D&D (2024) Class spell lists and pact magic are back!

[Chaosmancer]

no, I am saying your vote was understood as ‘no’ by WotC.

Show me the document that states so. Show me any shred of evidence beyond "I, Mamba believe this is true"

You can't just state what other people you've never personally met totally decided in a situation you were not part of.

You do not have to try to explain to me that the survey result is not the same as your vote, I am aware of that… I was already aware the last two times too

And yet you keep making the same claims...

I would say I am very confident in it, yes. Now show me why you think it did not get counted as a ‘no’, I gave you my reasoning and so far you have shown nothing to contradict it. Just saying ‘I disagree’ won’t change my mind.

Because Wizards could look at a single vote of Dissatisfied, along with comments explaining why it was dissatisfied, and take that as a sign to follow the revisions laid out in the comment. This is perfectly logical for them to do. A single vote of dissatisfied does not a 50% rating mean.

ESPECIALLY when it needs to be taken in with context of the class rating, and the other ability ratings.

well, I have shown the logic, as far as I am concerned it is absolutely sound.

Is your only counter ‘you do not know what WotC does’? Then we can end it right here. Neither do you, and out of the two possibilities I consider mine far more likely. If you cannot even address my concerns by doing more than say ‘you cannot be certain’, then that will not change.

Also, if you make a claim, which you do, you too NEED to show proof, just like you are demanding it from me. Saying you do not need to does not change that. At most you can say you have confidence in them, or are not convinced by my arguments. But if you say that there is no issue and WotC understands the votes correctly, you have a burden of proof just like me. It is not just ‘extraordinary’ claims that require evidence, all claims do.


I made my case, repeatedly, now refute it, saying ‘WotC smart’ is not even close to that

Considering you have repeatedly shown a shocking lack of understanding of statistically significant data, and continue making claims based solely on your own opinions? Yeah, I really don't need to do much here.

Your claims include:

• WoTC being either incompetent or malicious
• Being able to read WoTC's minds on their interpretations of data you have never seen
• Knowing their secret internal data processing procedures, because obviously they are the same as what you would do
• Your own opinions on what things mean
• A general dissatisfaction with the results of the survey you claim is flawed
• Logic based entirely on "everyone else thinks the same way I do, because that is logical"

Why would anyone take that seriously when compared with a decade of research done by professionals, working for a massive corporation, that seems to be leading them to unparalleled financial success?

if a process that should always lead to >70 leads to both 80 and 20, it very much is flawed.

No. Because you are assuming a process always succeeds. Which is false to begin with. And you then are following it up by ignoring that the process was not followed in that first iteration. The Champion Fighter and Berserker Barbarian did not undergo this process before.

Also, I am not saying they should abandon the playtest altogether, I am saying they should ask better questions. Not sure how that turned into ‘abandon as fundamentally flawed’…


even if it were it would not matter, the playtest is not over. It only needs to prevent the final result from being below 70, not everything along the way. That is not how any of this works

How does it not matter that nothing in the process yet has ranked in the 20's, yet you want to claim it is flawed because it leads to results between 80 and 20?

Yes, you have claimed "they should ask better questions" but you have ignored any thought that you might not be correct. And I often find myself frustrated by people like you, who with seemingly so very little knowledge on a subject, just decide to declare "Obviously they should do it better! Because I know how to do it better!" But you seem to not know how little you know about... any of this. I don't want to be mean, but you seem to be seriously claiming that finding 1 person in a group of ten who was confused means you need to investigate a 40,000 person survey for "obvious flaws". 10 people alone is 0.00025% of the survey. That is a quarter of a quarter of a quarter of a quarter of a quarter of a quarter of a percent. Even the best, absolute BEST statistical data in the entire world has a variance of 1%.

You'd need to find AT LEAST 400 people who were confused to have even a CHANCE at it making a single percentage point difference. And the difference between keeping the ability and losing it, between 70% and 50% is 20 points. That is 8,000 people who were so confused as to put data in that appeared to not want something they wanted. While NOT registering that inconsistency in their comments or their overall rankings of the class.

You have 1 person. Perhaps as many as 5. And you want to use THAT as evidence that WoTCs processes for data interpretation and collection need a complete overhaul. 5 people, when you would need 8,000.

 

[mamba]

No? Like, not even a little bit? Like, this shows a disturbing lack of knowledge of how statistics work?

let’s say you would not expect to find any in a sample of size 10, does that work for you?

Again... no? Like, let us say I walk into a room and out of five people, one of them is ill. Does this mean I should begin investigating for a global pandemic? After all, in such a small sample size as 5, I found 1 person who is ill. Therefore we should investigate, right?

very different things, you expect people to be sick, they are every day without there being a pandemic.

You do not expect your poll to ever be misunderstood, and yet we found a case where it was this easily

If one person randomly exploded in the streets, would you not expect an investigation? This is much more what we found than your sick person comparison

Except... no. We shouldn't. One person in a single group of five indicates NOTHING about a population of hundreds, let alone thousands or millions.

that is why I said you cannot extrapolate to how widespread it is, but you absolute have found something worth investigating

 

[mamba]

Show me the document that states so.

you know as well as I do that there is no document to show. I made a case based on what is available, you have not made a counterargument based on what is available. You just keep repeating ‘but WotC smart tho’

And yet you keep making the same claims...

different from the one I repeatedly refuted and you always keep bringing up for no good reason

Because Wizards could look at a single vote of Dissatisfied, along with comments explaining why it was dissatisfied, and take that as a sign to follow the revisions laid out in the comment. This is perfectly logical for them to do. A single vote of dissatisfied does not a 50% rating mean.

and how many people fill out the text box? 5%? What do they do in those cases where it isn’t?

and continue making claims based solely on your own opinions?

sounds more like you… I at least explained my reasoning

You never mentioned anything at all other than ‘WotC knows what they are doing’. If you cannot engage with my arguments and just keep on repeating one line, then I am moving on

How does it not matter that nothing in the process yet has ranked in the 20's, yet you want to claim it is flawed because it leads to results between 80 and 20?

the 20s are in the past, it did not stop it then. That nothing in the playtest ranked that low is not the result of the playtest, it gets no credit for that. That should be obvious… the playtest can only kick in after the first proposal was published, not before

Yes, you have claimed "they should ask better questions" but you have ignored any thought that you might not be correct.

it’s not my fault that you are unable or unwilling to produce anything worthy of consideration. Half of your post was just you misrepresenting my position, the other half boils down to ‘WotC knows what they are doing’ in 10 different ways

Unless you get to the point where you make an actual case to refute what I wrote, instead of just unsubstantiated claims, there is not really anything to discuss, we are just going in circles

 

[Amrûnril]

Pause.

Why are you saying the cut-off is arbitrary? 70% approval is not arbitrary. It is seeking more than 2/3rds of the approval. It was chosen very specifically and very carefully.

Are they treating a 5-point scale like a binary? No, actually they aren't. Because it ISN'T a binary. 80%? Keep as is. 70% probably keep as is, but maybe tweak. 60%? Needs a bit of work, but we can salvage it. 50%, probably should scrap it. That is not binary. That is a range.

So, at this point, you have created two out-right falsehoods. And you will now use this false base as a jumping off point to declare something else, but your foundation is too weak to support anything.

The data the developers are actually receiving is A% strongly disapprove, B% disapprove, C% neutral, D% approve, E% strongly approve of a given game element. They way they are describing the data is X% approve, Y% disapprove. Each data point (survey response) is being treated as having one of only two values, despite being collected on a five point scale.

While there are certainly reasons to look for supermajority approval, the specific numbers are arbitrary in that there's no intrinsic reason 60% is a better cutoff than 55% or 66% percent. What's more problematic, though, is that these cutoffs are being used to infer something the survey never asked about. Whether a design idea merits further experimentation is a fundamentally different question from whether playtesters are happy with its current implementation, and I see no reason to believe a fixed percentage cutoff can reliably translate one to the other.

 

[mamba]

The data the developers are actually receiving is A% strongly disapprove, B% disapprove, C% neutral, D% approve, E% strongly approve of a given game element

there are only 4 options, no neutral, not that this really changes anything about your argument

 

[tetrasodium]

Yes thank you for re-posting the graphic I posted. Which doesn't show the thing we're talking about (they show no increase in the thresholds nor why there would be an increase or any mention of discounting). And yes it matters. It matters because if they're doing that, they're doing that to realize the opinion that's being expressed, which is the purpose of the surveys.

You seem to be under the impression there is something bad about being an optimizer. I think that is a bad opinion.

There is no reason to talk about non-homogeneous groups like "optimisers" because the weighting does not distort their vote as a group. At no point in our exchange on polling weights and math have I been talking about "optimisers".

I have been talking about math through this entire exchange. Things like the graphic were referenced because we are arguing over basic facts that are clear and in the open to degrees that simply do not support your maybes and what ifs about a scenario that never occurred. Wotc weighted the voting. Because of that weighting they need to sanitize for the repeatedly defined munchkin brigade* and not just say ,"well 70 percent was the target"... There is no indication that they even consider the distortions caused their own weighting.

* a group who can be expected to reliably vote against any check to or reduction on the power of a PC build no matter the benefits reasons or things enabled by said check on power for the game as a whole.

 

[Charlaquin]

Another thing to consider is that WotC are not just looking at approval percentages in a vacuum. They’re comparing them to approval percentages in the big 2022 survey. If a changed feature scores lower than the analogous feature did in that survey, naturally they would drop it, even if it’s 70% or higher, because clearly the old version was more popular.

 

[mamba]

Another thing to consider is that WotC are not just looking at approval percentages in a vacuum. They’re comparing them to approval percentages in the big 2022 survey. If a changed feature scores lower than the analogous feature did in that survey, naturally they would drop it, even if it’s 70% or higher, because clearly the old version was more popular.

I mentioned that a long time ago, they should if they measured absolute satisfaction, yet they never said they compared against existing ratings, it is always 70% improve, 80% keep.
This also works both ways, they then should accept proposals with a rating below 70% in some cases, yet they never do.

This (fixed percentages) should only be the case if they measured relative satisfaction, i.e. X% like the new option better.

I believe they are measuring relative satisfaction, if not intentionally, then at least because that is how people rate. They compare against what they have today, they do not rate in a vacuum.

I assume WotC know this and that is why we have fixed percentages, yet not once did they say that this is what they are doing. Another disconnect.

Care to explain why what they say they do and what they actually do does not match? It doesn’t under your interpretation or mine… I can explain it, but what is the explanation under the ‘WotC smart’ doctrine?

 

[Remathilis]

Another thing to consider is that WotC are not just looking at approval percentages in a vacuum. They’re comparing them to approval percentages in the big 2022 survey. If a changed feature scores lower than the analogous feature did in that survey, naturally they would drop it, even if it’s 70% or higher, because clearly the old version was more popular.

My guess is a rule has to pass three thresholds for adoption:

1. Does the rule score well on its own?
2. Does the rule score better than its 2014 counterpart?
3. Did it score better enough to warrant the disruption of changing it?

So let's take bardic inspiration. It scores X in the 2022 survey. They update it in UA 2 and it scores Y. If X > Y, 2014 wins. If X < Y but not significantly, 2014 wins. If X < Y by a large amount, Y wins. Y has to beat X by a margin that makes it worth it. In the case of inspiration, it didn't beat X (or not significantly enough) to warrant keeping. It simply wasn't worth any potential incompatibility issues for a few percent gain in approval.

So something like replacing pact magic or the three spell lists could absolutely score well in the survey, but either not better than the 2014 version or not significantly better enough to make the change worth it. A rule has to beat not just its own approval threshold but the rule it's replacing and the inertia to change it. Which is why the changes we've seen have been on no brainier things (berserker, beastmaster, champion) or wildly popular changes (weapon mastery) but the contentuous ones are failing.

 

[Mistwell]

There is no reason to talk about non-homogeneous groups like "optimisers" because the weighting does not distort their vote as a group. At no point in our exchange on polling weights and math have I been talking about "optimisers".

It sure seemed like "optimizer" has serious overlap with what you're calling "munchkin" right?

I have been talking about math through this entire exchange. Things like the graphic were referenced because we are arguing over basic facts that are clear and in the open to degrees that simply do not support your maybes and what ifs about a scenario that never occurred.

I posted that graphic, not you. And I said "I think they said thy increased the threshold but it's a vague memory and I am not sure" to make it utterly clear I am not sure. And now, without quoting that part, you re-posted the graphic and claimed somehow the graphic I posted doesn't support the comment I made, which I never claimed it did, and claimed I am making a "what if scenario" out of it, which I never did. Why did you just do that?

Wotc weighted the voting.

Did they? Show me proof of that. That graphic isn't showing weighting, it was my vague comment about it remember?

Because of that weighting they need to sanitize for the repeatedly defined munchkin brigade*

No they don't. You have not in any way established they even exist, much less are bad or need to be weighted. That's the entire thing we've been discussing and now you're just asserting it's all as you say and must be adjusted for?

and not just say ,"well 70 percent was the target"... There is no indication that they even consider the distortions caused their own weighting.

There isn't any indication of weighting yet either, right?

* a group who can be expected to reliably vote against any check to or reduction on the power of a PC build no matter the benefits reasons or things enabled by said check on power for the game as a whole.

Yes and that view is as valid as any other view. You are arguing there is some reasons to "weight" or adjust for that view - with no justification behind your argument. It's absolutely a valid perspective, and should be equally represented in any decisions as any other view. Why is YOUR view more valid than the view you're describing? You have yet to show why this is somehow an inherently bad view for anything, You're just asserting it and then declaring it to be so.