User:Guy/CR

Everyone knows 5e's Challenge Rating is flawed. This stems from several problems, a few of which I rant review below.

To summarize there's too many inherent problems to truly solve everything; at least not without completely rewriting the system and changing how classes and monsters work. But there's a few partial solutions you can take to help swallow the bitter pill.

Primary Principle: Damage In, Damage Out

• Challenge Rating is based entirely on (a) how much damage is dealt/taken, and (b) chance damage is dealt/taken.
• Condition immunities as just one example have absolutely no bearing on CR, even though immunity to being paralyzed alone thwarts the most infamous boss-stomping strategy there is.
• This is the root cause of most other flaws.

Secondary Principle: Action Economy

• In other words: when are four actions not better than one? When is hit points split four ways better than a single big bag of hp? The devs don't appear to know for sure and to be fair I don't either. (Also to be fair XGtE and some official splat tried to accommodate for this after the fact.)
• Action Economy wouldn't matter if the game actually played out to the Primary Principle and there was never AoE, but that's not how the game works.

• AoE: In 5e, the only reason a dragon will win a fight against 10 mooks is because of fire breath. The bounded accuracy and other concepts of 5e mean every creature is going to make a significant difference unless it gets nuked by fireball or whatever your favorite AoE is, and there's no toggle for AoE on either side.
• Save or Suck: If three 4th-level goblins all cast hold person on the 20th-level fighter every round, then wail on him with sticks and stones whenever he's paralyzed... well, yeah.

Partial Solutions

• Replace Multiattack with actually getting several actions per turn, or let Legendary Actions actually be any action. Discussed more later.
• Make AoEs more realistically balanced and rated, especially on the player side. Fireball is the most infamous example for shattering the ceiling even the DMG suggests for 3rd-level spells: 6d6 is suggested for a 3rd-level spell like this, while 8d6 is suggested for a 5th-level spell.

Saves Should Matter

• From a pure numbers standpoint, every player-character has a higher chance of failing a random save than of being hit by an attack roll.
• This is obvious just from comparing a "low" AC of 15 to a "typical" saving throw of +1, let alone including advantage/disadvantage and flanking and whatever else.
• Yet CR assumes a +8 to hit is the same as a DC 18 save. Bonkers.

Partial Solution

Take the CR chart and lower the save DC standards while raising the attack bonus standards. Considering this is all based on proficiency bonus, a good starting point would probably be based on that. E.g., at +6 prof, split the difference to increase hit standard by 3 and decrease save standard by 3. It would mean re-rating every monster, though, and at that point we might as well just replace the CR system entirely. But I at least don't have the time or energy for that.

Stunlocking

• The most glaring issue, especially beyond early levels.
• Any single monster will rapidly be denied its actions and movement as the PCs whale on it mercilessly.
• While stunlocking with hold person or hold monster is the most infamous example, the root of this problem is generally the monster failing saves and not having the means to deal with a failed save. Even entangle, a 1st-level spell, can throw a boss fight in the trash.
• Because of the PRimary Principle, monsters who actually can resist this (with high saves, condition immunities, or special features) do not have a higher CR. Having more hit points (the main thing driving up defensive CR) has no real effect; it just means the monster might get stunlocked for 3 turns instead of 1.

Potential Solutions

• Actually give a damn about saving throws in CR calculation, considering a save is what almost always what actually cripples bosses. Right now the difference between 25% to succeed a save and 75% chance to succeed a save is maybe worth one CR increment, possibly two. Yet +4 AC (only 20% difference) is worth two. Still not feasible considering it would mean replacing CR completely.
• Replace Multiattack with actually just taking multiple actions. Multiple Dashes to overcome kiting (at a cost), multiple actions to break through entangle (at a cost), an extra Disengage to escape three PCs smashing it from every angle without wasting the entire turn, etc.
• Most creatures above about CR5 should have a "Legendary" toggle (or just call it a boss monster like a normal person) that activates whenever it is the main threat in an encounter. The toggle grants it something akin to Legendary Resistance & Legendary Actions.
• Every Legendary monster should at least have advantage on saves against being charmed, paralyzed, and stunned (or more)—if not outright immune.
• Boost the power of Legendary Resistance:

• Normally LR 3/day adds 90 hp in CR calculation. Instead of doing that, actually add 90 hp, and have using LR cost 30 hp. If you're desperate you can even increase or outright remove the daily limit. As a baseline I think it should have a number of uses equal to the monster's proficiency bonus.
• Add other forms of save resistance, such as taking damage to gain advantage on saves; have advantage on saves against [specific conditions].

• Boost the value of Legendary Actions:

• Every creature with Legendary Actions should have at least one of them afford a new saving throw to end all ongoing effects that can be ended with a save.

• Actually nerf the various spells and features which nerf the boss to death. Hold person and hold monster are hilariously overpowered for their level, yet so boring in practice (imo) that I'd throw them out wholesale, but they're just the icing on the cake. Covering everything would be too tedious for this rant.

Player Levels

Basically, levels are not even remotely equal, even though they're the core representation of player strength.

So CR is damage input and output. Yet the damage output between a 1st-level variant human fighter and a 4th-level human fighter is... maybe a difference 1 damage on average? This kind of disparity is across the board. Original rangers basically mostly advancing in damage output once they hit 5th level. The only thing really going up is hit points, which do matter, but much less than the difference between fireball and burning hands; or two attacks and one attack; or hold person and sleep. The only class which goes up somewhat consistently across every level is rogue.

As everyone knows by now the PC power spikes are generally at 5th level (the biggest), but also 11th level and 17th level. You can see this a little bit in the CR table, and is (in my theory anyway) the main reason the scale is so wonky instead of just being (X) × CR = damage. So at least the devs were aware of it. And iirc they chose power spikes as a design feature... although I disagree with that, personally.

Tangentially related, but 5e's design means two 2nd-level characters will be vastly more powerful than a single 4th-level character on the same scale; same with three 5th-levels and one 15th-level. Same applies to CRs, which is why you need to ramp up the encounter level so high with kobold club or whatever. The system could be designed this way, to be as intuitive as 1+1=2, but the devs didn't consider that enough of a priority I guess.

When it comes to players, these points are the main ones that can really be considered without decaying player freedom. You can't really stop one player being a combat-optimized build and another player being a Cha rogue. Compared to 3e and (imo) even PF2, 5e at least does a relatively good of saving players from themselves... to an extent, so at least there's that.

Potential Solutions...?

Can't really do much without rewriting (almost) every class, or fundamentally changing the structure on which almost all of 5e is built. It's an idea I play with a lot but I don't have enough for cocaine to follow through.

Encounters per day

Most modern campaigns have 0-2 encounters per day, whereas the DMG pretty clearly states 6-8 per day is the expectation.

Hindsight is truesight but did they really think 6+ combats per day was going to be the norm? I really don't see a reasonable solution to this, especially in the modern culture of 5e.

Magic items, feats, multiclassing

All of these are technically optional rules and are not factored at all into player power for CR calculation. Meanwhile I've never seen a table last long without using all of them.

• Yeah, the Sharpshooter feat which more than doubles your damage output at the cost of hitting 25% less often? Not even considered. Not even if you're a variant human who gets it at 1st level.
• The flame tongue which adds 14 or more damage output per turn to your attacks (assuming at least Extra Attack)? Not a factor.

Potential solution

• Wait for 5.5e coming out in... what was it, 2024?