ICGA WCCC and WCSC in Santiago de Compostela

[AlexChess]

The point isthat Stockfish is only strong on average, because that is what Elo is, an average. It is not reliably strong. Therefore it has a significant probability to loose, in a small number of games. And of course the Elo advantage on the competitors is so small that you would need many thousands of games to see it as an average.

The Stockfish developers don't like that. So they only want to participate in tournaments that measures average performance over an insanely large number of games. Because they know they only stand little chance to win a tournament over 8 games.

Then tell me what your 8-game tournament is supposed to measure? CPU noise?

I agree with Anthelion, 8 games aren't enough. Anyway there is no doubt that Stockfish is #1 even not participating to the WCCC. It's SPCC to STATISTICALLY confirm that, with 15.000 games the error margin is almost null and right now Torch 2 is 35 ELO weaker, Fritz 18 is 3157 according to FGRL and Fritz 19 2023 WCCC is not even considered by the rating lists.

[Graham Banks]

Ginkgo Valencia = Fritz 19

[hgm]

I agree with Anthelion, 8 games aren't enough. Anyway there is no doubt that Stockfish is #1 even not participating to the WCCC. It's SPCC to STATISTICALLY confirm that, with 15.000 games the error margin is almost null and right now Torch 2 is 35 ELO weaker, Fritz 18 is 3157 according to FGRL and Fritz 19 2023 WCCC is not even considered by the rating lists.

Better in what? Winning 15,000-game matches? Who would ever want play such a match?

Like in any sport, to win a championship you need to perform the few times when it counts. Stockfish cannot do that. When you play in the finals for the world cup soccer you'd better score goals. Whining afterwards that you are number one on the topscorers list for last season won't do any good.

[mar]

Then tell me what your 8-game tournament is supposed to measure? CPU noise?

the point of a tournament is not to measure anything, but to decide standings of an event, that's the beauty of it

[Modern Times]

the point of a tournament is not to measure anything, but to decide standings of an event, that's the beauty of it

Yes, and it is how you perform on the day. Very much like football. You might have a bad day and lose to someone you should probably have beaten. You only get one chance. You don't get to replay the game 100 times !

[Viz]

WCCC analogue to football will be to play 1 minute games so any team that is somewhat competent will draw every other team.
And then go to tie breakers which feature 1 penalty shot from only 1 side - if they score they win, if they don't score they lose.
Would be a great tournament with a lot of meaning indeed.

[Graham Banks]

WCCC analogue to football will be to play 1 minute games so any team that is somewhat competent will draw every other team.
And then go to tie breakers which feature 1 penalty shot from only 1 side - if they score they win, if they don't score they lose.
Would be a great tournament with a lot of meaning indeed.

It's no different to any other tournament as far as what could happen.

[hgm]

WCCC analogue to football will be to play 1 minute games so any team that is somewhat competent will draw every other team.
And then go to tie breakers which feature 1 penalty shot from only 1 side - if they score they win, if they don't score they lose.
Would be a great tournament with a lot of meaning indeed.

Even 'grossly exaggerated' is too weak a qualification for that. Even at standard time control Soccer can be just as boring as Chess, when both teams are bent on 'keeping the zero'. To win a tournament you have to be prepared to take a certain risk. If all participants have a 100% probability to draw, a strategy with a 2% win probability and a 3% loss probability, which on average is weaker, offers a much better chance for winning an 8-player round-robin:

If you also play the 100% draw strategy, and the tie breaker is basically a lottery, you only have 12.5% chance to win. The 2-95-3 strategy, however, still has 0.95^7 = 69.8 chance for all draws, so that it reduces your chances to win by tie break to 8.7%. The chances that you achieve one win and 6 draws are 7*0.95^6*0.02 = 10.3%, so that these two potential outcomes alone already give you a 19% chance for winning.

It is not the fault of the organizers when the participants prefer to play boring Chess. It is just like with Houdart, who explicitly stated that he would categorically refuse to participate with Houdini in any tournament that was not rigged to guarantee him the win.

[mar]

The computer community nowadays is dominated by people that are only interested in destroying others, and under those conditions a cozy get-together of programmers is not a format that attracts many participants.

I have the opposite feeling - the days when others were actually destroyed (Vas) are gone,
nowadays people seem to cooperate more while still competing in a friendly way.

[hgm]

Well, I don't see much friendliness here.