Reply to thread

Message

<blockquote data-quote="Blue" data-source="post: 7780111" data-attributes="member: 20564">See, right here is a great example of why this data is dangerous when not ibnterpretted right.There is literally NO WAY to look at characters and know if there is a featless game.&nbsp; Only the opposite.&nbsp; But a character without feats doesn't even necessarily lean (&gt;50% correlation) in the direction of a featless game until probably the 2nd or maybe third 3rd ASI.&nbsp; And they have told us they have less data there.&nbsp; Even then it just means it's more likely, not that it's true.Trying to use that to collaborate self-selected survey results (a system with well documented biases) that most games are featless (which it doesn't meansure does nothing.&nbsp; It's actively harmful because it tricks people into thinking there is evidence from multiple sources saying the same thing so it's likely true.&nbsp; When they don't say the same thing, and there are issues with the data collection.&nbsp; D&amp;D Beyond also has self-selection issues for who uses it.&nbsp; No one in either of my groups do.&nbsp; In addition, some types of data collection is compounded on that some offered content is free and some is gated by cost.&nbsp; Feats are one of these.&nbsp; Between the basic rules and the OGL, only the Grappler feat is listed.&nbsp; So anything they say about feats is even more suspect - either it's using the full player base, many (the majority?) won't have access to feats (outside Grappler) or they are reducing to a fraction of the sample size to those who purchased that content.So we have two pieces of data, both self-selected and one further distorted by gatekeeping, that are sorta-adjacent but are verifiable not the same thing, but because they are presented as such we get a lot of smart people thinking that we have confirmation about the majority of games being feat-free.But we don't.And that sort of bias confirmation is the danger of appoaching these as anything more than they are.&nbsp; Taken as they are it's great data.&nbsp; But it's not a representitive of games as a whole.&nbsp; Any more than Bob is the most popular character name in actual play.</blockquote>

[QUOTE="Blue, post: 7780111, member: 20564"] See, right here is a great example of why this data is dangerous when not ibnterpretted right. There is literally NO WAY to look at characters and know if there is a featless game. Only the opposite. But a character without feats doesn't even necessarily lean (>50% correlation) in the direction of a featless game until probably the 2nd or maybe third 3rd ASI. And they have told us they have less data there. Even then it just means it's more likely, not that it's true. Trying to use that to collaborate self-selected survey results (a system with well documented biases) that most games are featless (which it doesn't meansure does nothing. It's actively harmful because it tricks people into thinking there is evidence from multiple sources saying the same thing so it's likely true. When they don't say the same thing, and there are issues with the data collection. D&D Beyond also has self-selection issues for who uses it. No one in either of my groups do. In addition, some types of data collection is compounded on that some offered content is free and some is gated by cost. Feats are one of these. Between the basic rules and the OGL, only the Grappler feat is listed. So anything they say about feats is even more suspect - either it's using the full player base, many (the majority?) won't have access to feats (outside Grappler) or they are reducing to a fraction of the sample size to those who purchased that content. So we have two pieces of data, both self-selected and one further distorted by gatekeeping, that are sorta-adjacent but are verifiable not the same thing, but because they are presented as such we get a lot of smart people thinking that we have confirmation about the majority of games being feat-free. But we don't. And that sort of bias confirmation is the danger of appoaching these as anything more than they are. Taken as they are it's great data. But it's not a representitive of games as a whole. Any more than Bob is the most popular character name in actual play. [/QUOTE]

Verification