Originally Posted by
Adamas102
What you describe in the second paragraph is pretty much what the tool does. It uses inputs and baselines that can be changed to accommodate different settings, and spits out metrics based on how prevalent traits are within the group and compared to the baselines. So your example of an "entire cast of overweight, black, transgendered people" would garner a lot of "low" scores because of the lack of variety within the group. That doesn't mean it's a value judgement on those traits, it's just telling you "diversity amongst this cast of characters is low" and then it's up to the developers to decide what to do with that information. It's also not about saying "you have a problem" but rather "there might be room to develop these characters more".
0 isn't "bad" and 10 isn't "good". The aim isn't to get an "amazing score" because that's not how the system works. They're really just an indication of how prevalent certain traits are given the parameters that are set beforehand. 0 means common, 10 means rare. And yeah, I've said it before, the "scoring" system as it's presented isn't great (partially because of how easily misconstrued it is by people who don't even read the article), but no matter what they went with the smooth brains would still try to use the system to grade themselves and then bitch about how it's unfair.
With the last example you gave comparing a "cis, white, male cast" to a "black, female, lesbian cast", the tool is going to give both groups low scores because internally there's no diversity there. That might be fine depending on the setting the game takes place in, and from there the tool would also gauge a number of other traits that could tell you "hey, looks like you copy/pasted the template for "surfer dude-bro" or "sassy black woman" for all of these characters. Is there more room for variety here?".