eXTReMe Tracker
Front - Tournaments - Search - Rating - Reports - Antichess - Link
  Funny European counterexamples

Funny European counterexamples

09.04.2011 /   Andrejic, Vladica (2170)

How well the people of the European Chess Union set criteria at the European Championship is best seen on some examples that will illustrate here.

04.04.2011 /   Andrejic, Vladica (2170) : Above or below the line
05.04.2011 /   Andrejic, Vladica (2170) : European Championship Scandal

We wrote two reports about recently completed European Championship in Aix Les Bains (22.3-2.4.2011). The first explained that Jakovenko earlier qualified for the World Cup, so it opens a place for Ivanisevic. The second report indicated the absurdity of the criteria (from the first report) qualifing for the World Cup 2011

This time I will explain why the appeal to the European Chess Union makes sense (we expect an official protest of grandmaster Nielsen), although the players did not protest at the beginning of the tournament. At the beginning we first read about what exactly is written in the tournament rules. In Article 8 of the tournament rules ([ECU Tournament Rules B6.2] Ranking) stands:

The order of players that finish with the same number of points shall be determined by application of the following tie-breaking procedures in sequence, proceeding from (a) to (b) to (c) to (d) to the extent required: a) Performance Rating; b) Median-Buchholz 1; c) Buchholz; d) Number of wins. In case of (a) the highest and the lowest rated opponent will be deleted and the maximun rating difference of two players shall be 400 points. In case of unplayed games fot the calculation of (a), (b) and (c) the current FIDE tournaments rules shall be applied.

The question is how these rules should be interpreted. At first sight there are two logical options. First, the one that turned out to be "official", at which they are removed the opponents with the strongest and the weakest rating in the account for an average rating, and their results, too. The second option does not remove the results, but performance rating is calculated on the basis of percentage achieved of the total results.

Let's look at the following experiment. Suppose a player opponents had two players with the same rating, while other opponents have a higher rating. Which one is the opponent with the lowest rating? Since that case is not explained in the regulations, it is logical that calculation does not depend on the choice of one of these two (for example, as is the case with Buchholz coefficient), and it is certainly not the first official option, but the other one. This is not so hypothetical situation, and I'll illustrate it with the following example.

Example 1: Radoslaw Wojtaszek
Grandmaster Radoslaw Wojtaszek in the first two rounds had the weakest opponents by rating. In the first round his opponent was IM Kevin Roser (FRA, 2423) and the game was draw, while in his second round opponent was IM Serghei Vedmediuc (MDA, 2422) and he won that game. Now, suppose a situation in which his opponent from the second round had only one rating point higher, 2423 instead of 2422. Who is now the weakest rated opponent when both are 2423? If you declare that the weakest is the second round opponent, then a victory against him is removed and it is identical to the final standings in which Wojtaszek won a silver medal with the quasi-performance 2826. However, if you declare that the weakest is the first round opponent, then a bad draw against him is removes, and the result is 83% (7.5/9) instead of 78% (7/9), which means an increase on average rating of 2606 to 273 (instead of 220), which gives quasi-performance 2879 and a gold medal, because the champion Potkin has "only" 2849.

This funny example shows that declaring the lowest of the two with the same rating determines whether someone will win the gold or silver medal. Perhaps the initiators of the current system mean that there is no weakest in that case, so nothing is deleted. Who would know? In any case, a logical interpretation of the rules leads to quite different final standings of the tournament. Since all of the top players played all 11 games, it will be the final standings consistent with the column "average" from a table in our previous article. The most important details are that Moiseenko deserves the bronze medal insteadof Polgar, and World Cup spots are for Parligras, Esen, Nielsen and Chparinov (instead of Riazantsev, Lupulescu, Mcshane, and Fridman). We give a few more funny examples of quasi-performance.

Example 2: Alexander Motylev
Ivan Ivanisevic was the opponent of Alexander Motylev in the penultimate round. Both were in a critical group, and the stadings was not in Motylev favor (38 points lower quasi-performance than Ivanisevic). Motylev failed to win this game and it was a draw. In the last round Motylev made a quickly draw with Lupulescu, and Ivanisevic save a draw after big fight against Smirin (who has 38 rating points more than Lupulescu). After that, Motylev's quasi-performance enormously incresed so that Ivanisevic has overtaken in the final standings. How is this possible? Simply, Lupulescu with 2626 was the strongest opponent to Motylev, so the draw is deleted instead of victory over the second strongest L'Ami with 2623. For example, L'Ami unfortunately lost in the Dutch league of Bitalzadeh (2422), and without that he will be the strongest opponent to Motylev, and Ivanisevic would be far ahead.

Example 3: Peter Heine Nielsen
Suppose that, by some miracle, grandmaster Nielsen in the final round had a opponent with rating 1200 instead of Ivan Saric with 2626, and achieved the same result, which is draw. In that case this draw will be deleted, instead of a win over Petrisov with 2385. However, the victory against Petrisov made significant rising of percentage, more than average rating decreases by 2385 instance. This gives a quasi-performance 2717 (instead of the existing 2703), which will be enough to qualify to the World Cup 2011.

Sapienti sat