# BCcard Cup World Baduk Championship

__Keywords__: Tournament

Korean international tournament created in 2009 and sponsored by Beyond Card. In the first edition, ten players were seeded (Lee Changho, Lee Sedol, Kang Dongyun, Cho Chikun, Iyama Yuta, Gu Li, Chang Hao, Zhou Junxun, Cho Hunhyun and Won Sungjin), 54 players come from preliminaries held on 23, 24 and 25 February 2009. Play conditions are: 1 hour per player + 30" byo-yomi; komi is 6.5 points.

The tournament starts on February, 28th and the final will be a best of five match (from 1 to 5 May).

Korean announcement and results.

Ed. | Date | Winner | Runner-up | Result |
---|---|---|---|---|

1st | 2009-04-04 | Gu Li | Cho Hanseung | 3-1 |

2nd | 2010-04-27 | Yi Se-tol | Chang Hao | 3-0 |

3rd | 2011-04-28 | Yi Se-tol | Gu Li | 3-2 |

4th | 2012-05-16 | Baek Hongsuk | Dang Yifei | 3-1 |

The tournament was discontinued after four editions.

## General discussion

valerio: Over 3000 Korean amateurs participated in various tournaments and 20 of them qualified to the preliminaries (20 amateurs and 257 pros, no western players).

tapir: Anything (more) about the early preliminaries? Non-korean amateur players etc.?

valerio 2nd edition will start 2010-01-11

## Ama vs Pro discussion

tapir: In the preliminaries there were 216 + 40 players winning overall 40 + 108 + 54 games, that is 202/256 (=0,79) per capita while the 20 amateurs in the preliminaries won 18 games. That is 18/20 (=0,9) per capita. That is overall the amateurs played better than we might expect from a random sample of 20 participants. :) Please add this to Amateurs versus Professionals!

xela: Sorry if this sounds like a dumb question, but I don't understand these statistics. A game of go involves two people; normally one person wins and the other loses. So how can the winning rate per capita, averaged over all players in the tournament, be anything other than 0.5?

Herman: Because per *capita* is something else than per game. There were 256 participants, who played 226 games in the preliminaries. But I don't think that it is actually meaningful. For example:

If someone wins the Honinbo title 4-1, then there were 5 victories in the final, for an average of 2.5 victories per capita. But if someone else wins the Honinbo title 4-3, there were 3.5 victories per capita in that final, so the players in the second final scored better per capita than those in the first. Were they therefore better players?

So I think the meaningful statistic would be: How many games did the amateurs play, and how many of those did they win?

valerio: The 20 Amateurs played 33 games, winning 18 of them. I thik this is the right count. The ratio is 54,54: good aniway.

tapir: In a knockout tournament system the number of wins per player is significant (the player with the most won games wins the tournament, and the honinbo with four wins is surely better than the runner up whether 3 or 1 win, it's ordinal scale), that's why there are rating tables by won games and by winning percentages etc. (54,54% is better than overall 50% winning rate as well, wins per player is just another way to express the same) The point is: the selected amas did well in the preliminary, better than average pros :)

Herman: Yes, number of wins is significant in a pure knockout. But this was not pure, because 80 of the 256 players had to play a preliminary round, with the 40 winners joining the remaining 176 players in the real knockout (216 -> 108 -> 54). Which means that two players that both have two wins have not had an equal performance if one of them played in the prelim. Of the 20 amateurs, 3 played in the prelim, with 2 going through to the final 216. From there, in the pure knockout, the 19 remaining amateurs played 30 games and won 16 (53.3%), while the 197 remaining professionals played 162 games, winning 65 (40.1%). I'm not disputing that the amateurs scored better than the professionals, but I think numbers like 53% and 40% gives a better idea of performance than 0.79 and 0.9 per capita, because per capita figures depend on the number of rounds played.

- I doubt that these calculations are correct, in a pro vs. pro game there should be at least one pro win :) (132 + 14)/(264 + 30), counting wins per player for a certain population is way easier, that said. we all agree that statistically the best amateur players are on average better than pros. (they still lose the tournament to the best pros, but that doesn't matter here :)

Herman: D'oh! I had a feeling I was doing something wrong! :-( Disregard what I said :-)

Herman: Some more statistics:

- Of the 256 participants, 80 had to play a preliminary round (31%).
- Of the 20 amateurs, only 3 had to play a preliminary round (15%).
- Of the 256 participants, 54 got through to the final 64 (21%)
- Of the 20 amateurs, 5 got through to the final 64 (25%)
- Only counting from the last 216, 54 got though (25%), with 5 Of the 19 remaining amateurs (26%)

So on all counts, the amateurs did slightly better :)