And just like that. 2 great builds have vanished. A loss for Build Diversity.

[Obtena.7952]

@"JusticeRetroHunter.7684" said:I think you've got it backwards. Nobody here thinks that except me and possibly you so far... and this entire time, is what I've been trying to prove against Obtena and others who think diversity doesn't even change at all...And Obtena doesn't really bring up good talking points imo. For example she just asked:

" For instance, at what point does diversity become 'bad'? What does 'bad' even mean?"

We've already said earlier that diversity is not black and white and that diversity lies on a spectrum between 0 and infinity...so how can that possibly be a good talking point? It's a loaded question, that will lead only to terrible discussion because there is no

bad

or

good

... there is only

more

and

less

.

HOLD the F on ... my discussion points are WAY more relevant than arguing the area under a curve and other academic math trivialities :astonished:

You're whole argument (whether you use math or not) is that this change is bad because "diversity" ... except that completely ignores ALL the other reasons the change was made in the first place that appear to be MUCH more important to Anet than diversity is for this case. You simply focus on the math and diversity because your position falls apart if you don't.

And as for example you bring up ... if you don't ask yourself if you are in the range of your diversity index where a negative diversity change is bad for the game compared to not making the change that results in the negative diversity change ... then you have NO argument at all.

[JusticeRetroHunter.7684]

@"Obtena.7952" said:HOLD the F on ...

Edit this out, moderators are scanning this forum and people are getting banned, don't want you to get banned either (i've already got like 3 warnings)

my discussion points are WAY more relevant than arguing the area under a curve and other academic math trivialities :astonished:

Your position is one of nihilism. "Anet made the change for their own reasons therefor anything we could say about it is meaningless and pointless."

Not saying that's not true...But it's just as bad of a discussion point as anything else because there is nothing to discuss about it. If i eat a peanut butter sandwich, you can tell me about how unhealthy it is to eat, and i could just say "doesn't matter what you say because i'm the one choosing to eat it" which is pointless argument at best, especially for any practical application.

Edit : Also, you are saying that the math here is trivial... you shouldn't be embarrassed to say that something is complex or too difficult to do.. Like if it is so trivial for you than you should be able to provide some sort of mathematical proof that can end this conversation right now...which you haven't been able to do... God did it, and so far i think the conversation we've been having has been the most constructive on this forum on the topic of diversity there's ever been about this game.

[Obtena.7952]

@JusticeRetroHunter.7684 said:Your position is one of nihilism.

No, it's one of realism and of a practical nature. I'm not rejecting diversity as a factor for game change ... I'm simply more inclusive of all the other ones you are ignoring to make your case. There isn't anything to discuss about all the other factors that affect game changes except for diversity? That's just not true. You just don't have a math language to talk about them.

And for your information, to ME this math IS trivial ... and to be fair, I already provided you an example where the math shows a game change doesn't always result in a diversity increase ... you just ignored it because you didn't know how to counter it with your fabulous math; you simply dismissed diversity as 'complex' at that point and continued ... and to be even more clear, the other guy also gave you an example.

This discussion IS NOT about math ... it's about understanding you can't predict how people react to game changes to assume diversity will decrease AND understanding there are more factors that go into a decision to make a game change other than diversity ... EVEN if you are not willing to talk about them.

[God.2708]

@Obtena.7952 said:This discussion IS NOT about math ... it's about understanding you can't predict how people react to game changes to assume diversity will decrease AND understanding there are more factors that go into a decision to make a game change other than diversity ... EVEN if you are not willing to talk about them.

Making a system, however rough shod, to predict how people react is mathematics. The model doesn't need to be perfect or even precise. I already presented a rather rudimentary one where some Anet dev is looking at #'s of each class pre and post patch change. I'd argue if you do that enough times it very much can be 'predicted' in the sense that you can gauge the odds of a something happening. You of course can't know for sure, but that's not the point of probability.

I also don't think it's particularly fair to claim Justice thinks diversity is the only change that goes into decision making. This is just their niche and they're trying to define it as clearly as they can because they think it has an important impact. There's merit to some people having a really precise point of view in one particular subject just like there's merit to people trying to summarize the broad picture. IMO, the latter often do themselves a disservice by never having looked at the precise elements of their picture and end up misunderstanding the broad picture.

@JusticeRetroHunter.7684

So had an awful class that I spent the entire time thinking about this and decided to do a super reduced case to make sure we are on the same page because I'm not sure we are.

So there are 10 people, and 5 builds. We will spread these people out to start as 0-4-3-2-1. This gives an index (I'm going to use index and not inverse index. So 1 is all the same, 0 is diverse) of4^2+3^2+2^2+1^2 / 10 ^ 2becomes30/100 -> .3

So there's a list of questions we can ask from this.

1. If we remove an option and everyone changes choice, does the result -trend- towards less diverse? The answer to this is YES. The reason for this is the minimum index in our example is .2 (every selection has 2). If we remove an option the minimum now becomes 2-3-2-3 or .22. The maximum remains 1 (everyone picks the same option). NOTE: This only applies when builds available is less than the population.
2. If we remove an option and only those who had selected that option before change choice, does the result -trend- towards less diverse? Here, I believe the answer relies on whether the option that is removed has a number of people on it greater than or less than average. IE in a system where 0-10-0-0-0 is select and you remove the option of 10, the resulting index will -probably- be less than 1. But if you go to our example of 0-4-3-2-1 and remove the option that has one person, the resulting system is either the same diversity (that person moves to the build with no other players) or decreased diversity (any other build)

There's other questions (A pertinent one is probably If we remove an option, everyone changes choice, and the population is less than the number of available builds does the result -trend- towards less diverse?) but that IMO is enough to answer the things I was trying to wrap my head around. The more problematic layer to this is that there is more builds available than players (but builds that are viable are less than players) so indexing players playing builds might not be measuring the diversity we want to begin with. Because whilst removing altruism scrapper 'shrinks' the available pool of builds and thus ensures a probabilistic decrease in diversity, it might ADD additional builds to the pool which would increase the odds of an increase in diversity.

This conflict is likely where the desire to bring in player behavior is, but like I said, that leaves the math behind and goes into science.

[JusticeRetroHunter.7684]

@"God.2708" said:So had an awful class that I spent the entire time thinking about this

lol i'm so sorry man. >.<

1. If we remove an option and everyone changes choice, does the result -trend- towards less diverse? The answer to this is YES. The reason for this is the minimum index in our example is .2 (every selection has 2). If we remove an option the minimum now becomes 2-3-2-3 or .22. The maximum remains 1 (everyone picks the same option). NOTE: This only applies when builds available is less than the population.

That's pretty solid proof.

2. If we remove an option and only those who had selected that option before change choice, does the result -trend- towards less diverse? Here, I believe the answer relies on whether the option that is removed has a number of people on it greater than or less than average. IE in a system where 0-10-0-0-0 is select and you remove the option of 10, the resulting index will -probably- be less than 1. But if you go to our example of 0-4-3-2-1 and remove the option that has one person, the resulting system is either the same diversity (that person moves to the build with no other players) or decreased diversity (any other build)

Again...this is pretty solid. nice.

There's other questions (A pertinent one is probably If we remove an option, everyone changes choice, and the population is less than the number of available builds does the result -trend- towards less diverse?) but that IMO is enough to answer the things I was trying to wrap my head around. The more problematic layer to this is that there is more builds available than players (but builds that are viable are less than players) so indexing players playing builds might not be measuring the diversity we want to begin with. Because whilst removing altruism scrapper 'shrinks' the available pool of builds and thus ensures a probabilistic decrease in diversity, it might ADD additional builds to the pool which would increase the odds of an increase in diversity.

I'm gonna take the time to sit on this question, cause it's quiet valid reasoning. But it doesn't sit right with me, As the number of possible builds shouldn't matter, much like the Simpson Equation, where the number of possible species to exist doesn't have an effect in the equation. There must be a valid reason they don't have it as a parameter. I believe it's because in biology, they know that the diversity space for number of possible species is probably near infinite, and there is always less population then there are number of possible species. This question is a huge can of worms, because it puts into question, why would dominant builds(or species) exist if there is enough space of possible builds(species) to make that could just as likely force an increase in diversity?

But ya man, great conclusions and proofs there...this is actually really kitten good i'm glad you spent the time to think about this in so much depth.

Edit: Okay, did a bit of research, and although i need to do more, it seems that this problem largely has to deal with lack of additional parameters. The additional parameter here would be competition.

Basically, we can't properly explain dominant species(builds) in diversity without talking also about competition, which we've talked about briefly at the beginning of this thread.

But basically there is an index for "dominance" that you supposed to ALSO use when you take an index for diversity. FML.

The math on dominance is just too complicated for me to process right now at this point in time, so i'll just provide a link to a lecture summary pulled from a scientific paper from J.P. Grimes and then leave it at that.

Exert from a lecture from the College of Agriculture and Life Scienceshttps://cals.arizona.edu/classes/rnr555/lecnotes/22.html

"In his 1973 paper (Nature 242:344-347), Grime asked "Is there a correlation between the appearance of these characteristics [dominance] and a decrease in species richness because subordinates are eliminated by dominants?"Grimes linked dominance and diversity w/ competition:The species that prevail are those that are best equipped to capture resources and maximize production: high competitive ability is represented by a combination of genetic characteristics...."