I noticed on IGS that players from the Americas seem to be much stronger in the DDK than players from Japan with the corresponding rank. Since the ranking is based on the people you play against and since people mostly play live games against people in more or less the same time zone, it seems like it is possible to have ranks on the same servers that have different values for different time zones. I guess this is not an issue for OGS since most of the games are correspondence.
IGS has also a “Japanese rank” mode where Japanese players get a displayed rank several stones higher than their real one, to make them feel less sad about their inflated ranks vs rest of the world. I’m not sure which one you would see and compare.
3 Likes
Now that sounds like a great concept. Shouldn’t be restricted to one country. Display a rank 5 stones higher than the real one and we all will be so much happier! I’ll become a dan finally!
6 Likes
I’m not sure. You can find many Japanese players on IGS at any time. Ranks could be different if Japan were divided into two populations which don’t mix, those who play in the morning and those who play in the evening.
Some time ago I had a losing streak after playing games in the morning (European time), so I thought players were stronger in the morning than in the evening. Then I had a winning streak and started to think the opposite…
2 Likes
I have noticed this since handicaps are automatic at IGS and the handicap I get is usually fair against players from the Americas (Canada, US, Mexico, etc.) but it seems to be way to high against the DDK players from Japan despite the fact that they have formally the same rank as the DDK from the Americas
Try Fox or Tygem.
1 Like
Surprising! I didn’t know that.
How is it handled when their rank rise up later?
I don’t know the details, maybe the offset changes based on rank e.g. 10k → 5k Japan, 5k → 1k Japan, 1d → 4d Japan, 10d → 10d Japan.
What about their opponent ranks? Not very credible if they play a 10k when being a Japanese rank 5k.
Count me skeptical. There are too many statistical pitfalls to parse the data this finely.
1 Like
Some years ago OGS ranking was tough. Being a 4k meant more to be near dan and i guess that was maybe some consequences of correspondance games. Were OGS players disturbed by this? I don’t think so. I’m really surprised if a special Japanese ranking has been introduced on OGS to have their rating here look like their rating home. Can’t everyone be under the same roof?
Not to digress, but it’s surprising to me how many folks love to say that ranks in Asia (or maybe just Japan?) are inflated. A frog in a well does not know the great sea…
On the original topic, I don’t know about IGS, but I feel the same whenever I change time zones as well. Now in Asia (homebase in North America) and had a losing streak when I first got here, but things later stabilized. I think it’s a combination of luck and personal winning / losing streaks.
By country, I understand it’s mainly Japan that has the inflated ranks.
By server, Fox is known to have inflated ranks compared to OGS.
2 Likes
Also, kid’s ranks in Chinese go schools are probably pretty inflated, which was my experience playing at a little one in Beijing in 2020-2021. The teacher in the school, a strong dan amateur, explained how it worked in his school, and gave me the impression that was the norm across the many go schools for kids. It’s pretty easy to imagine the market forces at play that drive the inflation, which I think was probably the cause for inflation of Japanese ranks.
2 Likes
Some Chinese go players did report on this evolution.
Recently to become 5d before the university is no more rewarded with some bonus points helping to join the best of them, so the pressure has been quite alleviated by elimination of the main motivation, and the rating system may cool down and return to some more reasonable graduation.
Last a small school with 1 dan teacher is quite… a small school (with limited perspective on more big official schools, like where the trimester regular tournament are usually organized.)
15 yrs ago i participated regularly in these tournaments (like 8 games/trimester you needed pretty good results in more as one event).
Things may have quickly evolved like everything in China but to resume, they were hundreds of contestants for reaching 2d 3d… For 4d was another matter, like a small classroom. For 5d like a study group.
IMHO children have a bit of fear and less experience against adults of same level but they had a lot of solid knowledge to rely on. And they will make perfect partners for blitz , their favorite speed.
I’m still surprised that all the go schools will overrate their students before the first stages of the graduation tournaments. (Not impossible but amazing)
Something similar but through other benefits (social status). I don’t see any direct link between both countries, so I’m guessing that wasn’t in your idea.
KGS did that one year, as an April Fools special. Everyone was 9 dan that day.
6 Likes
My understanding for Japan is that, if this is true, it may have happened because the various Kiin make money from folks “ranking up” (each “rank-up” certificate costs money).
I say “may have happened” because I haven’t found any articles in Japanese describing this (mind you I didn’t search much).
Personally, having started playing Go in Japan, I think the EGF and AGF ranks are deflated. The idea that a 1d in Japan is 6k EGF makes little sense to me.
It’s like “black belt” in martial arts. I’ve practiced both karate and kendo to a proficient level. 1d is very doable if you put in the hard work. Also, 1d is far from mastery. It just means you have a strong foundation. I’m impressed by 5d karateka, not 1d.
I think it’s a bit of both: a bit of rank inflation in Japan and some rank deflation in Europe / NA.
I’d be interested in knowing more about how AGF and EGF set their levels to be so strict, and why they continue to do so when Asia ranks, where the Go/Weiqi/Baduk populations are greater, has a different distribution of ranks.
Rank deflation hurts the game. The organizations would benefit from making 1d more attainable.
Should we start selling certificates?
Seems difficult to deflate/inflate when all the spectrum is occupied. You want inflate a 6k to 1d . What about the 1d? The 5d?
Seems to be a perilous task when it is already correctly working and adressing everyone from beginners to the top. (But yes 1d seems to be deflated from your interpretation of black belt)
Korea is the tough one, not the western world.
On the first table Goquest are completly wrong putting highest at 2200 which is weak (more like 2500-2750)
There are many inconsistencies so very recommended to read all the comments/warning
1 Like
I think Korean ranks are even stricter than EGF ranks?
2 Likes
It’s a question of changing the distribution, not just saying X + Y.
It works for determining relative strengths, but not from a motivation perspective.
If we just want to have strength comparisons, then why don’t we just use the Glicko calculation?
While I’m not an expert in “goal-setting theory”, the “kyu-dan” ranking could be altered so that it promotes more attainable goal for intermediate / lower-advanced players. That’s where it’s broken imo.
Good point. I hadn’t seen that. It’s great to see the nuance shared there.
Frankly, I would love for folks to look internally and question whether deflation “in the West” (forgive this expression, just looking for something very general) is a problem vs pointing fingers at the East and saying that ranks are “inflated” there.
I’d also be curious to learn more about Korea. I haven’t played on Tygem / WBaduk in a while, but when I used to play, my rank wasn’t very different from Fox.
1 Like