugh as an example of advancement, which actually makes a good example of my point. Take a look at it, but as you examine it, note that its CR only increases by +2. Yes, its attacks do go from +4/+4/-2 - 1d6/1d6/1d4 to +14/+14/+11 - 1d8+4/1d8+4/1d6+2, yes, it gains 79 hp, yes, its saves go up by +7/+2/+4, and yes, it gains two attribute bumps and three (two of which are suboptimal) feats, all in exchange for a mere +2 CR.
So in conclusion, Challenge Rating and Encounter Level are in desperate need of repair, and the problems inherent to them are present in all levels of play; high-level games simply have a way of making those problems very clear. No matter how much Pathfinder fixes the PC classes, if it doesn't fix the challenges they face, the system will still be riddled with problems.
-Matt
Yes, they can be salvaged.
The thing is, they work as a relative rating of power... meaning a CR 10 monster is a lot tougher than a CR 5 monster. That's a useful number, it is just that everything beyond that falls apart.
I think for the system to be considered complete, Pathfinder needs to enumerate the "hidden" factors, you know, the ones we keep explaining in these threads.
For example, many of us are familiar with the concept of the Elite Array, and that PCs should be using that for abilities. Newbies don't know this. We know that we should adjust for characters with above Elite abilities (basically every party I've ever SEEN) but we don't know by how much.
3.5 puts you in charge of managing these factors, but never explains them in any detail beyond "you should probably take these into account."
Items too. This has to stop.
Fixing it is a huge undertaking, but they're "rebalancing" the CRs for the new Bestiary anyway, it might be a very ideal time to go through and take the guesswork out of encounter balance.
Give us a point system for how high Abilities affect APL — and for pete's sake stop pretending that everyone knows to play the elite array. Nobody does, and then they complain that it's broken!
Give us a point system for items and buffs. Make APL based on reverse-engineering the party's final stats after items and buffs if you have to... That's the only thing that really makes sense.
Add to that a little checklist of what pertinent abilities the party has access to (such as fly, or DR -piercing weapons) and ACCOUNT for these... more later.
I agree with Toyrobots. (that sounds silly) It is time for a Unified Theory of Challenge Rating, and you are just the people to find it. Even if it winds up being a little work, if it makes matching parties to encounters more accurate, it will be worth the time I have to invest.
I agree with Toyrobots. (that sounds silly) It is time for a Unified Theory of Challenge Rating
:D
Anyway, I don't want to just sit around and gripe about it. Let's put some muscle in and figure out: how do we balance encounters?
I know this issue is close to Mattastrophic's heart, so I hope he doesn't mind the thread-jack. I look forward to his opinion on the matter.
As a GM, what do you think needs to go into a "long form" encounter balancing calculation?
Ability scores are a chief concern, because (if I understand correctly) it isn't so much the scores as which scores. A Strenght 20 on a Fifth level wizard doesn't affect things much, methinks.
I think we should have a long-form system that takes into account not just level but Class (unless you think the classes are truly balanced), gear, and ability scores. Anything else that needs inclusion?
One approach involves monsters having multiple CRs to reflect what kind of party they pose a challenge for. For example, a flying creature might have two CRs, one for a flight capable or powerful ranged party, and another for a heavy melee party. This approach would probably require an expansion of the encounter balancing rules... what we have now is nice and short, but it doesn't really work very well.
I'm beginning to think the best system will be a "challenge block" for monsters instead of a single number. The party as a whole is tallied up against these ratings based on their total attributes, including buffs, and equipment. This is not a number for the players, but for the GM, and therefore it should accurately reflect the challenge. Good decisions like a timely buff should change the PC's effective party level so the GM knows what sorts of dynamic changes will make the encounter easier or harder.
Wow, that's a tall order.
Director of Game Design
|
Although I applaud the notion here, I am concerned that this has been tried before, many times in fact, and has never succeeded. Monsters, even those of the same CR, and not created equal and work quite differently based on the number of monsters, the number of PCs, the classes of the PCs, the general environment, and a host of other wildly variable factors.
I tried recasting the system a bit before the Alphas, but abandoned it because it just started getting too complicated and hence not really worth it. Ultimately, this part of the rules needs more advice. The encounter system is far from perfect, but its entire point is to give you a baseline to work with, one that you can tailor to fit your party. This is going to mean different things based on group and play style and neither of those are something we can call "from the bench" as it were. To top it off, if we recast the CR/EL system too far afield, it will invalidate nearly every encounter in print today.. and rebalancing every encounter is not very compatible.
That said, I am always up for hearing new ideas, but I want to keep things realistic here.
Jason Bulmahn
Lead Designer
Paizo Publishing
Fair enough.
Even so, there are ways that the "advice" presentation could be improved over 3.5... DO mention things like flight, and DR, and how they can make an encounter go from impossibly hard to pathetically easy.
Classes and Equipment can get left out, but I really feel we need to quantify how "Big abilities" mess up CR. Even a simple guideline like every 5 points above the Elite Array increases the effective CL by 1. Might be more fair to spread that out over the party...
The problem is, even people who know better routinely end up with higher attributes than the game is designed for. That's the inherent flaw with a baseline, even an elite baseline: the players want to be better than the baseline, even if all it does is make paperwork for the GM.
A quick rule of thumb that the GM could do once to say "hey, all my PCs are 30 point purchase, so for this campaign they have APL +2" would make players think twice about forcing the issue.
(Okay, so maybe I'm a little bitter from personal experience when all of my players work to force the ability scores ridiculously high... but this is a common enough experience for veteran GMs, unavoidable for novice GMs, and undesirable for young GMs. Should these groups really have to work harder to build encounters? )
I feel the CR system works fairly well, as long as its short-comings are acknowledged (as some others pointed out they need to be), and advice given. While crazy examples can be found, most of the time it's a decent measure before adding all the other factors. Tune it, don't scrap it.
Comments in bold
The Inherent Problems of Challenge Rating
-The rules offer no guidance on how to adjust encounters for different party sizes.
They say to consider the APL+1 if the party is 6 or more.-An encounter that takes place behind the great double doors, where the PCs can load up all their rounds/lvl buff spells, is a wildly different level of challenge than an encounter that takes place in the middle of the night, when two-thirds of the party is sleeping.
I fully agree one encounter is much more difficult than the other. But difficulty is not actually the measuring stick of CR. The question is whether the *resources* used up is different - in one it's buffs, in the other it's probably a lot more healing afterwards.-Inconsistency among monsters with a given Challenge Rating.
... Ogre Mage and a Nine-Headed Hydra are both CR 8.
I'm sorry, which one is supposed to be the strong one? The one with 9 attacks on every round and every attack of opportunity and regrows heads, or the one that flies around invisible at will, casts like a sorcerer one level higher than its CR and hits like, well, an Ogre?-Inconsistency within the monsters themselves.
A Human Cleric15 is a wildly different challenge when he's pre-buffed than when he's unbuffed.
If the party is *planning* to catch a cleric unbuffed then they deserve the easier fight they get. Adjusting CR's down for a party's good planning (whether attacking at the right time, keeping "Fly" potions around, or DR beating arrows) is straight out wrong. The Party is expending resources *before* the fight to improve their odds.-If I give a Glabrezu a melee weapon, it suddenly can switch out its two Claw attacks (+18/+18 - 1d6+5) for three weapon attacks (+20/+15/+10 - weapondamage+15, reduces pincer attacks to +18), significantly increasing its ability to damage PCs just by picking up a weapon of negligible GP value.
Actually, vs. AC 25 you're right - he does about 30% more damage. Vs. AC 30 though he would do about 10% less damage.-The monster advancement rules are busted.
Agreed (and I love them :)