The COMMUNITY Tier List is HERE!

General Discussion
21

Me and some other people apparently lol

22

I calculated a score for each character based on winning, losing, and even matchups. His score was more than 2 standard deviations away from the mean. No other character was that high.

It’s been a while since I took stats, so I hope this is all correct. :slight_smile:

1 Like
23

Because data, that I have yet to see, was said to be used. Data is used to back up evidence/facts. Since I have yet to see it, I have reason to doubt. Also, as I have said, it still doesn’t account for players (and it likely never will, since everyone just “assumes” that the player plays at the best possible level, which in itself, is an impossibility, since everyone is fallible and makes mistakes). In other words, it’s faulty data that’s inconclusive.

24

That’s my point.

25

I know. You guys were credited in the character match ups. What I’m asking is more of a math equation. Like what makes a character S tier. Is it the .9 interval test. I’d like to know @llPaulBll

26

Apparently everyone thinks fulgore has no bad matchups. I admit. I chuckled.

27

What are you looking for dude, screenshots of the messages people sent me on Twitter? Do you think I made up the data or something? Why do you always do this?

Here’s the aggregated data file if you’re really that curious.

2 Likes
28

What is chart confidence?

1 Like
29

Well, Nicky himself provided the main chart for Fulgore, so. :slight_smile:

3 Likes
30

My opinions on the tier list:

Jago should definitely be higher than C. I feel like he’s one of the better balanced characters in the game, not to good but not to bad. I think he should be moved up to B.

I feel like Combo should be moved up to C tier. I feel like maybe he’s one of the weaker characters in the game, but not enough to be put in the bottom. I’ve personally had a lot of “success” with Combo.

I think Gargos is better than C, I think maybe he should swap positions with Hisako (I don’t feel like she’s as good as the other characters in B, to be honest. Most Hisako’s I fight online, even the Gold/Killer ranked ones, tend to be easier fights).

I feel like Spinal is better than D tier, and that maybe he should swap positions with General Raam (who I personally feel isn’t as good as a lot of characters in this game).

I don’t think that Omen is an A tier character. I feel like he should be swapping positions with someone in B tier, but I don’t know who
 I know that sounds weird, but basically what I’m saying is that I don’t think Omen is quite that good and their are better choices for A tier.

31

Compares against the MUs provided by other people. Confidence is % of matchups that agree exactly. I need to loosen the requirement there to consider a difference of .5 to be relatively confident.

Let me know if that isn’t clear. I know it isn’t immediately.

2 Likes
32

Yes it is. I’m glad you were able to calculate this properly. Now how many SD do you consider it to be OUTSIDE of A tier? somewhere less than 2 SD?

33

Eh, I’m not fully on board with it. But, it’s no concern of mine.

34

So, the so-called “data” stems from a few pro/good players for the specific characters? And they’re just match-up numbers based on how they feel? That’s it!? That’s NOT the entire community. And doesn’t account for everything a character can do in every single posssible situation/scenario. This is like taking the advice of 50 really smart people that say an apple’s bad for you simply because they don’t like the taste of it - that’s not how studies are done with data. To make the data legit, you need a larger and more random pool.

I suggest we start a poll with every single possible match-up combination and ask in each 1 who everyone thinks will win, lose, or tie in that particular match-up, and let it evolve overtime as more people contribute. If the game or a character gets updated, then the poll (or parts) gets a reset. It’s simple, it’s random, it includes a large pool of people from all skill ranges. That way you really get what the community thinks.

It’s either that, or go with looking strictly at the players themselves.

Otherwise, you’re just looking at the opinions of some skilled/popular people who THINK they know what they’re talking about (which won’t always be true) and others that will blindly agree with them, just 'cause.

35

A tier is between 2SD and 1

2 Likes
36

I think you misunderstood “community.” The intent is not to poll everyone. In fact, that would pollute the data since 90% of people aren’t anywhere near expert level. This isn’t a trial or observational study. It’s an aggregation of expert opinions.

I appreciate your desire for accuracy, but you’ve misunderstood the motivation behind this project.

6 Likes
37

Then the lower the tier the closer it goes to the mean right?

38

I like this idea, but it serves an entirely different purpose.

39

Fine, but it still doesn’t make it even remotely true, since as I have routinely said, it doesn’t account for individual player skill. You have to consider ALL of the variables in play.


and don’t start calling a few so-called experts “the community” - they’re only a very small slice of the community.

40

Right, and under the mean would be the 2 lower tiers. Outliers well below mean would be F tier, but there aren’t any with this set of data.

1 Like