Fastest, youngest and mostest

[John Fairbairn]

An update to an earlier report I made somewhere: Yuki Satoshi has now achieved his 1,000th career win and so has become the youngest and the fastest to do this in Japan, beating Cho Chikun's record, and hsi winning ratio is out of sight.

He beat Yata Naoki on 2011-04-11 to reach the milestone. At that point his career record was 1000 wins, 371 losses and 2 jigos, for a winning ratio of 72.9%. At this point he has been a pro for 27 years and 1 month (he became pro at 12) and was ten months shy of his 40th birthday. Cho was 43 years 1 month when he set the previous record.

Cho still has the record for the most career wins - currrently 1365, but in a race with Rin Kaiho, now on 1364.

Fourteen players have now reached 1,000 wins in Japan. Sakata Eio was the first, in 1984. In some ways that the most impressive case as he was playing in a period of much fewer tournaments. As a measure of how much he stood out, it was over ten years before another player reached 1,000. However, Sakata was 64 when he millideified.

Yuki is the third Kansai Ki-in player to the target, after Honda Kunihisa and Hashimoto Shoji. You might expect Hashimoto Utaro to be in there, and indeed he did win 1,082 games out of 1,535 in the GoGoD database, but that includes some handicap games and games not counted as official in his early days. Close but no cigar. BTW Go Seigen didn't even play 1,000 games, but this week we have just found another game missing from his collectanea.

[ethanb]

Sakata was 64 when he millideified.

If Sakata's 1000th win made him 1/1000th of a deity... does that mean 1 million wins as a professional makes somebody a god?

[tchan001]

Sakata was 64 when he millideified.

If Sakata's 1000th win made him 1/1000th of a deity... does that mean 1 million wins as a professional makes somebody a god?

With a lot less than 1,000,000 total professional games collected by GoGOD in their database, the short answer is "yes" because you would have to have a very long life to reach that goal.

[John Fairbairn]

If Sakata's 1000th win made him 1/1000th of a deity... does that mean 1 million wins as a professional makes somebody a god?

Or think millipede rather than millimetre? Actually, a GoGoD milligod, while a joke, is strictly someone of whom we have 1,000 games rather than 1,000 wins, but that's still no joke in the professional world.

[Aeneas]

Or think millipede rather than millimetre?

So "millideified" = blessed with a thousand gods = extremely lucky?

[gowan]

An update to an earlier report I made somewhere: Yuki Satoshi has now achieved his 1,000th career win and so has become the youngest and the fastest to do this in Japan, beating Cho Chikun's record, and hsi winning ratio is out of sight.

He beat Yata Naoki on 2011-04-11 to reach the milestone. At that point his career record was 1000 wins, 371 losses and 2 jigos, for a winning ratio of 72.9%. At this point he has been a pro for 27 years and 1 month (he became pro at 12) and was ten months shy of his 40th birthday. Cho was 43 years 1 month when he set the previous record.

Cho still has the record for the most career wins - currrently 1365, but in a race with Rin Kaiho, now on 1364.

Fourteen players have now reached 1,000 wins in Japan. Sakata Eio was the first, in 1984. In some ways that the most impressive case as he was playing in a period of much fewer tournaments. As a measure of how much he stood out, it was over ten years before another player reached 1,000. However, Sakata was 64 when he millideified.

Yuki is the third Kansai Ki-in player to the target, after Honda Kunihisa and Hashimoto Shoji. You might expect Hashimoto Utaro to be in there, and indeed he did win 1,082 games out of 1,535 in the GoGoD database, but that includes some handicap games and games not counted as official in his early days. Close but no cigar. BTW Go Seigen didn't even play 1,000 games, but this week we have just found another game missing from his collectanea.

I don't know how amateur tournament rating systems work but they have something to do with probability of winning. WOuld Yuki's amazing 72% win ratio mean that he is a rank or more stronger than the 9-dan number would indicate?

[Kirby]

To me, the "stuff" that 9d "indicates" doesn't have an upper bound on strength.

[Laman]

I don't know how amateur tournament rating systems work but they have something to do with probability of winning. WOuld Yuki's amazing 72% win ratio mean that he is a rank or more stronger than the 9-dan number would indicate?

i guess this ratio is not far above other top players' performance. i don't have an access to any database but just a brief look at Mr. Kin's page, most top players have over 60% of wins (in just year, possibly they have even more included games from the time when they were rising to 9d). and by checking EGD (sample player), it seems that similar scores are achieved also by top amateurs

the reason is that top players don't compete only among themselves but play also many weaker opponents, so they collect surprisingly high overall performance

for rating systems, i think Yuki also wouldn't stand out very much, because the higher your rating is, the slower it rises and the faster it can decrease

[hyperpape]

Yuki's numbers are very good, even among 9 dan professionals. As of 2007, Kobayashi Satoru had the best win rate of any Japanese player with 900 wins, and his rate was 68%. Cho U reached 700 wins with a rate of 73%, Takao reached it with 72.5%.

There's a trend towards better percentages as time goes on, but Yuki (38/39) is a bit older than the current crop of top players.

[gowan]

the reason is that top players don't compete only among themselves but play also many weaker opponents, so they collect surprisingly high overall performance

Perhaps so, but I don't think that top title holders play very many tournament games with players ranked below 7-dan. The top players are seeded out of the preliminary rounds where the low dan ranks play. It would be interesting to see what the win percentage of people like Kobayashi Koichi and Cho Chikun was during the years when they held most of the Japanese titles. In the past when the Oteai determined players' ranks some of these star players zoomed up the rank scale losing very few games. In the rating tournament they played only players within a rank or two of themselves.

[John Fairbairn]

Just as a snapshot (not sure it means much), in 1989 the top Japanese players had the following win-loss ratios (note the relatively small number of games):

Kobayashi Koichi 21-14 = 60.0%
Cho Chikun 29-18 = 61.7%
Rin Kaiho 42-20 = 67.7%
Takemiya Masaki 32-21 = 60.4%
Kato Masao 22-25 = 46.8%
Otake Hideo 39-15 = 72.2%
Fujisawa Hideyuki 25-16 = 61.1%
Sakata Eio 15-8 = 65.2%
Ishida Yoshio 26-19 = 57.8%

The first five held all the major titles (Kob had three). Otake won the IBM Cup. The other three are added for interest.

[Kirby]

Is it easy to get the same data aggregated over a longer period of time?

[Matti]

Sakata was 64 when he millideified.

If Sakata's 1000th win made him 1/1000th of a deity... does that mean 1 million wins as a professional makes somebody a god?

Suppose one plays 10+10 minutes lightning games, 40 games a day. It then takes about 68 and half years to play a million games.

[Laman]

Perhaps so, but I don't think that top title holders play very many tournament games with players ranked below 7-dan. The top players are seeded out of the preliminary rounds where the low dan ranks play. It would be interesting to see what the win percentage of people like Kobayashi Koichi and Cho Chikun was during the years when they held most of the Japanese titles. In the past when the Oteai determined players' ranks some of these star players zoomed up the rank scale losing very few games. In the rating tournament they played only players within a rank or two of themselves.

you have a point. but still, if there is (on the top) a group of players, all with positive score, some of them have to get their wins from players outside the top group. i should obtain myself the GoGoD so i could make statements stronger than based purely on logic

actually this is why rating systems should give more information than only win/loss ratios, because win against higher rated player is worth more than against a lower rated player, while ratio may be the same. so we can determine which of the very strong players with impressive win/loss ratios really beats the strongest opponents

PS: not that i would have wanted to undervalue Yuki's performance, he is surely currently one of best Japanese players. i just thought that his win percentage alone might not prove that he is stronger than others (as you wrote)

[Vesa]

Is it easy to get the same data aggregated over a longer period of time?

It is easy, you just need to have Nihon Ki-in's Igo (Kido) Yearbook for each year.

Cheers,
Vesa