Trout, thanks.

Do you happen to know, or is there a reference page on the formula of their rating numbers ?

This was produced by Dr. Bae Taeil using repeat calculation(I am not sure what it means.). It is not official ranking but Korean Baduk association use this table to produce their Korean Ranking.

I think this looks like a elo system.

Uberdude, Thanks.

I'm guessing the ELO equations also completely ignore the time settings of each game. The rating systems of KGS, IGS, AGA, BGA (?), too.
Or the prize money of the tourney, although this one does have some weightings for different stages of the tourney.

There's only so much one can try to squeeze into the calculations.
One has to stop somewhere.

Now I'm curious about WeiqiTianDi's annual (?) rankings of the pros.
How their tables compare to this one.

Although there isn't an explicit weighting for the the time settings, I imagine "First Class" tournaments will tend to have longer time limits than "Second Class" or "Third Class". Similarly later rounds of tournaments may have longer time limits. So longer games will tend to have a higher weighting factor.

In chess they have separate tranches for Classical, Rapid and Blitz, all based on time settings. The differences between the various tables can be quite startling, although the very best players seem mostly to vary within a range of maybe up to 5 places. There are players who see themselves as e.g. Blitz specialists and will enter every Blitz they can. Conversely, many players seem to dislike and so avoid one or more categories. No doubt this affects the reliability of the tables, but it is unclear how much.

The most important point is that the Classical (long time limits) table is the most respected one.

Nice.

36 Korean players, 61 Chinese players, and 3 Japanese players? Is this a surprise?

_________________
12k: 2015.08.11; 11k: 2015.09.13; 10k: 2015.09.27; 9k: 2015.10.10; 8k: 2015.11.08; 7k: 2016.07.10 6k: 2016.07.24 5k: 2018.05.14 4k: 2018.09.03 3k: who knows?

Not particularly. The only surprise for me was 4 of the top 5 were Korean. This is a bit different from
http://sports.geocities.jp/mamumamu0413/total.html

With 3 of the Top 5 were Chinese.
Park Junghwan
Ke Jie
Shi Yue
Kim Jiseok
Tuo Jiaxi

There is a big difference especially for Tuo who is 18th on this one and 5th on the Japanese site.

goratings.org has
1 Park Jungwhan 2580 2015-07-17
2 Ke Jie 2553 2015-07-19
3 Lee Sedol 2515 2015-07-17
4 Shi Yue 2503 2015-07-11
5 Kim Jiseok 2501 2015-07-20
6 Tuo Jiaxi 2493 2015-07-11

So is there some Korean bias on De Bae Taeil's ratings?

Between China, Korea, and Japan, there will likely always be bias

_________________
be immersed

There could be some bias in that Chinese tournaments in which Tuo Jiaxi does well might have a lower weighting in the Korean rating system than in the Chinese/Remi's/mamumamu's. Or it might simply be that the collection of games used is different and the Korean rating doesn't include some small Chinese tournament he won a lot of games in.

And Kang Dongyun is getting a big bonus from Korean matches.

My recollection is that Dr. Bae Taeil's system does a lot of "stuff" in terms of weighting to adjust an ELO style calculation. I don't recall ever seeing an English language discussion of whether that "stuff" makes sense or not.

_________________
Occupy Babel!

I'd really love to know more about his method. Does anybody have any reference (even in Korean?). It seems my system ranks Japanese players consistently higher than his ranking.

Rémi

Gogameguru gave a link to a a paper he wrote in English, I believe, on the post "30 of the world's best go players" 2011-2012

http://english.baduk.or.kr/sub03_04-1.htm

The comparison of the theoretical winning rates against the actual winning rates is also very interesting.

_________________
On Go proverbs:
"A fine Gotation is a diamond in the hand of a dan of wit and a pebble in the hand of a kyu" —Joseph Raux misquoted.

Thanks for your reply.

I had noticed the link, but I am not sure this is the algorithm that was used. That page mentions a variation of incremental Elo. But that does not fit with the "repeat calculation" description of trout.

I believe incremental rating methods cannot work well when a strongly connected group of players (Japanese pros) play rarely against the rest of the world.

I don't see a Korean bias from Dr.Baeil's ranking system. Tuo's ranking did go up significantly. I dont think he did great on international tournament so what do you expect? Only one thing that all the ratings agree upon: PJH is no.1 right now.
Personally I see KeJie becoming no.1 pretty soon though.