The tiebreaks at the Candidates Tournament are ludicrous, although I’m surprised to admit that it’s not as ludicrous at it seems at first glance.
It’s absurd to determine who will challenge for the World Championship due to Most Wins, or any other tiebreak system in a Round Robin.
The only reasonable way to break a tie in an event of this magnitude is to play a tiebreak match for it.
I could easily end the blog here, and the majority of the chess playing community would agree with me. However it’s not quite as simple as it seems.
What does it mean that there is no tiebreak after tomorrow’s games? It means that these are going to be the most intense classical chess games that you’ve ever seen
Karjakin, Mamedyarov and Caruana are going to be fighting like wild animals in a classical game, because there is no chance for any of them to coast to a tie and aim to win a tiebreak match. This means that the audience is going to be treated to the most dramatic day of chess that we’ve seen since at least 2016. If a tie resulted in a tiebreaker match, you would be likely to see more conservative play among the leaders at almost every point throughout the tournament.
The key point is: By removing any tiebreak match, and instead using a tiebreaker, even if it is a flawed tiebreaker, the classical chess play becomes much more exciting.
This is a win for fans, but at the same time we are talking about a very serious topic. We are talking about who will go on to challenge Magnus Carlsen for the World Chess Championship. It’s not enough that it’s exciting, but it also has to be both fair and logical. It is fair as all players play by the same rules, but it is not logical. There is no real reason why someone with more wins should qualify over someone who has the same score. There is certainly no logical reason why a Sonnenborn-Berger score should have any effect at all.
Fortunately I have the perfect solution, and one that I believe has been proposed before, but absolutely should be a staple for all future Candidates Tournaments.
The tiebreak should take place before the first round!
The above system is used in some tournaments to determine the draw, such as who gets the extra white and which pairing numbers and etc. But in this case it should be used as the tiebreak.
For example, you could have a one day round robin tournament with a time control of 15+2, or you could even make it a double round robin that takes place over two days. The winner of this event wins on all ties, the second place finisher wins on all ties against lower placed players and etc. In fact with this format, the entire field would effectively start the Candidates a half point behind the winner of the rapid tiebreak tournament.
This means that everyone will be playing catch up from the very start, resulting in a clear cut standing in every round. Whenever there is a tie at the top of the crosstable, you will always know who is ahead based on their performance in the rapid tournament. It’s simple for fans and adds another day or two of exciting chess for everyone to enjoy.
This is a much better system than the current one because:
- It ensures the classical chess that takes place is maximally exciting, because seven of the eight players will always be clearly behind the leader.
- It’s fair to all players
- It’s logical, unlike our current system. The players who win on tiebreaks will have clearly earned the right to their victory.
This is the second most important event in chess and the idea of what to do on a tie needs to be taken more seriously, instead of the lazy solution that is currently being used.
Greg Shahade 
I die a little inside anytime I see a Sonnenborn-Berger being applied to a round robin
LikeLike
This has indeed been proposed before, though I don’t remember where, and as soon as I heard of it, it was immediately clear to me that this is the obviously correct way to implement tiebreaks. I have no idea why it hasn’t become common.
LikeLike
I do not like this tie break system. First, you are using a rapid game for a tie breaker for a classical tournament. These are two different types of chess. Why not use a coin flip, a game of checkers, or a game of paper, rock and scissors to be the tie break. Makes as much sense.
Second, it is not fair. As you said, every player is behind the leader, even if they have the same score. How is that fair?
A fair system would be a one game playoff. Winner breaks the tie. If a forced draw, the player with the Black pieces wins. No agreed draws.
LikeLike
I think the tiebreak rules for the Candidates was perfect. No blitz/rapid games determining places for a Classical chess tournament. Head-to-head was the first tiebreaker, then most wins, then, and only then, does Sonnenborn-Berger come into play. What’s wrong with encouraging players to play for wins?
Maybe instead of S-B, the third tiebreak should be average rating over the last two years (since that is the time between World Championships now)?
-Matt
LikeLike
Whole tournament is unfair really. You win when your opponent blunders and draw when he doesn’t so it’s all luck as to who gets the most blundering opponents. Why slow chess round-robin is so flawed. Ding Liren was the only undefeated player, 1 win and all the rest draws.
Bobby Fischer had it right when he requested a series of knockout matches. Surprised though he felt the Russians were arranging quick draws with each other rather than throwing games to the one they wanted to win whilst playing steady chess against him to get draws instead.
Speed up classic chess time control to 30|45 though. That is still classic – anything where the numbers add to more than 60 is classic and those add to 75.
Head to head, 2 games a day. At least 6 games, then pairs of 2 until a winner. Switch order of who gets white first so ABBA style. After a certain number of games maybe switch to rapid, 15|30 maybe but no blitz or bullet please and certainly no armageddon.
Incidentally: for the reasons above I stated Magnus Carlsen may well not win a Swiss tournament but give him a series of knockout matches and he’d beat all-comers.
LikeLike