ELO ratings data was compiled using Demolyzer 2.0.

I am interested in gathering season 3 practice game MVDs so I can compile 'pre-season ratings.' If anyone wants to help, that would be much appreciated. I will do what I can to scan the servers for games, but if you have some feel free to stop by #demolyzer and I can give you an FTP account to upload to).

Thanks -Cyanide

The reason why ganon's team is so low is due to Phantasy. He was near top in div2, but in div1 he only played 1 game and did relatively poorly, which really hurt his div1 rating. My guess is had he played more games he probably would have a higher rating.

This was his one game:



These are the ratings for the entire Season 1 and Season 2 combined:

What about Div2 ?

Man did I suck that bad?

Very interesting!
Reminds me of the days when I played NBA Live and NHL Hockey and focussed on boosting players' stats :-)

This is interesting, but it's worth keeping in mind the huge effect that your team mates performance and your role in the game have on your statistics, eg. kloze is 7th while fix and darkki are in the bottom 12. Generally speaking, ranking systems in team sports are only usually good when applied to teams with consistent lineups and not the individuals within said team.

Stev,

The iELO algorithm I used actually takes into account your teammates' iELO as well.

iELO does the following:

1. Takes into account your teammates' iELO
2. Takes into account all opponents iELO

So, it is a very good measurement of 'performance.' Keep in mind that is past tense. This is not a measurement of future skill or capability.

If you only play 2 games and do bad in both, obviously your rating will be low. If you play an awesome team and do well but still lose, you can still have a good rating. But if you play against a low iELO team and lose bad, your rating will get crushed.

The reason darkki and fix are so low, is due to two reasons:

1. They had a bad season 1, and got blown out most games, even against other poor teams (thus magnifying their low rating)
2. They only played season 1, so there were no 'good' games to make up for their poor ones.

Here is a perfect example of how iELO works well, even if you are on a bad team. Below are rating results from North America's Stronger Than All 4v4 tournament season 1. Notice squeeze and serp had bad records and teams. But, iELO takes that into account and they were rated accordingly.

Defcon 5:

Here are ratings for Season 2's Div2:

downloading v2.0, it's amazing!!! congratulations man!

The equalize teams feature would be über to add to servers.
Wet dream with not having to rpick a million times until everybody is satisfied.

Needs loads of work I reckon, but still...

Love this tool!

Thanks for posting div2!

This is an amazing program. However I still think it misses quite a bit of player skills in TDM. I'm hopeful we can think up a way to capture iELO (Teamplay) to add to it. :-)

BTW, does anyone know if/when stats.quakeworld.nu is coming back?

stats.qw.nu works for me.