Thursday 30 December 2021

There are no perfect tie-breaks

 The topic of suitable tie-breaks has popped up again, after a 4 way tie in the 2021 World Rapid Championship. The regulations (published beforehand, and used in previous events) specified a playoff in the case of a tie for 1st. The controversy was that only the top 2 players (on tie break) were involved, which in this case were Abdusattorov and Nepomniachtchi, whith Carlsen and Caruana missing out. 

Despite the players being aware of the tie-break system, Carlsen immediately criticised the rules, and was backed up by other players. Pushing back were a number of arbiters/officials who pointed out that the previous tie-break were changed at the request of the players, and had been used in the previous event.

Having done quite a large amount of work in tie-breaks there are a few things worth pointing out

  1. There is no perfect tie-break system, even for similar types of event
  2. No one cares about the tie-break method until they are disadvantaged by it
  3. Everyone thinks that the solution to tie-break issues is to simply use a different tie-break
  4. Not even FIDE can sort this issue, as at least 3 different FIDE Commissions believe they are the responsible body for defining tie-breaks (and I have been a member of 2 of them!)
Due to the issues with tie-breaks, the ACT Chess Association made a decision a very long time ago to avoid them wherever possible. For the ACT Chess Championship (and the Belconnen/Gungahlin Chess Club) the title of champion was simply shared. If there was a necessity to separate the players (eg to see who qualifies for the Australian Championship) then a playoff match (at the same time controls as the championship) was organised.

On the other hand, the ACT Junior Chess League has always used a similar playoff system as the World Rapid, with on exception. The first tie-break is direct encounter, but if that doesn't work (either due to players drawing in the tournament, or a multi-way tie), then a playoff match between the top two finishers (on tie-break) takes place. If it is good enough to decide the ACT Junior Chess Championship, then it is good enough for the World Championship as well!

1 comment:

Garvin said...

My opinion is that tie breaks for first should not be used. As stated, there is no satisfactory method to split ties, especially for first.

Also, in important tournaments such as these, there is no harm in having a four person or so playoff. Will it take a bit longer than the current method, yes, but it is certainly 'fairer' than any system that eliminates one or more players based on a particular tie break method.