2014 Bilbao Masters - Round 6

Aronian breached Anand's defenses today, winning a nice game with White. If we use the traditional scoring (win = 1, draw = 0.5, loss = 0), then they tied for first place. But the tournament uses a different method that has gained some traction over the last few years. With wins yielding 3 points and draws 1 point, Anand is the clear winner. In this case, it functions the same way as the tiebreak system that awards first place according to the most wins. But when two players are tied, then giving the prize to the one with the most wins is the same as giving it to the one with the most losses - rather nonsensical. Even worse, in the 3-1-0 system, it is possible for someone who is trailing according in the traditional score to win the tournament. It does have the advantage of discouraging quick draws, but there are better ways to do this. A while ago I wrote an essay on this, but it was never published.

I've been making tournament forecasts for about a year by now, so sometime soon would be a good time to review the predictions and see how well they matched up with reality. Also of interest would be the effect of the new draw rate model on the forecasts. It should lead to marginally better results, but the differences will probably be imperceptible. Unfortunately my ambition for new projects greatly exceeds my spare time; I don't know when I'll get around to testing if 1.e4 or 1.d4 has a higher draw rate.

2014 Bilbao Masters - Round 5

Today featured a pair of quiet draws, so Anand clinched at least a tie for first place. Aronian can only catch up if he wins his game against Anand tomorrow; according to the model, there is a 24% chance of that happening.

2014 Bilbao Masters - Round 4

The Vish wins again! His impressive 3.5/4 places him a full point above his nearest competitor, Aronian.

Because of ties, the first column of probabilities sums to more than 100%.

2014 Bilbao Masters - Round 3

Aronian and Anand drew a rather uneventful game; Anand's chances continued to improve since he maintained his lead.

The chances sum to more than 100% in the first column and less than 100% in the second column since multiple players can tie for first.

Importantly, these winning chances are based on the traditional scoring system: 1 point for a win, 0.5 for a draw, and 0 for a loss. The prizes are awarded based on the alternative method that gives 3 points for a win, 1 for a draw, and 0 for a loss.

2014 Bilbao Masters - Round 2

The favorites won today, so their chances improved in the forecast:

Since multiple players can tie for first, the first column adds up to more than 100% and the second column to less than 100%.

2014 Bilbao Masters - Round 1

Anand won a nice game, taking the lead as Aronian and Vallejo drew.

Since a tie for first is possible, the first column sums to more than 100%.

2014 Bilbao Masters

Lest any post-Sinquefield Cup withdrawal set in, we now have the Bilbao Masters coming up. It will be a 4 player double round robin starring Aronian, Anand, Ponomariov, and Vallejo. Anand and Aronian are certainly the favorites, but don't be shocked if Ponomariov or Vallejo wins. The player to watch? Anand. His victory in the Candidates Tournament was very impressive; in 14 games against the best in the world, he was never in any serious danger of losing. He hasn't played in any tournaments at classical time controls since then, so we will want to see if he can continue to perform at this level. If he can, Carlsen will have a much more difficult task in their rematch for the World Championship. 

Almost all of the remaining elite players will be participating in the European Chess Club Cup. That will also be held in Bilbao, but I don't have any predictions for it; as discussed in my post about the Olympiad, team events pose some special difficulties for forecasting. 

As usual, the fact that players can tie for first implies that the numbers in the first column sum to more than 100%.

2014 Sinquefield Cup - Round 7

Once Vachier-Lagrave played Rf1, giving up castling, I realized that Caruana was going to reach 7 - 0. Amazing. There is just a 0.19% chance that Carlsen will catch up with Caruana and tie for first; that would involve Caruana losing his next 3 games while Carlsen wins every remaining game.

Before the tournament, if you asked me how likely Caruana's 7 - 0 start would be, I would have answered 0.003%. This probably occurred only once in the 30,000 simulations I ran in order to generate the original forecast. According to the model, Caruana has a 14% chance of defeating Carlsen tomorrow, so the chance of (dare we contemplate it?) 8 - 0 would have been about 0.00004%. Chess history is unfolding before our eyes.

2014 Sinquefield Cup - Round 6

Thanks to another Caruana victory, the forecast is not very exciting - realistically, there is no chance that anyone will overtake him in the standings. It feels like we are watching chess history firsthand - that the 2014 Sinquefield Cup will be remembered in the same way as 1994 Linares, Fischer - Larsen, Fischer - Taimanov, 2005 San Luis, and other fantastic performances.

A more interesting question was brought up in the live commentary: before the tournament, what was the probability that Caruana would start 6 - 0 ? According to my model, the answer is 0.011%, which is about 1 in 9000. Even if we expand the question to "what is the chance that one player - not necessarily Caruana - will begin with 6 - 0?", the chances still have to be below 0.1%.