I wrote the first version of this plan for a world chess championship in February 2005. This document was not for publication but was circulated amongst a small group of players and officials. At the end of the Linares tournament Frederic Friedel discussed it with FIDE deputy president Georgios Makropoulos and vice president Zurab Azmaiparashvili. Since then, FIDE has announced the forthcoming World Championship to be held in Argentina during September-October 2005. I cannot say whether my ideas had an influence, but this event bears a remarkable similarity to my suggestion! However, FIDE have not (yet!) taken on board my ideas for a qualification system for future World Championships. This plan may be of some interest to the chess public, so I give it below.
In a further development, a meeting has recently taken place between FIDE and the ACP, and one of the suggestions made in that meeting has a bearing on my ideas. I will comment on this at the end.
First of all, here is what I wrote in February, slightly abridged:
From 1948-1990 the World Chess Championship was organised by FIDE (Fédération Internationale des Échecs), the International Chess Federation. During this period, the FIDE World Championship was universally recognised as the legitimate world chess championship. However, in 1993 a breakaway world championship was founded, and since then the chess world has been divided. FIDE has continued to organise world championships, although using a quite different system to that employed previously. Their system has resulted in a series of title-holders not ranked in the world top 10, seriously undermining the legitimacy of the FIDE World Championship title.
In 2002, an agreement was signed in Prague to reunify the World Championship, signed by most of the leading players and chess organisations. However, this agreement was under pressure from the start and has now collapsed completely. One of the main reasons for the collapse of the Prague agreement was that it involved a complex series of matches stretching over a considerable span of time. While some of these events were organised, others could not be, and since all the events were connected, the failure of just one of them led the whole structure to collapse.
A World Championship should be organised once every two years. It would take place during what is currently an inactive part of the chess season. September would be a good choice, but there are other possibilities. This World Championship would consist of a single event, an eight-player tournament in which each player meets each other player twice. By condensing the World Championship into a single event, the problems which have plagued previous structures are avoided, and sponsorship would not be diluted by having the publicity shared amongst several events.
For the first event, the players should be selected on the basis of Elo ratings. The figure used would be the average rating of players over the preceding 12 months, weighted according to the number of games played in each period.
An obvious problem with a qualification system based on Elo ratings is that once a player has achieved a high rating, he may choose not to play further in order to ensure qualification. It seems to me that if someone wishes to play for the world championship, then his qualification should be based on actually playing rather than sitting at home, so for subsequent world championships, the figure used would be the Elo rating (as calculated in the previous paragraph) plus an activity bonus.
The activity bonus would be the total number of rated games played since the previous world championship divided by two. The activity bonus would be limited to 75 points. Thus a player who has been active recently would be ranked higher than an inactive one, even if their Elo ratings were the same. This system stimulates chess activity in general, and it particularly encourages young and improving players who might qualify by being very active. It also ensures additional publicity for the sponsor by creating a race for qualification in the several months before the event itself. It is important to note that the activity bonus is only used for the qualification calculation; it does not affect a player’s Elo rating and Elo ratings would continue to be calculated in the usual way.
The winner of one tournament would not be given an automatic place in the following one. This does not happen in other sports and a free place would only encourage inactivity. The calculations would be made three months prior to the event and invitations and player contracts issued as soon as possible thereafter.
There are 14 rounds. The schedule would be
| Day 1 | Arrival day and drawing of lots |
| Days 2-5 | Rounds 1-4 |
| Day 6 | Rest day |
| Day 7-10 | Rounds 5-8 |
| Day 11 | Rest Day |
| Day 12-15 | Rounds 9-12 |
| Day 16 | Rest Day |
| Days 17 | Round 13 |
| Day 18 | Round 14 |
| Day 19 | Possible tie-breaks and prize ceremony |
| Day 20 | Day of departure |
The time-limit for the games should be 40 moves in 2 hours, then 20 moves in 1 hour, then 30 minutes for all the remaining moves. From move 60 onwards each player will receive an increment of 10 seconds per move.
In the event of a tie, there must be a procedure for tie-breaks.
4.1: If the tie is between two players, the tie-break will consist of two rapid games played at the rate of 30 minutes per player plus 10 seconds per move.
4.11: If there is still a tie then two further rapid games will be played at the rate of 10 minutes per player plus 5 seconds per move.
4.12: If there is still a tie, then the players will play a series of blitz games at the rate of 5 minutes per player plus 1 second per move. The first decisive game will decide the title.
4.2: If the tie is between more than two players, then the tie-break will consist of a single round-robin rapid event at the rate of 30 minutes per move plus 10 seconds per move.
4.21: If there is still a tie, then if this resulting tie is between two players then it will be resolved as in 4.11 and 4.12 above.
4.22: If the resulting tie is between more than two players then it will be resolved by a single-round rapid event at the rate of 10 minutes per player plus 5 seconds per move. If there is still a tie, and if this tie is between two players then it will be resolved as in 4.12 above.
4.23: In the extremely unlikely event that this tie is still between more than two players, then single-round robin blitz events will be held with the same time-limit as in 4.12 until a single winner emerges.
In view of the chaotic situation in the chess world, a World Championship will be recognised by the participation of all the leading players in an event which is designed to produce a fair winner. It is critical that the credibility of the event be established without a doubt, so that it is recognised as the World Chess Championship. It is unlikely that this can be achieved on the cheap.
I would estimate the cost of the event to be 2.5 million dollars. The prize fund would be 1.5 million dollars, which could be apportioned as follows (in dollars):
| 1st: | 650,000 |
| 2nd: | 300,000 |
| 3rd: | 150,000 |
| 4th: | 120,000 |
| 5th: | 100,000 |
| 6th: | 80,000 |
| 7th: | 60,000 |
| 8th: | 40,000 |
The reasonable returns even for the lower places ensure that qualifying places are regarded as very valuable, and therefore players will be encouraged to play more to get their ratings up and achieve a good ‘activity bonus’. The race to qualify will generate additional publicity for the main event.
The remaining 1 million dollars would be for organisational expenses, including hire of venue and publicity. A good website with live games and commentary is a must for an event of this type.
Chess is a game with a wide following, not only in the West but also in many other parts of the world. There is currently a boom in playing chess on the Internet and chess software often occupies leading positions in software best-seller lists. However, this boom has not been reflected in top-level chess due to organisational problems. This proposal offers a chance to solve the problems surrounding the World Championship, while at the same time offering a sponsor a chance to create a massive impact with a unique event.
That is what I wrote in February. In May a meeting took place between the ACP and FIDE. On the ACP website, the minutes of the meeting are given and include the following: “ACP agrees and adds that inactive players should lose rating points. FIDE agrees.” This may sound similar to the ‘activity bonus’ idea presented above, but in fact it is very different and, in my view, a very bad idea. No details are given of the ACP proposal, but for the sake of argument let us assume that a certain number of points are deducted for each inactive period.
Many chess players on the Elo list (even grandmasters!) may have a period of inactivity for a variety of reasons; for example, career, family, illness, pregnancy or child-raising. I should explain that at the moment, if you are inactive for some time then rating is not published, but it remains the same and if you resume playing again then you return at your old rating. It is surely desirable to tempt inactive players to start playing again, but forcing them to start again at a lower Elo rating and work their way up again is a powerful disincentive to resuming play. Can you imagine the reaction to “Well, Garry, of course we’d love you to start playing again, but you’ll have to start again at 2200.” Judit Polgar proved that you can have an inactive period and return again at full strength, and there is no special reason to suppose that a period of inactivity is necessarily damaging.
I could perhaps understand this proposal if there were some grave abuse being perpetrated in the chess world based on inactive players retaining their rating, but I cannot see anything like this. Therefore, this proposal would seem to have no positive effect, but an unquestionable negative effect. It seems to me that players should have the freedom to choose when and where they will play without fear of negative consequences, and I am surprised that the ACP, which is a players’ organisation, should be so keen to undermine this freedom.
I recall that several years ago there was a proposal, which again came from the players’ side, to strip grandmasters of their title if their Elo rating fell below 2500 (this was enthusiastically supported by Nigel Short, for example). Fortunately, this proposal didn’t get anywhere. There must be some strange quirk about chess players which makes them so enthusiastic about punishing their colleagues. Perhaps a psychologist can shed some light on the matter!
Dr John Nunn
Tell us what you think
We are interested in your opinion on the situation in world chess
and the proposals that have been presented here by John Nunn.
