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

[Mistwell]

You still do not understand samples and quality control. If the survey worked, there would not be such 'noise'.

No. That's impossible. You're claiming surveys can be perfect. That's...the opposite of everything we know about statistics and why and how they are used.

I think we're done. I think everyone is very clear on where you are coming from.

 

[mamba]

No. That's impossible. You're claiming surveys can be perfect. That's...the opposite of everything we know about statistics and why and how they are used.

Perfect in what way? Accurately capturing the opinion of the person filling it out? Sure, that kinda is the basic necessity, yet many fail even at that. Perfectly representing the population at large? Much harder to impossible, but that is not what we are discussing here, we are discussing the former…

 

[tetrasodium]

I literally said everyone, and never mentioned GMs. You've fabricated a strawman. GMs are part of everyone.



IF you are are GM where that class was playtested, you have just engaged in a playtest of that class. It's a neutral question. It's not tailored to the player. Nobody is more equal than another. GMs are not being oppressed or dismissed by the playtest. In fact, the instructions under "How to Playtest This UA" talk about DMs and creating your own adventure for the purpose of playtesting.


I am absolutely not dismissing or discrediting criticism. I have my own criticism. But when you make a public post here on a message board intended for a back and for conversation, you don't get to complain about getting feedback on what you're saying.


Yes, that was me. That, because someone might like both the old and new rule exactly equally and exactly as favorably, it is no harm to those people's views to assume that maybe you need to pad to account for that. You were saying some votes should be dismissed which WOULD harm the views of those voters and I explained how that's a different situation - dismissing voices who care about their view being heard is not the same as dismissing those who suffer no harm from it and are fine with it.


That's you putting your ethical view on other people inappropriately. No, you as a DM do not have a more important voice that others who consume this product. But EVEN IF YOU DID (which you do not) it would still be wrong to split it not between "DM vs Player" but instead between "DM vs a specific type of player you personally don't like." There is absolutely nothing objectively good or righteous or just about your view. It's you saying your view is objectively better than the views of others for no reason other than it's you.

No, it's deeply irresponsible to assume a type of player's views is a "distortion" because you disagree with it. You're essentially saying, "I don't like the color blue so anyone who likes the color blue should be "corrected" as a distortion because I don't like the color blue and therefore blue is a distortion." It's circular logic and has no basis in any kind of ethical view.

The statistics might not be easy but you again miss the ball, we are literally talking about the impact of the weighting you yourself claimed to be occurring not preference.. The distortion taking place in

Spoiler: this statement

You've lost sight of the ball. The munchkin brigade voting against any slight to power came up because of your earlier claim that wotc lowers the veto threshold by padding the breakpoint for approval to correct for responders who will vote positively for anything at all. The GM needs to consider all of the players at the table & can be expected to have a good probability of voting for healthier options a reasonable percentage of the time rather than the more selfish ones of respondents that are exclusively players & only interested in their own power. Correcting for the distortions is responsible polling when those distortions are caused by their own weighting & not about "discounting" anyone.

is wotc's own weighting allowing the previously described munchkin brigade to have an outsized result when voting against any reduction in power no matter the reason or benefits of that result. That distorted & outsized ability to veto it is a direct result of the weighting being done

Spoiler: here

I have a vague recollection many years ago - and I am not confident it really is just vague - that at some point one of the Devs said some percentage of fans will give full approval to anything WOTC puts out in a UA and so they needed to increase the scale to compensate for those "sure I'm good with anything" people.

If the survey metrics apply weighting to normalize against the impact of people who would vote for anything the weighting doesn't end there. Once you apply such weighting it will also create a favorable weighting to amplify the vote of any group who would vote against anything meeting certain criteria like those who would vote against any power reduction unless steps are taken to weight against it.

 

[Mistwell]

If the survey metrics apply weighting to normalize against the impact of people who would vote for anything the weighting doesn't end there. Once you apply such weighting it will also create a favorable weighting to amplify the vote of any group who would vote against anything meeting certain criteria like those who would vote against any power reduction unless steps are taken to weight against it.

No, that doesn't follow from what you're claiming it follows from. The standard was hear every voice - discounting a voice in half that is equally good with the old rules or new rules is hearing their voice.

You're trying to discount a voice in a manner which does exclude hearing them. Which is meaningfully different.

 

[Remathilis]

No. That's impossible. You're claiming surveys can be perfect. That's...the opposite of everything we know about statistics and why and how they are used.

I think we're done. I think everyone is very clear on where you are coming from.

I believe the phrase you were looking for is "There are lies, there are damned lies, and then there are statistics."

 

[tetrasodium]

No, that doesn't follow from what you're claiming it follows from. The standard was hear every voice - discounting a voice in half that is equally good with the old rules or new rules is hearing their voice.

You're trying to discount a voice in a manner which does exclude hearing them. Which is meaningfully different.

When talking about a group with an amplified voice with their vote because of weighting being done... That would be the result of counter weighting against other weighting that was increasing their voice. It would also be the result of simply saying "hmm, lets run the numbers for this so called nerf without the weighting correcting for people happy with anything". Normalizing an element that is given outsized clout is "discounting" it.

 

[Ashrym]

You still do not understand samples and quality control.

That's not how samples and quality control work. We're talking about the law of large numbers here. A person might see a small sample that violates the minimum quality guidelines and do a deeper dive, but that individual 1/10 in such a small sample means nothing when the larger representation demonstrates something else.

If we see a statistical analysis that shows 990/1000 and take a sample of 10 were 1 is an issue that cannot change the 99% to 90% and that 1 out 10 is considered a one-off. When that 1/10 is coming from and even larger sample it's even less significant to the total. Depending on context (that you don't have) behind that outlier it could be an opportunity for improvement but that still won't change the overall statistic. That's an outlier.

 

[cranberry]

Ultimately it doesn't matter.

For a substantial percentage of the RPG players, D&D is the one and only (RPG) game that they would ever consider playing.

So, unless WoTC's implements changes that make the game completely unplayable and unfun, people will just grumble about what they don't like, and then show up to the next session.

Such are the advantages of being the industry's 800 pound gorilla....

 

[mamba]

That's not how samples and quality control work. We're talking about the law of large numbers here. A person might see a small sample that violates the minimum quality guidelines and do a deeper dive, but that individual 1/10 in such a small sample means nothing when the larger representation demonstrates something else

If we we had a larger representation that shows otherwise I would agree, then the small sample is contradicted and an outlier.

We do not have a larger sample at all however. All we have is a small sample that given its size really should not contain any issues, and yet it does.

Since this is all we have, the next step would be to take a larger sample. Until this is done, all we have is an indicator that this should be investigated.

Please provide the large sample that shows the small sample wrong. Until then… dismissing the only sample you have, because you do not like its result, is not a valid approach to this

 

[Mistwell]

When talking about a group with an amplified voice with their vote because of weighting being done...

All groups have their voice amplified to the same degree. These munchkins (as you call them - I think it's a terrible and inherently biased description) have the same voice as anyone else.