Another “classical” tournament to be decided by Rapid/Blitz playoff. As far as I’m concerned Ding has won. He’s ahead of Carlsen on both tiebreak and performance rating.
Perhaps if promoters and FIDE stop having playoffs, the players will try harder to win games knowing that they may lose via TB system.
Both of which are effectively meaningless in a RR.
But in many European tournaments Sonneborn-Berger tiebreaks are used to determine final placement and prizes, both in RRs and open events. All tiebreaks have a component of unfairness. Playoff games to determine a winner are “fairer” but the use of blitz and Armageddon games to decide seem anti-climactic, especially when Carlsen has proven so good at them.
But Solkoff or Modified Median can at least be claimed to measure which player had the tougher schedule—each has situations where they can be unfair, but usually get things in at least a reasonable order. The only real “merit” that S-B has is that it can be applied in a RR. Most Blacks is probably a fairer tie break, particularly for super GM’s.
Modified median is fine for a twelve round event, but for a 4 round tournament that gives you only two data points and the likelihood of still having ties. My preference for small Swiss events is Solkoff, cumulative, head to head, and then a coin toss. There is no time to run playoffs; everyone wants to go home early anyway. The custodian wants us out so he can go home, too.
Despite the criticism of a few online vultures who are looking for blood, there is nothing wrong with a tie for first and a split of the money after a hard fought tournament. The artificial means used to give away first prize do not show that any of the tied players is better than the other(s) in that particular tournament. Carlsen knows he outclasses his fellow players at the tiebreak schemes.
For a Swiss, maybe, but none of those systems (Solkoff, median, modified median) work for a round robin, do they? All players with the same score would have the same tiebreaks, too. And those with higher scores would have lower tiebreaks, and vice versa.
Sonnenborn-Berger breaks ties in a round robin, but it rewards inconsistency. A player who defeats a high-scoring opponent and loses to a low-scoring opponent has better S-B than one who does the opposite.
Bill Smythe
“Modified” median only tosses high and low for players at even score. For players with plus score it only drops the low score. (Not-modified median is what throws out high and low for everyone). I agree that MM may be a bit iffy at four rounds, but I would still prefer MM followed by Solkoff even for that.
Right. It has about a much merit for a tie breaking system as adding up the number of letters in the names of the players you beat.
For an RR, the following would make sense as the first three breaks (as actually indicating some merit in the result):
Head-to-head
Most blacks
“Solkoff” only of players against whom you played Black
The third one would probably end up eventually breaking almost all ties.
Then use Solkoff or Modified Median. Better than BS Rapid and Blitz after playing classical games, either Swiss or round robin. In this case Ding wins the playoff, but I certainly don’t care to even look at the playoff games. Your view may differ which is fine.
Another option is to declare joint or multiple co-champions and split the prize money between them. IMO, that’s better than a Rapid/Blitz playoff.
Ding-A-Ling Liren Shares Sinquefield Cup First Place With Magnus
xpertchesslessons.wordpress.com … th-magnus/
Unlike other games in which lucre is the end and aim, [chess] recommends itself to the wise by the fact that its mimic battles are fought for no prize but honour. It is eminently and emphatically the philosopher’s game. – Paul Morphy
Have them declared joint winners and split the prizes equally.
Excuse me. YOU were the one who brought up (bad) tiebreaks and said “As far as I’m concerned Ding has won” on their basis.
The whole point is that Solkoff, Modified Median and Median don’t apply to RR’s since they would give the same value to everyone in a tie. S-B gives different values for tied players (maybe), but does not really differentiate on anything that is meritorious.
Using Solkoff, Median or Modified Median in a round robin does not break any ties (or at least no ties between players who played the complete schedule).
Cumulative may break ties between players who had a different win/loss distribution (winning round 1 and drawing round 2 is better than winning round 2 and drawing round 1 - even if one of the two tied players played A in rd 1 and B in rd 2 while the other played B in rd 1 and A in round 2 - resulting in both of them having drawn the same player and beaten the same player). The justification for cumulative is that winning earlier means you get tougher players later and thus have a tougher schedule than a player who did not win as much earlier - an argument that is inapplicable to a round robin.
Three-way ties might not be broken (first, second and third pairing seeds split with each other and sweep the rest of the field ). It is probably a valid statement since you used “almost”.
Except it would only fail to break the tie if they all had the same number of blacks, and they each had one white and one black in the games among the three of them. With a random draw, the probability of all that isn’t high.
In the RR-tables for 7 or more players, the first, second and third pairing seeds/pairing numbers (not the same as the three strongest players) have 1-2, 2-3 and 3-1 pairings and all have one more white than black.
As I said “almost all” keeps it a valid statement. If you had gone over the top and just said “all” then the over-generalized statement would be refuted for some specific special cases.
Sorry. I thought you meant the top three players (since splitting with each other and sweeping the rest would seem reasonable only in that case). The probability that the three strongest players would randomly draw numbers that would give them an equal number of Blacks with BW among the three is a bit less than 1/8 in a fairly large field (a lot less in six player field).
I agree strongly with Mr. Lafferty. Indivisible prizes should be limited as much as possible.
Alex Relyea