Sunday, 3 April 2011

Some tie-break systems aren't as good as others.

There probably isn't such a thing as a perfect tie-break system. Some work better than others in certain circumstances, but fail in others. However, to at least try and make it work, an organisers should be (a) consistent and (b) publicise it well in advance.

For example, the junior event I directed last week used what is the 'standard' tie-break system for ACT Junior events. In the event of a tie for first (or a title), the two highest players on countback (Bukhholz, Median Bukhholz, Progressive) first compare head to head, and if they either did not play, or drew, play a G/15m single game, with draw odds going to the player with the highest tie-break. While that loads the odds in favour of the player with the best tie-break, it does give the other player a fighting chance of winning the title (as indeed happened last weekend). This system has been in place for a number of years, and is reasonably well know (at least amongst players who have been involved in such ties).

In contrast, the organisers of the European Championship sprung a new tie-break system on the players, one which the majority of players seemed to be unaware of until just before the closing ceremony. The tie-break (to decide qualification spots for the World Cup) had Tournament Performance Rating (TPR) as the first method. However this was modified to drop the lowest (probably good) and highest (probably bad) rated opponents from the calculations (including any points scored). This had the effect of dragging down some TPR's of players who defeated their highest rated opponent (eg Gawain Jones against David Navarra). In the confusion it appears that some players believed they were off to Russia for the World Cup, only to find out later they weren't. I guess the European Chess Union will eventually be the final arbiter of the results.

No comments: