D&D 5e has failed

[Ignimortis]

I would like to correct my previous post. There are, in fact, a few lines about spellcasting and resistances/immunities. The only actual numbers given are "Resistance improves EHP by this much depending on CR, Immunity by this much". There are no rules for multiple resistances, and spellcasting is "to be considered during assigning a CR". So basically you bullshit your way through making anything that's not a frontline bruiser tank-and-spank mob.

[Username17]

Probably the weirdest thing about 5e's Challenge Rating non-system is that when you encounter more than one enemy the total XP value gets multiplied. So a CR 4 creature is worth 11 CR 1/2 creatures, but if you actually faced the CR 1/2 creatures 2 at a time the conversion would be a bit less than 8:1, and so on. This means that the one fucking thing that CR might actually be useful for - meaningfully comparing large numbers of low level bullshit to one medium level bruiser - just isn't possible because large numbers just can't exist. 15 monsters has a x4 XP modifier, so even 15 CR 1 creatures is using up the XP budget of a CR 15 monster.

Now blah blah bounded accuracy blah - it is true that medium sized groups of dudes are more dangerous than they have any right to be in 5e. But it's also factually true that 15 CR 1 monsters is not going to accomplish dick diddly against an Adult Green Dragon. I mean, for fuck's sake. And more pressingly, that's pretty much the end of the challenge curve and we still haven't gotten to groups of enemies too numerous to fit into a fucking van. Fuck.

-Username17

[Darkholme]

So what would you say a better creature design system would look like?

I feel like (were I designing such a system from scratch) it would center on coming up with a way of measuring everything in terms of HP and Damage Per Round, possibly with a Kills Per Round & Rounds Per Kill (as relative numbers that compare vs an average HP of an X level PC) worked in.

Maybe track average performance per round, max performance per round, and minimum performance per round on a hit? I feel like the best approach would be a bunch of formula-generated numbers, ideally calculated by a webapp or something similar.

You could (through) math, figure out the effective damage of a sleep spell. You could figure out the effective damage of a healing spell (by comparing it to the damage lost by not attacking, and by the damage you're not preventing).

It seems to me like you should be able to boil monster creation down to a couple numbers that measure threat and endurance to one creature or multiple creatures - albeit with math as a requirement.

[Jason]

So what would you say a better creature design system would look like?

I feel like (were I designing such a system from scratch) it would center on coming up with a way of measuring everything in terms of HP and Damage Per Round, possibly with a Kills Per Round & Rounds Per Kill (as relative numbers that compare vs an average HP of an X level PC) worked in.

This. In the end, even stuns and dominations can be seen as damage increase or damage mitigation. The only real isse would be "save or die" effects. Then again, those should be avoided, regardless.

Doing all of that requires a very form grasp of the system's probabilities, though.

[Darkholme]

This. In the end, even stuns and dominations can be seen as damage increase or damage mitigation. The only real issue would be "save or die" effects. Then again, those should be avoided, regardless.

I dunno - I don't mind SoS effects, especially if they take a few failed saves with increasing bad stuff before you're totally screwed. I also don't mind things like stacking debuffs (a better implementation of ability damage, IMO).

Doing all of that requires a very firm grasp of the system's probabilities, though.

Absolutely. But the official developers tend to have that. Even at Paizo, where design is always super haphazard and sloppy, they have a massive spreadsheet/database with the monsters.

But really, you just need to understand the number scaling of the player characters, and then you can plot that out on some graphs, and design your enemies against that progression, no?

Like, to me this isn't rocket science, it's data processing. You need to have some sample characters of each level, in a spreadsheet, preferably a few of each level, the more classes the better, then you tell your spreadsheet to spit out your progression curves as a couple of well designed graphs.

But I mean - it's hardly a novel idea. BadAxe Games did that very thing to 3.5 with Trailblazer back in like 2008, and it made a hell of a lot of sense. They didn't layer on the "convert everything to be damage or damage reduction" step for balancing purposes, but that's what I would think you'd want to put on top for good design, you know?

But then, I've come to a bunch of game designy conclusions that run counter to what everyone in the industry seems to be doing (and I haven't published any games), so maybe I'm just wrong.

---

Tangentially random thought: The bonuses in 5e are super tiny compared to the RNG for the first half of the campaign, which I have not been a fan of for skills - particularly.

I've given a go to Ability + Proficiency(+5 flat) + Expertise (1/4 Lv Round Up) + 1/4 Lv Round Up), magic items grant partial or full proficiency or expertise. It's similarly bounded but with a higher lower bound for the proficient characters. It does require some tweaking, particularly of defenses, and enemy offense numbers at lower levels, but I would say I like it better than the baseline math.

But what about going GURPS-like? You could change the randomizer to 3d6[1-18, average 10.5], or to (1d6+2d8-2)[1-20 average 10.5], both weighted significantly closer to the mean, which would make bonuses matter more.

---

Also: Hi! I've lurked here off and on for like a decade, but decided to make an account and post about stuff.

[Jason]

I also don't mind things like stacking debuffs (a better implementation of ability damage, IMO).

That's easy to translate into damage and defense equivalents. They effect hit probability.

But really, you just need to understand the number scaling of the player characters, and then you can plot that out on some graphs, and design your enemies against that progression, no?

Assuming you have a properly balanced and linear progression, yes.

[Username17]

Making an internal system for balancing bruiser monsters is the easiest thing in the fucking world. You have some reference attack and defense ratings for different levels, and you average out rounds per win and rounds per loss and ask that the two be within a certain amount of each other to qualify for being a particular level. Having a system that's usable by the public is actually real hard - because asking people to multiply average damage values by hit rates to generate numbers that are compared to other derived numbers is too much math for the typical player. Many of your players are 12 years old - or 45 and just don't remember how to solve multi-stage algebra problems. But your fucking designers should be able to do this. If someone can't solve a DPS problem they have no business being paid money to design monsters in D&D.

Which means that the presented monsters and deign systems in 5e are exactly wrong. The Monster Manual should be able to deliver CR 4 monsters with all kinds of weird attack bonuses. A monster with +7 to-hit and a monster that does twice as much damage but only has +2 to-hit are interchangeable if your reference target has an AC of 18. The monsters made by the system for internal use should be able to have numbers all over the place, because if you actually do the algebra it's very easy to increase one term and decrease another. Stupid reductionist shit like "CR 4 Monsters have +5 to-hit" is something you should only see on the "quick and half-assed make-a-monster system" that is for external use.

-Username17

[OgreBattle]

Would just making a simpler to understand, simpler to scale system for D&D be a solution? I feel a lot of this comes from legacy instead of what is necessarily convenient for a tabletop RPG you pick up and play with buds.

[Username17]

Would just making a simpler to understand, simpler to scale system for D&D be a solution? I feel a lot of this comes from legacy instead of what is necessarily convenient for a tabletop RPG you pick up and play with buds.

No. We actually know what the simplest system is that works. You have a roll to decide whether you hit, and then you have a roll to determine how much damage is done. It doesn't get any simpler than that. But inside such a system, creating 'balanced' enemies requires you to calculate hit rates and damage per hit and compare those outputs in terms of rounds to victory and compare to rounds to defeat.

In essence you are solving a math equation that has four variables, four constants, and fractional coefficients. Most people cannot do that on the fly. Critical hits, secondary attacks, rider effects and so on can make things more complicated, but the bottom line is that most game players cannot solve for X in the simplest possible scenario.

[Grek]

I would submit that we actually do have a resolution system that's even simpler than that, AND allows balanced, on the fly monster generation: Mutants and Masterminds.

In order to create a balanced opponent, you need only make sure Effect + Accuracy, Toughness + Evasion and Willpower + Fortitude each sum to 2 x Desired PL. In order to make a coinflip* encounter, make there be equal numbers of Heroic NPCs to Heroic PCs. Five to ten PL-2 minions count for one Heroic NPC, depending on how much the party has invested in minion-clearing effects. A single PL+5 enemy is a doable encounter for a normal sized party.

*Coinflip is the expected base encounter for M&M, because there's a supheroic genre conceit where Defeat =/= Death. For D&D, you'd want to adopt a similar conceit or make the average number/strength of PC-equivalent monsters per fight go down. That doesn't change the resolution system though, it just means you have to tweak the guidelines.

[Tannhäuser]

I love M&M, but it's not different from D&D, in that you still roll to hit, roll for damage, and note down effects. It is in fact usually more of a pain in the ass than D&D is because you have to look up unique charts for degree of severity of damage instead of just ticking off hit points or applying a fixed effect.

[virgil]

I love M&M, but it's not different from D&D, in that you still roll to hit, roll for damage, and note down effects.

This level of reductionism is too far to be useful, IMO.

A really simple resolution system would be something like the Amber RPG, where both sides point out any conflating environment/tactical factors (possibly a relatively fixed list to make things less MTP), and then the higher fixed number plus modifiers wins.

[Username17]

I would submit that we actually do have a resolution system that's even simpler than that, AND allows balanced, on the fly monster generation: Mutants and Masterminds.

In order to create a balanced opponent, you need only make sure Effect + Accuracy, Toughness + Evasion and Willpower + Fortitude each sum to 2 x Desired PL.

This is totally wrong. This is an example of a "half assed approximation that can be used by most players because it doesn't use complex math." Its not actually that good at creating balanced opponents. Damage and Accuracy are still multiplied together, not added together. So getting +1 to-hit when you hit on a 16+ is worth 20% increase in damage, getting the same bonus when you hit on an 11+ is worth a 10% increase in damage, and getting the same bonus when you hit on a 6+ is worth a 7.5% increase in damage. It's non-linear, and linear tradeoffs are always inexact approximations. Mathematically, they cannot be anything else.

-Username17

[Grek]

M&M also uses 1d20 for damage. Every +1 on the attack roll is a -1 on the damage roll. Every +1 on an 11+ roll implies a -1 on a different 11+ roll. Your peak effectiveness is therefore against whichever opponent has a defense tradeoff that matches your offense tradeoff - the 10/10 guy wants to fight against the 8/8 defense enemy, while the 15/5 guy would rather hit the 12/4 enemy. It's balanced because you don't know whether the enemy is going to be 8/8, 7/9, 12/4 or 4/12 in any given encounter - when any of those are equal possibilities for enemy defenses, any analogous offensive tradeoff is equally valid.

[Username17]

M&M also uses 1d20 for damage. Every +1 on the attack roll is a -1 on the damage roll. Every +1 on an 11+ roll implies a -1 on a different 11+ roll. Your peak effectiveness is therefore against whichever opponent has a defense tradeoff that matches your offense tradeoff - the 10/10 guy wants to fight against the 8/8 defense enemy, while the 15/5 guy would rather hit the 12/4 enemy. It's balanced because you don't know whether the enemy is going to be 8/8, 7/9, 12/4 or 4/12 in any given encounter - when any of those are equal possibilities for enemy defenses, any analogous offensive tradeoff is equally valid.

No. The simple fact that the Toughness save has a degree of failure and the to-hit is binary means that isn't true.

If it was two binary rolls, then you'd be right (and the game would be unbelievably shit), but the fact that one guy hits a quarter of the time but knocks the opponent out half the time when they do and the other guy hits almost all the time but never knocks their opponent out means that the 5/15 and 15/5 guy aren't actually balanced.

-Username17

[Grek]

High Accuracy guy is applying Toughness penalties, which makes it harder to say in any given case whether Effect or Accuracy is really coming out ahead. But that doesn't actually matter, because it just applies a weight in favour of Effect over Accuracy (or possibly the other way around) and it doesn't make you want to go 100% Effect (or 100% Accuracy) every single time. Eventually the marginal utility for more effect goes negative and you find the local maxima. No matter where the peak of that utility curve ends up being for a given matchup, it'll be different for an opponent with a different defense tradeoff in a different matchup. That's why you can just say "Give it PL appropriate numbers and stuff will work out basically fine." Picking the party's tradeoffs a matching game where you get a number of guesses equal to the number of PCs, and the prize is ending the fight one turn sooner instead of one turn later.

[Username17]

High Accuracy guy is applying Toughness penalties, which makes it harder to say in any given case whether Effect or Accuracy is really coming out ahead. But that doesn't actually matter, because it just applies a weight in favour of Effect over Accuracy (or possibly the other way around) and it doesn't make you want to go 100% Effect (or 100% Accuracy) every single time. Eventually the marginal utility for more effect goes negative and you find the local maxima. No matter where the peak of that utility curve ends up being for a given matchup, it'll be different for an opponent with a different defense tradeoff in a different matchup. That's why you can just say "Give it PL appropriate numbers and stuff will work out basically fine." Picking the party's tradeoffs a matching game where you get a number of guesses equal to the number of PCs, and the prize is ending the fight one turn sooner instead of one turn later.

I'm on record as not liking Mutants and Masterminds very much, and the painful simplicity and lack of meaningful choices in combat is a big reason why. But we're already at the point where actually calculating what numbers mean in terms of actual average turns to victory is something that is way beyond what most players are capable of. Like, you haven't even done the math and you're talking this system up.

Yes, it's possible to give a list of pre-calced attack and damage numbers for each level and tell people to pick one. That's actually fine. It works basically well enough for bruiser monsters in most cases and it's simple enough for players to use at the table without having to do multi-variable algebra. But that's not the game designer's toolbox, the game designer is supposed to actually math hammer these fucking things.

-Username17

[Darkholme]

I'm on record as not liking Mutants and Masterminds very much, and the painful simplicity and lack of meaningful choices in combat is a big reason why. But we're already at the point where actually calculating what numbers mean in terms of actual average turns to victory is something that is way beyond what most players are capable of. Like, you haven't even done the math and you're talking this system up.
-Username17

After 6 months of M&M I came to that same conclusion and gave my books away to a friend in exchange for a promise that he would never ask me to play the system again. I was super hyped about it in Character Creation, but I hadn't realized just how flat & bland the gameplay would be.

In essence you are solving a math equation that has four variables, four constants, and fractional coefficients. Most people cannot do that on the fly. Critical hits, secondary attacks, rider effects and so on can make things more complicated, but the bottom line is that most game players cannot solve for X in the simplest possible scenario.
-Username17

Very true - but, you don't normally need to design monsters on the fly, in your head. You tend to do it in advance, and a digital tool that does this kind of math wouldn't be terribly complicated to make, and once you have one, using the thing would become pretty quick.

To fully idiot proof it you'd probably want a webapp (and you could even simplify it down to sliders with a bunch of toggles), but you could get a quick and dirty version that an idiot could break by erasing calculations if you made it a spreadsheet, say, in google sheets.

Yes, it's possible to give a list of pre-calced attack and damage numbers for each level and tell people to pick one. That's actually fine. It works basically well enough for bruiser monsters in most cases and it's simple enough for players to use at the table without having to do multi-variable algebra. But that's not the game designer's toolbox, the game designer is supposed to actually math hammer these fucking things.
-Username17

That kind of reference table setup is pretty reasonable for the newbie DM who is designing a monster on a napkin an hour before the session, but yeah - not what you want out of professionally designed monsters. In my set of 5e products I've bought (PDF and otherwise) I completely skipped the MM, because I just don't think their monsters are well designed or interesting - They're almost all rather samey boring HP sponges.

[CapnTthePirateG]

HP numbers in 5e are batfuck insane.

A CR 1/8 monster is allowed to have over 30 hp per the DMG table. This would actually be forgivable if monsters did anything interesting or had any cool tricks, but they don't - and once you do this monster damage output needs to go way down. It's like they wanted to make a final fantasy game, but didn't actually understand how the health/damage asymmetry worked, or that eventually you'd be cheesing infinite spell slots and you can do shit like slap reraise on the entire party to deal with massive damage. Alternatively, incompetence, whatever. Amusingly, this leads to the scenario - despite Mearls heartfelt article about how grease, was like, bad for the game you guys - where the crowd control casters once again dominate (literally!) because hypnotic pattern wins regardless of how many hundreds of hp the CR 1 monster has, whereas evocation magic is really only great in the monsters' hands to devastate cloth wearing PCs. It's something the 3.5 crowd used to scream about, then 4e promised to end it (and didn't) and we're back to the part where writing fireball on your character sheet makes you a big sucker.

You pretty much have to write your own monsters at this point. DM Empowerment!

[Lago PARANOIA]

5E D&D's big innovation was to make straight hit-point damage without the rigmarole of save-or-dies a viable solution for parties.

This also applies to casters, too. If you really want to just straight-up do cynical hit point damage, you can if you put your heart into it. Everyone knows about the Sorlocks, but I'm playing an Evoker for Adventurer's League right now and I have a Ring of Spell Storing that I pay to put Find Greater Steed in it along with a Wand of Fireballs. Melf's Minute Meteors on a Paladin steed lets me do about 16d6 + 10 each round in an AoE. That damage will go up and by quite a bit once I grab Simulacrum and Overchannel. On a 'fuck everything, it's time to throw down' round, Hex + 5th-level Scorching Ray + Hexblade's Curse + Overchannel + Simulacrum will be an easy 2 * (6d6 + 107) damage round, with possible misses of course.

It's rather sad to note the number of high-level monsters in the 5E D&D who have a ton of anti-caster defenses but have no real answer to, say, an Eldritch Knight popping an Oathbow use, getting their Action Surge and Sharpshooter on, and then killing off the ancient white dragon/Astral Dreadnaught in two rounds. By themselves.

The end result of this is that high-level 'balanced' parties with only one or two crowd control caster tends to be weaker than a party that specs entirely for vanilla damage or entirely for puncturing saves.

[Ignimortis]

It's rather sad to note the number of high-level monsters in the 5E D&D who have a ton of anti-caster defenses but have no real answer to, say, an Eldritch Knight popping an Oathbow use, getting their Action Surge and Sharpshooter on, and then killing off the ancient white dragon/Astral Dreadnaught in two rounds. By themselves.

The end result of this is that high-level 'balanced' parties with only one or two crowd control caster tends to be weaker than a party that specs entirely for vanilla damage or entirely for puncturing saves.

That happens because the designers of 5e are dumb. They intentionally kneecapped every martial sans monk (monk has a working left leg) in terms of actual narrative power and replaced that with "hey, you do more damage than anyone if you've got a magic weapon!".

Martials are there to kill shit, casters are there to kill shit slightly less and do everything else times infinity better, because spells are still king and you either give "spells" to martials or rework the damn system from the ground up so that spells aren't the end-all, be-all for narrative impact.

[Darkholme]

I lean towards 'Give everyone magic' personally. *shrug*. The badass supernatural powers are why I'm playing this system. If you nerf it to be like Savage Worlds magic, I may as well just play Savage Worlds.

[Username17]

5E D&D's big innovation was to make straight hit-point damage without the rigmarole of save-or-dies a viable solution for parties.

How could you possibly list that as an innovation for 5th edition when it was the defining trait of 4th edition?

-Username17

[nockermensch]

5E D&D's big innovation was to make straight hit-point damage without the rigmarole of save-or-dies a viable solution for parties.

How could you possibly list that as an innovation for 5th edition when it was the defining trait of 4th edition?

-Username17

*1st

[CapnTthePirateG]

5E D&D's big innovation was to make straight hit-point damage without the rigmarole of save-or-dies a viable solution for parties.

This also applies to casters, too. If you really want to just straight-up do cynical hit point damage, you can if you put your heart into it. Everyone knows about the Sorlocks, but I'm playing an Evoker for Adventurer's League right now and I have a Ring of Spell Storing that I pay to put Find Greater Steed in it along with a Wand of Fireballs. Melf's Minute Meteors on a Paladin steed lets me do about 16d6 + 10 each round in an AoE. That damage will go up and by quite a bit once I grab Simulacrum and Overchannel. On a 'fuck everything, it's time to throw down' round, Hex + 5th-level Scorching Ray + Hexblade's Curse + Overchannel + Simulacrum will be an easy 2 * (6d6 + 107) damage round, with possible misses of course.

It's rather sad to note the number of high-level monsters in the 5E D&D who have a ton of anti-caster defenses but have no real answer to, say, an Eldritch Knight popping an Oathbow use, getting their Action Surge and Sharpshooter on, and then killing off the ancient white dragon/Astral Dreadnaught in two rounds. By themselves.

The end result of this is that high-level 'balanced' parties with only one or two crowd control caster tends to be weaker than a party that specs entirely for vanilla damage or entirely for puncturing saves.

Wait, hold the fuck up. Sure, you can combine splatbooks and magic items to create The Ultimate Nuker, but you can do that in 3e by being an incantatrix with a metamagic rod. 5e makes that worse by making sure you can't get magic items you need AND making multiclassing (I see those warlock class features in your wizard) optional.