D&D 5E (2024) Mike Mearls explains why your boss monsters die too easily

[EzekielRaiden]

D&Ds never really done what youre advocating because its basically impossible to do on a realistic budget.

Except they literally haven't done what I'm talking about.

They literally don't do actual, statistical testing. If they had, the "ghoul surprise" could not happen, because they would have already known, from doing simulations, that saving throws fail far too often and need to be bolstered.

This is why I say we need ACTUAL testing, not the garbage they keep floating. 4e came closest to doing it--but didn't quite make it.

Nobody, not WotC, not TSR, not Paizo, nobody actually does TESTING. They do vague touchy-feely surveys.

A bunch of this stuff really isn't hard to do. You just need to actually do testing, which means putting things through rigorous situations. It means actually DOING "white room theory" because you are literally in the white room, the design place before the rubber hits the road.

 

[GreyLord]

I fear that many designers are too afraid to kill sacred cows, forge new paradigms, and upset vocal subcommunities to create strong design goals and stick to them when things they don't like don't neatly fit in them

It's less then not playtesting right and more being unwilling to use what the testing acknowledges because it suggests something different.

It's only when the status quo is so messed up, like with Monks, do designers feel the need and mandate to do something different.

I mean...isn't one of the problems obvious....

Spellcasters have too many base spells slots and too few regained spells slots on "short rest".

Last time they tried to kill a few Sacred Cows (4e) it created such a great outcry which probably makes them want to shy away from trying anything similar to it, or even remotely resembling that attempt ever again.

 

[GreyLord]

Except they literally haven't done what I'm talking about.

They literally don't do actual, statistical testing. If they had, the "ghoul surprise" could not happen, because they would have already known, from doing simulations, that saving throws fail far too often and need to be bolstered.

This is why I say we need ACTUAL testing, not the garbage they keep floating. 4e came closest to doing it--but didn't quite make it.

Nobody, not WotC, not TSR, not Paizo, nobody actually does TESTING. They do vague touchy-feely surveys.

A bunch of this stuff really isn't hard to do. You just need to actually do testing, which means putting things through rigorous situations. It means actually DOING "white room theory" because you are literally in the white room, the design place before the rubber hits the road.

The problem is bounded accuracy.

What do I mean by that?

I mean they wanted to have bounded accuracy. They demanded it. In some form or another, the designers were bound and determined that this was what they were going to have. It didn't matter what anyone else said, or what real feedback may say, they had to have that in their system.

In that same way, there were things that were not up for debate. They were not going to change them that drastically (they may make a small change, but major changes such as doing entirely away with their favorite ideas...that was not happening) no matter what anyone else said or what the feedback may come in as.

Look at it more this way.

Playtests today aren't really for playtesting the game. They have already playtested it to their own fashion in their private groups. The Public Playtests is to make sure it won't offend too many people. This means they can "tweak" something if people show displeasure with it, but that still isn't it's purpose.

The purpose of Playtests today, especially for WotC, is marketing. It is to publicize something and to build hype for it. It makes people feel like they are part of the process and valued (even though they really aren't), that they have a say and then to get excited about the product.

That's the purpose of the playtest today. They don't want to do actual testing as you suggest because (1. Last time they actually did something that changed things so massively brought about a pretty vile lashback and 2. it would disrupt the flow of what they are actually wanting to do. They don't want to change their system back to it's foundations and redraw it after making it already. A true playtest would be open to that, but not WotC's playtests. it's not it's purpose). It's purpose is purely for a marketing viewpoint and that does not require the vigorous testing you are suggesting...and may even be contrary to it's goals!.

 

[EzekielRaiden]

The problem is bounded accuracy.

What do I mean by that?

I mean they wanted to have bounded accuracy. They demanded it. In some form or another, the designers were bound and determined that this was what they were going to have. It didn't matter what anyone else said, or what real feedback may say, they had to have that in their system.

In that same way, there were things that were not up for debate. They were not going to change them that drastically (they may make a small change, but major changes such as doing entirely away with their favorite ideas...that was not happening) no matter what anyone else said or what the feedback may come in as.

Look at it more this way.

Playtests today aren't really for playtesting the game. They have already playtested it to their own fashion in their private groups. The Public Playtests is to make sure it won't offend too many people. This means they can "tweak" something if people show displeasure with it, but that still isn't it's purpose.

The purpose of Playtests today, especially for WotC, is marketing. It is to publicize something and to build hype for it. It makes people feel like they are part of the process and valued (even though they really aren't), that they have a say and then to get excited about the product.

That's the purpose of the playtest today. They don't want to do actual testing as you suggest because (1. Last time they actually did something that changed things so massively brought about a pretty vile lashback and 2. it would disrupt the flow of what they are actually wanting to do. They don't want to change their system back to it's foundations and redraw it after making it already. A true playtest would be open to that, but not WotC's playtests. it's not it's purpose). It's purpose is purely for a marketing viewpoint and that does not require the vigorous testing you are suggesting...and may even be contrary to it's goals!.

Folks have, very recently, told me that this is not true and that the playtest really was there to test things to see if they would work. That it wasn't performative.

So...like I completely agree with you! But there are people, even today, who adamantly insist that this isn't true and that the playtesting really is sincere, actual playtesting.

 

[Minigiant]

Except they literally haven't done what I'm talking about.

They literally don't do actual, statistical testing. If they had, the "ghoul surprise" could not happen, because they would have already known, from doing simulations, that saving throws fail far too often and need to be bolstered.

This is why I say we need ACTUAL testing, not the garbage they keep floating. 4e came closest to doing it--but didn't quite make it.

Nobody, not WotC, not TSR, not Paizo, nobody actually does TESTING. They do vague touchy-feely surveys.

A bunch of this stuff really isn't hard to do. You just need to actually do testing, which means putting things through rigorous situations. It means actually DOING "white room theory" because you are literally in the white room, the design place before the rubber hits the road.

It's less that they don't do testing is that the zeitgeist and mentality of D&D is that it is "an art and not a science" and that "tables can fix the simple stuff". So rarely do designers bludgeon the rules with the results of the surveys and testing.

So a lot of "looks correct" , "matches how I like to see it" and "close enough" and BOOM!

 

[Minigiant]

I mean they wanted to have bounded accuracy. They demanded it. In some form or another, the designers were bound and determined that this was what they were going to have. It didn't matter what anyone else said, or what real feedback may say, they had to have that in their system.

In that same way, there were things that were not up for debate. They were not going to change them that drastically (they may make a small change, but major changes such as doing entirely away with their favorite ideas...that was not happening) no matter what anyone else said or what the feedback may come in as

The bolded part. For every Major RPG.

 

[Parmandur]

I mean they wanted to have bounded accuracy. They demanded it. In some form or another, the designers were bound and determined that this was what they were going to have. It didn't matter what anyone else said, or what real feedback may say, they had to have that in their system.

I mean, praise for bounded accuracy is pretty universally praised: I would reckon the real feedback at the time was similarly positive. Even the post-5E games we are seeing now stick to it.

 

[GreyLord]

I mean, praise for bounded accuracy is pretty universally praised: I would reckon the real feedback at the time was similarly positive. Even the post-5E games we are seeing now stick to it.

Was it though?

According to whom?

Who gave out that information?

What is it's source?

If I made an Apple Pie and said everyone loved it...that's what everyone said. Does that mean everyone loved it?

On BA...

There were some complaints about it...but they were minimized. They actually had to change the numbers a little bit...but if you look at what they did...they changed the appearance of the numbers slightly, but in reality...they didn't change a thing. And that's after admitting that there were some complaints with their universally loved BA during their original public playtest.

They were not about to tear it to the ground and try something else, or give it a multi-pronged approach to try multiple options. That was never going to happen. When given the option, all they did was a small tweak that really didn't do much and call it good. And suddenly...that solved the problems?

On tweaking information and the sources...

They were trying to sell ocean front property in Iowa as well...I didn't buy it.

They then told me that since I wasn't happy with that offer they were more than happy to change it up and sell me ocean front property in Hartwhistle UK for a pretty penny. They felt that was enough of a change that I should be happy with getting ocean front property there.

PS Edit: In other words - TLDR - I don't know what the actual reactions originally were. They had a more solitary source. I suppose it depends on whether you trust marketing to tell you the truth...or that instead they will tell you what you want to hear in order to sell you something. BA may have been loved by a majority (though we know there were complaints enough to cause a tweak), but whether it was or wasn't isn't something that I've ever seen the information on.

 

[Staffan]

I don't think that's true.

One group wants it to be hard to recover to 100% if you have 6 encounters
One group wants you to have 100% power every fight but only have 2 encounters..

That's not the read I'm getting. One group wants attrition of offensive power throughout the day*. You use a fireball on those 8 orcs – great, that made that encounter easy, but now you're down one fireball. You need to carefully ration when and how you use your magic (because it's mostly spells that are affected by this). This also needs to be weighed against hp loss: if you didn't blast those orcs out of existence, how much damage would they have done? And would that damage require more resources to heal than the fireball cost? In this model, each encounter is mostly a foregone conclusion – the question is whether you're ahead or behind on expected resource use, not whether or not you'll win that fight. And later in the day, that's when encounters become threatening for real, because you've used up your reserves.

The other group wants Maximum Effort, all the time. Much like you go into each boss encounter in World of Warcraft with full hp, full mana, and with all your cooldowns ready, they want the same for each encounter in the game. You probably have fewer resources immediately available, but you only need to worry about them for this one fight. There might be some healing surge-like mechanic that limits the number of encounters you can endure, but until you hit that limit you're golden. In this model, each encounter is threatening in and of itself and you need to use your resources to defeat it. But that's fine, you'll get them back.

(A sub-group of this is the Draw Steel type, where you go into each encounter with few resources but build them up within the encounter. Maybe you'll have to make do with fire bolt for a round or two before you have enough juice to cast a fireball.)

This does not have to do with how many encounters you can handle in a day. It's about how much juice you have for each encounter. The first group has a big juice tank but has to make that tank last all day. The other maybe has more of a juice glass, but can get free refills at the counter once you're out of combat. The problem with 5e is that it's designed for the first type, but many people play it as if it's the second type and then need to rest all the time which breaks both verisimilitude (because of the 5-minute adventuring day) and the game.

* or week or whatever time unit you want to have for recovery.

 

[Parmandur]

Was it though?

According to whom?

Who gave out that information?

What is it's source?

If I made an Apple Pie and said everyone loved it...that's what everyone said. Does that mean everyone loved it?

On BA...

There were some complaints about it...but they were minimized. They actually had to change the numbers a little bit...but if you look at what they did...they changed the appearance of the numbers slightly, but in reality...they didn't change a thing. And that's after admitting that there were some complaints with their universally loved BA during their original public playtest.

They were not about to tear it to the ground and try something else, or give it a multi-pronged approach to try multiple options. That was never going to happen. When given the option, all they did was a small tweak that really didn't do much and call it good. And suddenly...that solved the problems?

On tweaking information and the sources...

They were trying to sell ocean front property in Iowa as well...I didn't buy it.

They then told me that since I wasn't happy with that offer they were more than happy to change it up and sell me ocean front property in Hartwhistle UK for a pretty penny. They felt that was enough of a change that I should be happy with getting ocean front property there.

PS Edit: In other words - TLDR - I don't know what the actual reactions originally were. They had a more solitary source. I suppose it depends on whether you trust marketing to tell you the truth...or that instead they will tell you what you want to hear in order to sell you something. BA may have been loved by a majority (though we know there were complaints enough to cause a tweak), but whether it was or wasn't isn't something that I've ever seen the information on.

I mean, bounded accuracy is for my money the best decision WotC made with D&D since they bought the game. I don't see much criticism of it by and large, and it works great in practice. So I am very willing my to believe that WotC detailed feedback was positive in the same way that discourse about it has been for the past decade. What reason is there to think otherwise...?

Bounded accuracy restored the statistical bell curve from AD&D, fixing the central mathematical mistake of 3E design, while keeping the more intuitive "high number good" aspect of 3E. What's not to love, it serves both math and casual fun.