D&D 5E (2024) A critical analysis of 2024's revised classes

[FitzTheRuke]

This was one of the major problems with the surveys. Early on people would be thoughtful, critical, and honest in their feedback. Then, WotC scraps anything that didn’t get at least decent reception right off the bat, because the designers are on a deadline, and as @mearls has shared, past experience has shown that when an idea polls poorly, revising it almost never resulted in significantly improved scores. So, they figure no point wasting time on things that don’t get at least mixed results from the jump. But then players see that ideas they thought were promising but needed work got abandoned, and that erodes their trust in the process. So, they start responding more conservatively, being less critical of ideas they think are redeemable, in hopes that will result in them getting iterated on instead of abandoned. But by then the deadlines are getting closer, so the designers are pressured to also be more conservative with their designs, sticking closer to what they know the audience already likes so they can get the approval they need and move on. This further erodes trust in the process, and responses start getting more polarized. Everything you like enough to not want to see abandoned is a 9 or 10 out of 10; everything else is a 0 or 1 out of 10.

I think the new red/yellow/green method is an improvement over the system they used for the D&D Next and One D&D playtests, because it makes it more clear whether a low to middling approval is an “I don’t want this” or an “I want this, but I want it to be better.” Because the former isn’t worth wasting time trying to fix, but throwing the latter out with the bathwater leads to overly conservative designs and players getting frustrated with the survey process.

Absolutely! You nailed it.

 

[Blue]

A lot of the fighter/wizard squabbling is about genre and aesthetics more than it is about balance. It doesn't matter how much a fighter can do, if a wizard can cast wish or teleport or whatever, then there's a certain "high magic" vibe that a certain subset of the player base just cannot abide. And "play at lower levels" is not a really great way to mitigate that, since character growth is such a strong hook for play.

It'll be fixed again in 6e when levels only go up to 10.

Bit of a late reply, but this is exactly what happened.

Before 5e came out, the lead designer of 4e (Rob Heinsoo) and a lead designer of 3ed (Jonathan Tweet) came out with a D20 RPG called 13th Age, their "love letter to D&D" and the game that they wanted to run in their weekly Wednesday night slot. 5e shares a lot of sensibilities of streamlining, but without the need to keep certain sacred cows nor executive meddling they made the game they wanted. And it only had 10 levels.

They did, however, introduce to the game the concept of Incremental Advances. so before you level you have multiple points to pick up something from your next level, Made leveling a bit more granular and not as long of a wait for players between playing with new stuff.

Side note: 13th Age 2nd edition just came out in the second half of 2025 and I do recommend it. It's also math-compatible with the original edition, so all adventures. monster books, and the like are still viable to use.

 

[Zardnaar]

This was one of the major problems with the surveys. Early on people would be thoughtful, critical, and honest in their feedback. Then, WotC scraps anything that didn’t get at least decent reception right off the bat, because the designers are on a deadline, and as @mearls has shared, past experience has shown that when an idea polls poorly, revising it almost never resulted in significantly improved scores. So, they figure no point wasting time on things that don’t get at least mixed results from the jump. But then players see that ideas they thought were promising but needed work got abandoned, and that erodes their trust in the process. So, they start responding more conservatively, being less critical of ideas they think are redeemable, in hopes that will result in them getting iterated on instead of abandoned. But by then the deadlines are getting closer, so the designers are pressured to also be more conservative with their designs, sticking closer to what they know the audience already likes so they can get the approval they need and move on. This further erodes trust in the process, and responses start getting more polarized. Everything you like enough to not want to see abandoned is a 9 or 10 out of 10; everything else is a 0 or 1 out of 10.

I think the new red/yellow/green method is an improvement over the system they used for the D&D Next and One D&D playtests, because it makes it more clear whether a low to middling approval is an “I don’t want this” or an “I want this, but I want it to be better.” Because the former isn’t worth wasting time trying to fix, but throwing the latter out with the bathwater leads to overly conservative designs and players getting frustrated with the survey process.

Love or hate them youre not seeing the experimental books 2E and 3E had.

 

[ECMO3]

But it does indicate the results cannot be trusted as representative.

"Not trusting" the results is fundamentally different than "plenty of reasons to believe a sample is not representative"

 

[Seramus]

The surveys for UA were so short I don't know how they got any meaningful playtest data back. It was useful to point out any immediate and glaring issues, but not actual play data.

 

[I'm A Banana]

Before 5e came out, the lead designer of 4e (Rob Heinsoo) and a lead designer of 3ed (Jonathan Tweet) came out with a D20 RPG called 13th Age, their "love letter to D&D" and the game that they wanted to run in their weekly Wednesday night slot. 5e shares a lot of sensibilities of streamlining, but without the need to keep certain sacred cows nor executive meddling they made the game they wanted. And it only had 10 levels.

I liked a lot of 13th Age's ideas. Some felt a little too "immersion breaking" for me, since I am pretty strong Actor Stance in my D&D, but they were cooking. Escalating power was also a good idea.

But, yeah, I think the "wizard problem" is mostly solved if you move the high-level play into an opt-in area where strict balance in monster fights and dungeon exploration is not really a major concern and we treat high level magic more like the world-shaping thing that it can be (and give high level characters who don't use magic some ability to shape the world, too).

And I'm deeply suspicious of any appeal to some kind of intractable and dedicated pro-wizard "partisans" who just always want the wizard to be better than everything else.

 

[mamba]

I have a low opinion of the surveys and don't trust their results are particularly meaningful, but that's because I filled them out and most of the time, I had no idea what I was saying. I don't know if I ever got my opinion across.

If everyone had the same problem (or worse, were sure they were saying one thing, while WotC read it as another) then I don't know how they ever got much useful info out of them.

that is my problem with them, what does giving something a 2 mean? Do I not like it and want to keep what we have now, do I like the idea but think the implementation needs work? Will WotC understand it the same way I meant it?

After a few rounds I settled on only voting min (reject it outright) or max (like the direction, nothing else matters) values, as that is the only way for WotC to not get it wrong.

After the halfway point where they threw out every single thing that was halfway interesting, I no longer bothered taking a look.

To me the survey is useless for improving the game, what it does is identify the real duds that no one likes, and that probably is all WotC cares about anyway

 

[FrogReaver]

"Not trusting" the results is fundamentally different than "plenty of reasons to believe a sample is not representative"

you are using different words here. Different classes of people respond to the survey at different rates than is in the population. Even you didnt dispute this. That makes it by definition a non-representative sample.

 

[Paul Farquhar]

you are using different words here. Different classes of people respond to the survey at different rates than is in the population. Even you didnt dispute this. That makes it by definition a non-representative sample.

Which doesn’t make it useless, the people who respond are likely to be well qualified to comment on mechanical issues, but woefully out of touch when it comes to thematic aspects.

The surveys for UA were so short I don't know how they got any meaningful playtest data back

The more time required to complete the survey, the more you select in favour of very hardcore players.

 

[FrogReaver]

Which doesn’t make it useless, the people who respond are likely to be well qualified to comment on mechanical issues, but woefully out of touch when it comes to thematic aspects.

Right, I never said it was worthless. Just that it wasn’t representative.

The more time required to complete the survey, the more you select in favour of very hardcore players.

Exactly!