Let's Talk: 3ddy NA and JimMashima bots + Wavu Wank 100% proof of defective Tekken MMR and inferiority to Wavu Wank ELO

[C.C. 折オリ枝 の 夫]

Long title, I know, it ain't mine, I'm just stealing content as per usual, and since this time the one I'm stealing from wrote everything, I'll just copy paste, with a link at the bottom to OG poster.

-------------

Here are two Wavu Wank metrics of two 3ddy bots. 3ddy bots are just Eddy bots that hit 3.

https://wank.wavu.wiki/player/2q4iyjerLYHy

https://wank.wavu.wiki/player/55A8fa2EQR88

The ELO rating of the 2 bots are only 2 off. This is despite the two bots performing rather differently in the regions they are in with the NA 3ddySmurfy hitting red middle rank early with an early attainment of higher prowess and the Euro OG 3ddy hitting early red but once but being rather hard stuck on rank and prowess much longer than NA 3ddy. Despite all of this Wavu Wank uses ELO across their many matches and results them at only 2 off which is amazing.

In the current rank system NA 3ddySmurfy reached high prowess rather quickly. You can also see how all later matchups since the prowess upgrade will continue to match him up to higher ELO players. The T8 MMR system is doing a bad job matching up 3ddy due to the prowess system. But it wouldn’t only do a bad job with bots I am sure if folks use this Wavu Wank you can find many other examples of it mismatching human players. What is great about using the bot examples is we know they do not improve and they play a constant way each game. The only huge variable is matchmaking and opponent behavior. Amazingly ELO from Wavu Wank does a masterful job assigning play strength to the bots and does the same to the human opponents.

Anyhow it is proof that the current MMR system sucks with prowess. It is also proof that current MMR system can grossly mismatch you (there are tunnel vision dorks that keep insisting this is impossible due to their anecdotal bubble view of their own 100 games at +-2 ranks but we have Wavu Wank cataloging and ranking everything to show undeniable proof of this).

Unfortunately we still as a community haven’t had our MMR complaints heard particularly because many influencers fall into god ranks where their only big concern is match time and lag. IMO if dev team actually spent time analyzing Wavu Wank they can see the failures of current MMR holding back experience of lower level players. The complaints of lower level players is what started this journey to prowess MMR, but none of the lower level players are even gonna keep playing the game to point out the real flaws. Instead T8 dev team borked MMR off their own Eddy set to beginner mishaps and unfortunately this problem isn’t being reported enough! The failures of T8 prowess MMR is all contained within Wavu Wank ELO database and I hope Bamco and more influencers actually cared.

https://www.reddit.com/r/Tekken/comments/1clzi43/3ddy_na_and_jimmashima_bots_wavu_wank_100_proof/

-------------

Ok, well, from here on out, you know the drill, discuss.

[processor.dev1l]

Yeah, wavu wank is amazing, it uses system based on ELO chess but modified for any 1v1 competition and I honestly believe all fighting games should use it for matchmaking.

[C.C. 折オリ枝 の 夫]

Originally posted by processor.dev1l:

Yeah, wavu wank is amazing, it uses system based on ELO chess but modified for any 1v1 competition and I honestly believe all fighting games should use it for matchmaking.

It's Glico2, not ELO, but meh, same difference I guess since everyone just refers to it as ELO anyway....

[Arbor]

Someone also made a spreadsheet for wavu where you can see which characters give you the most problems in ranked.

https://www.reddit.com/r/Tekken/comments/1cm04pm/google_sheet_win_rate_vs_specific_matchups/

[processor.dev1l]

Originally posted by C.C. スパークル:

Originally posted by processor.dev1l:

Yeah, wavu wank is amazing, it uses system based on ELO chess but modified for any 1v1 competition and I honestly believe all fighting games should use it for matchmaking.

It's Glico2, not ELO, but meh, same difference I guess since everyone just refers to it as ELO anyway....

As I said, it's a system based on ELO, but modified for all 1v1 competition.

Source: https://en.m.wikipedia.org/wiki/Glicko_rating_system

[processor.dev1l]

Originally posted by blackzephyers:

Someone also made a spreadsheet for wavu where you can see which characters give you the most problems in ranked.

https://www.reddit.com/r/Tekken/comments/1cm04pm/google_sheet_win_rate_vs_specific_matchups/

Man, that's amazing! 💕

[C.C. 折オリ枝 の 夫]

Originally posted by blackzephyers:

Someone also made a spreadsheet for wavu where you can see which characters give you the most problems in ranked.

https://www.reddit.com/r/Tekken/comments/1cm04pm/google_sheet_win_rate_vs_specific_matchups/

Accordion to that thing, my highest winrates on Alisa are against Lili, Alisa, and Claudio, while my lowest is against Shaheen, however the Shaheen one is factually incorrect, because it only had the latest 1 Shaheen player I faced, all the ones from prior to that aren't in the database, so there's that.

So imho, should be taken with a grain of salt for the time being unless you only start playing ranked after March 27th.

[StRyK]

What are your complaints about MMR?

[Sirlion]

As always fans have to solve the devs' inability to do anything. I wasnt such a downer in T7 despite everything, and had plenty of fun until the matchmaking update.
This shows just how incompetent they are at tackling their own issues. Not like the Azucena WR3,2 update didnt show that already... or the tornado which they had to revert.

TL;DR devs dotn know ♥♥♥♥ about how their own game is being played, and keep making mistakes. They NEED TO REVERT THE MM CHANGE. They HAVE to. This needs to reach Twitter and every other platform until they do.

[StRyK]

Originally posted by Sirlion:

As always fans have to solve the devs' inability to do anything. I wasnt such a downer in T7 despite everything, and had plenty of fun until the matchmaking update.
This shows just how incompetent they are at tackling their own issues. Not like the Azucena WR3,2 update didnt show that already... or the tornado which they had to revert.

TL;DR devs dotn know ♥♥♥♥ about how their own game is being played, and keep making mistakes. They NEED TO REVERT THE MM CHANGE. They HAVE to. This needs to reach Twitter and every other platform until they do.

You have no idea how the rating worked in T7, and it didn’t work there at all, since it could be corrected by replacing files.

[SEATEK The_LEFTY]

We know it didn't work like this current system?

[no use brain]

That's just a sample size of 2, you would need way more bots to have a definite conclusion.

[AGA | prinnyuber]

Watchu talkin about C.C.? I don't have a voice. Tournament players represents us as an a-WHOLE! Therefore tournament players are the law of the land! So shall it be spoken, so shall it be written, and so shall it be done!

[Sirlion]

Originally posted by AGA | prinnyuber:

Watchu talkin about C.C.? I don't have a voice. Tournament players represents us as an a-WHOLE! Therefore tournament players are the law of the land! So shall it be spoken, so shall it be written, and so shall it be done!

Can you stop using ChatGPT for creating your posts? its super cringe.

[SEATEK The_LEFTY]

It's ChatGPT?

Damn.

[天佑中共🐥the light]

nah just prinny mocking yall mouse icon haver. by having T8 means u guys are willing to eat all the sht Bamco gushing out. cant refund? lmao gottem. gotta eat ze bugs bro
open wide~🤤 Fahk and Marduk are next