I'm feeling a bit nervous for Ke Jie's win streak in his current game against Yang Dingxin

But he won.

Go4Go seems not to be up to date, but may be they'll wait until this whole event is over and upload all games at once. I really wanna know Ke Jie's rating right now.
And Ke Jie's latest game, his 18th consecutive win, ended with just 123 moves...

Will play Xie Erhao next in round 6, and may play Tan Xiao in the final round.

And win!

More information: from http://sports.sina.com.cn/go/2017-07-08 ... 1139.shtml

Other players have lost at least two so Ke Jie is already the winner of National Games of PRC regardless of the final round outcome.
But of course people are still looking forward to it as a part of Ke Jie winning streak.

Xie Erhao was looking good midway through, but Ke Jie totally took over in the endgame. As a result, he won the tournament with a game to spare.

The last game will be against Tao Xinran.

Goratings has also updated to include the games from the first 4 rounds.

He just won by resign with black. 20 wins in the raw. It is getting insane!

News report on this http://sports.sina.com.cn/go/2017-07-08 ... 9875.shtml

His next schedule will be 15, 17 and 19 July when he'll likely to face very strong opponents such as Park Jungwhan, Tuo Jiaxi and Chen Yaoye.

While 20 consecutive wins is still less than half of 41 consecutive wins by Lee Changho (and just one third of 60-0 by Master) it's something worth watching very closely. But this happened after AlphaGo match is coincidentally or not is something to be debate.

How fast were the time settings for this tournament?

Fox go server set the room info as 1 hour maintime + 3 periods of 30 second byoyomi

For comparison, in its 60-game series against the Japanese National Squad DeepZen scored 51-9 and its longest winning streak was 17, assuming I've got the games in the right order.

At one point the humans won four in a row. Iyama lost again and it was the younger players who outfoxed DeepZen. It's not yet clear whether they have found a hook to hang new ideas on, but they certainly weren't relying on endgame cock-ups - only 6 games were counted up and the humans won only 1 of those..

Whether Ke Jie has likewise learnt something new is not clear to me either. He doesn't seem to have changed his style. Maybe he's just riding a wave of confidence.

From this interview http://sports.sina.com.cn/go/2017-07-08 ... 7776.shtml
When reporter ask if this is because of AlphaGo Ke Jie said that AlphaGo broaden pro's understand of Go. My impression is that even Ke Jie himself is not sure how much AlphaGo affect him.

Also, I quote your stat on winning streak to reddit forum, but I add that Nihon-Kiin only recognize Sakata Eio achievement (from facebook post) is because Kogishi Soji's achievement was before Nihon-Kiin was found, am I right?

Edit: though on Sakata's winning streak, reddit comment said they didn't count his tie with Go Seigen.

Not sure if someone could confirm the below:

(Unconfirmed) On Baidu Tieba someone mentioned Lee Changho's record is not legit because the Korean Go Association didn't count international games at the time - he actually lost in the third Fujitsu Cup in 1990 (against Satoru Kobayashi by 0.5 points), counting international games, the winning streak is actually 24. For the same reason, Lee Sedol's winning streak is actually 21.

Lee Sedol's interview in 2010 seems like he really think that he had 32 consecutive wins though

forum/viewtopic.php?f=13&t=307

I assume he uses the same convention as the official record. That 24 wins he referred to actually also excludes a game he lost against Lee Changho in a Haiti charity game (given that it wasn't really an official game I'm inclined to agree with the 24 wins personally).

From this source

https://sports.sina.cn/others/qipai/201 ... 883.d.html

1. Master 60 2016-2017 (so they don't count continuous win from different version then)
2. Iyama Yuta 26 2015-2016 (Nihon Kiin says he had 24, so he might win game that Sina recognized but Nihon not, or lose game that Nihon recognized but Sina not)
3. Lee Changho 24 1990
4. Lee Sedol 21 2010
5. Honinbo Shusaku 1842-1847, Go Seigen 1930-1931, Ke Jie 2017 20

There's something very wrong with that list.

For example, in the period quoted for Go's 20 wins (1930-03-19 to 1931-01-01) we have 33 games, and there was no game on 1931-01-01. I assume the latter date comes from a game known only to be in 1931. GoGoD lists that as 1931-00-00 but many people feel the need to add a full date to such year-only games and, hey presto, this one becomes 1931-01-01.

More seriously, in that period Go lost twice - to Murashima and Maeda - and had two jigos.

Shusaku's alleged record is also spurious. Sina gives the period 1842-07-08 to 1847-11-26, which is strictly wrong as these are lunar dates, but sticking with lunar we have Shusaku losing to Ota Yuzo on 1842-07-17 and 1842-07-27.

They have the period 2015-07-27 to 2016-02-01 for Iyama, but as far as I can see on 2015-07-23 he won his previous game (Agon-Kiriyama Cup). Furthermore, in that period we have 32 games. We can presumably exclude the Honinbos Pro-Am game (although handicap games are counted for earlier players. like Go Seigen), and presumably the two "Cool of the Evening Games". However, the Sina list uses the cut-off date 2016-02-01, apparently on the grounds that this is an NHK game he lost (i.e. same sort of time limits as the Cool games). But even that raises another highly questionable point. The NHK game was broadcast on 2016-02-07 but was actually played much earlier, on 2015-12-21. I can't see any obvious reason for excluding any other games, except for one more NHK game.

If unofficial games are excluded (and that's just a modern concept anyway), presumably we have to exclude Ke Jie's five games in the Go Rally. The time limits were very reasonable, but it was a small invitation event and the usual criterion for an official game is that the event has to be open to all that organisation's players (one reason for excluding international events).

Nevertheless, all the achievements listed are hugely impressive. Historically tongues have started seriously clucking in admiration whenever a player got a streak into double figures, and such streaks do seem to be far less than once-a-year events.

Does anyone have a list of the players and their ranks (not great as a measurement of strength but it'll do for those players whose name one doesn't recognise) beaten by Ke, the Lees, Iyama, Sakata etc? The thing that particularly impresses me about Ke's current run is the high caliber of the opponents he's beating with lots of top 10 in the world pros. It sounds like Lee Changho's may be less impressive if he was beating only Koreans (who had a few of the world top 10 in 1990 in Cho, Seo, Yu but mostly in Japan plus Nie/Ma in China). Sakata back in the 60s playing only Japanese doesn't detract as they were sole top dogs and I expect quite a few games were title matches against other top players of the day. Iyama beating up the other Japanese these days isn't so impressive as they are not top 10 (though just maybe could be in 8 hour title match games as other nationalities don't do them). Lee Sedol's could maybe have been against similarly tough opposition.

Here's Ke Jie's list:
1. Weon Seongjin 9p
2. Kang Dongyun 9p
3. Chen Yaoye 9p
4. Gu Li 9p
5. Chen Yaoye 9p
6. Gu Li 9p
7. Tuo Jiaxi 9p
8. An Dongxu 6p
9. Kang Yootaek 8p
10. Huang Jingyuan 3p
11. Gu Zihao 5p
12. Guo Wenchao 5p
13. Tang Weixing 9p
14. Li Zhe 6p
15. Chen Yaoye 9p
16. Fan Tingyu 9p
17. Tang Weixing 9p
18. Yang Dingxin 5p
19. Xie Erhao 5p
20. Tao Xinran 6p

Every player can only play the hand dealt to him, and Ke is still going, of course, but personally I'd rate Yi Se-tol as having had the toughest hand so far. There were some make-weight games but a big proportion of his run came in world championships (BC Card, Fujitsu, Chunlan) and he had to face world-champion luminaries such as Chang Hao, Pak Yeong-hun, Kong Jie and Zhou Junxun. And let's not forget he's also the only player to have beaten AlphaGo.

Iyama's run was heavily loaded with title-match games but was all domestic. Yi Ch'ang-ho's run was almost entirely domestic and inclined to lower grade competition.

Sakata shone in the days of leagues when all games had to be against top-calibre competition. Modern players get the benefit of more early-round games against rabbits.

All of this must be seen against a background of chalk, cheese, apples and pears, of course, and was Barry Bonds a better slugger than Babe Ruth? Some of the runs here may need to be asterisked likewise.

According to Nihon Ki-in, Sakata won 29 consecutive games.
According to my data these wins (between 1963-10-11 and 1964-07-29) were against

Takagawa Shukaku (9p) - 11 games
Fujisawa Hosai (9p) - 5 games
Shimamura Toshihiro (9p) - 2 games
Fujisawa Shuko (9p) - 2 games
Ohira Shuzo (9p) - 2 games
Handa Dogen (9p)
Hashimoto Shoji (9p)
Sato Sunao (9p)
Kada Katsuji (8p)
Okubo Ichigen (7p)
Rin Kaiho (7p)
Otake Hideo (6p)

But again we have to exclude unofficial games (which boosts the records for some other players). Sakata lost in mid-run against Handa Dogen in a high-profile East-West Japan telephone match billed as Judan versus Meijin.

If we do exclude such a high-level game, surely we'd have to exclude Ke Jie's rally games?