MLily main tournament

[trout]

Throw out stats when it is single game match. At this level of game, anything can happen.

[emeraldemon]

Semifinals!

Ke Jie (#1, 3566) vs Park Yeonghun (#17, 3443):
That's right, Ke Jie is now ranked #1, the best go player in in the world, one single elo point above Park Jungwhan!
Park Yeonghun also moved up a few spots and is currently at his highest-ever rating, surpassing his previous peak way back in 2005.
Ke Jie should win 67% of the time according to WHR. They have played each other twice, in the 2013 Chinese A league. Ke Jie won both times.

Lee Sedol (#3, 3530) vs An Sungjoon (#40, 3356):
WHR predicts Lee Sedol to win 73% of the time. They have played 4 times so far, and the record stands 4-0 Lee Sedol.
An Sungjoon is 6p, I think he will be promoted if he wins this game, since he will be a finalist for an international title.

If these matches go as WHR predicts, Ke Jie and Lee Sedol will face each other for the first time in the MLily final, with WHR giving a scant 55% edge to Ke Jie.

As an aside, if you want to evaluate the accuracy of probabilistic predictions like these, check out Brier Score and friends.


Edit: Just noticed that the semifinals are best-of-3 matches. That means (according to the model) Ke Jie and Lee Sedol have 75% and 82% chance of advancing, respectively.

[Uberdude]

Given Lee's greater experience of multi-game finals* and ability to use the extra time well, I would give him a slight edge in the final. Ke Jie is a hot potato though.

* I think part of this experience will be the ability to learn from mistakes that cause a loss in an early game thus making each game not an independent event with the same probability of win.

[Uberdude]

I didn't mean to seem insulting. There does seem to be common misunderstanding about expected results. So I was just trying to point out that if it had been taken all the way would have become clear to people that contrary to the expectations of most, those differences in rank were not enough to make the most likely single outcome other than "pure guess".

Here's what I get for all the cases:
4/4 correct: 17%
3/4 correct: 40%
2/4 correct: 32%
1/4 correct: 10%
0/4 correct: 1%

So those win rates were enough to make the most likely single outcome 3/4 not the 2/4 of guesswork (thanks Ke Jie). That's not really so surprising after you already have 47% (thanks for spotting the mistake yoyoma) used up, though before I'd calculated those percentages I would have said 'pass' if you asked me whether 3/4 or 2/4 correct was more likely as I'm well aware binomial coefficients balloon in the middle, whereas for 4 80/20 matchups I would be pretty confident to say 2/4 would be unlikely by intuition, and yoyoma's calculations show that's the case, with 60/40 being the turning point at which 3/4 overtakes 2/4.

[trout]

Reminder!!!!!!!!!!!!!

Semi final - 11/22/15, 11/24/15, 11/25/15
Lee Sedol vs Ahn Sungjoon
Park Younghoon vs Ke Jie

[hyperpape]

In the meantime, the ratings have changed a good bit:

Ke Jie - 3601 (+ 35)
Park Yeonghun - 3473 (+ 30)
Lee Sedol - 3490 (-40)
An Sungjoon - 3366 (+ 13)

The same players are the favorites, but Lee didn't do so well through October and November, while Ke Jie is still on fire.

WHR and/or the goratings database makes kibbitzing about ratings hard: the website now says that Ke Jie hit 3565 on 2014-08-03 and has been steadily rising since.

[trout]

Semi final - 11/22/15 1st game of 3;
Lee Sedol defeated Ahn Sungjoon by resign.
Ke Jie defeated Park Younghoon by resign.

2nd game - 11/24/15

[seigenblues]

FYI, Myungwan Kim will be doing live commentary of game #2 between Ke Jie and Park Younghoon tonight, starting at 7pm PST, you can watch live at http://youtube.com/c/USGOWeb/live or http://twitch.tv/usgoweb.

As a bonus, he'll review game #1 during the lunch break, between 8:30pm and 9:30pm PST.

See you there!

[trout]

Semi final - 11/24/15 2nd game of 3;


Park Younghoon defeated Ke Jie by resign.
Park Younghoon vs Ke Jie - 1:1


Lee Sedol defeated Ahn Sungjoon by resign.
Lee Sedol vs Ahn Sungjoon - 2:0
Lee Sedol advances to Final!!!!!!!!!!!!


3rd game:11/25/15

[Uberdude]

Park's first game with Ke was interesting as the opening was similar to Ke's vs Lee Sedol recently, see viewtopic.php?p=195684#p195684. Did Myungwan Kim comment on this?

[trout]

Yes, he did.

https://www.youtube.com/watch?v=HLVDTsNPDBE

[trout]

3rd game:11/25/15
Ke Jie defeated Park Younghoon by resign.
Ke Jie advances to Final!!!!!!!!!!


Final(5 games): Ke Jie vs Lee Sedol - 12/30/15, 12/31/15, rest of games will be played in new year.

[Lucian]

This should be an exciting final, like a mini-jubango, considering their previous meeting. I just saw in this article:
http://www.wbaduk.com/news/news_view.as ... index_div=
the following statement by Ke Jie:

-How do you expect the final against Lee SeDol 9P?
Lee SeDol 9P said his winning rate is 50%, but I think it is just 5%. His time is gone.

Is this translation accurate? It sounds astonishing to me. Maybe Macelee or Trout can make some light about this interview, using the primary sources

[wolfking]

-How do you expect the final against Lee SeDol 9P?
Lee SeDol 9P said his winning rate is 50%, but I think it is just 5%. His time is gone.

Is this translation accurate? It sounds astonishing to me. Maybe Macelee or Trout can make some light about this interview, using the primary sources

It is more or less accurate, although it did not quite convey the jokingly manner. Ke Jie was asked about his outlook for the final. His answer was: I've heard Lee Sedol said his winning odds are "5成" (Chinese usually describes the odds in a scale of 10 but omits the 10), I think it's "5成" but maybe the total is 100成. So basically he's joking "Yeah I agree with him that his winning share is 5 (if the total is 100)". He did finish by saying "I think it's about time for the legend to make his curtain call." So he definitely expressed extreme confidence in this brief interview right after the win against Park.

[macelee]

I just watched the video record of that interview. Ke Jie did say that Lee Sedol's chance is 5 out of scale of 100. He immediately laughed and said something incomprehensible. It is apparent to me that was a joke. As suggested by wolfking Ke Jie invented the term '100成' which doesn't exist in Chinese language. He did show his respect to Lee Sedol by implying Lee's career being 'legendary'.