Government funding for chess

[Mike Gunn]

My idea would have to be combined with a reform of the membership structure to have 2 classes of member: club/ league members and those who play in congresses too. (=bronze and silver?). You would adjust the membership fees to bring in the same income as now.

[Roger de Coverly]

Would the players all have to be Gold ECF members?

That is an ECF internal rule which it could change. I think also that arbiters have to be present, at least notionally.

Given that you can now play in tournaments against players with ratings in both systems where your performance increases your ECF rating and reduces your FIDE one, the more parallel games that are FIDE rated, the more the gap can widen. This may, probably would, change if FIDE's revaluation of 1000 - 2000 players to 1400 - 2000 goes ahead and they enhance the ratings for new entrants by the dummy player device.

[Ian Thompson]

(which would have to adopt an U2400 rule to make 3 hour games FIDE legal)

Not at all. Any evening games involving 2400+ players would simply not be submitted for FIDE rating - which I believe most of those players would prefer anyway.

If one player in a game was under 2400 and the other over 2400 that probably wouldn't go down very well with the lower rated player if they didn't lose.

[Paul Cooksey]

I've have been meaning to post on the new ECF memberships scheme for ages. But particularly in the light of this announcement I don't know enough to say much sensible. However abolishing Gold to get more games FIDE rated was already being considered.

The plan in the last ECF budget was to grow membership income from 235k to 306k by 2025. Coincidentally the international budget for 2025 was 71k.

[Ian Thompson]

This would have the advantage of unblocking England's current problem with inaccurate FIDE ratings for most of our players because they simply don't get enough FIDE rated games.

Would it? Isn't there a risk that players who currently only play in local leagues and therefore don't have a FIDE rating will get a rating based on a small number of games which is therefore not going to be reliable?

[Roger de Coverly]

The plan in the last ECF budget was to grow membership income from 235k to 306k by 2025.

How would they plan to do that without putting prices up and perhaps abolishing the freebies? Are there signs of a 1972 style boom or even a 1993 one?

[Mike Gunn]

Accuracy is down to the number of games played (whatever the system) so a FIDE rating based on few games will be just as inaccurate as a national rating. However, I do prefer the FIDE system of just rating games against players who already have an established rating rather than the ECF approach.

(There is an unstated assumption in my proposal: if we FIDE rate everything then we don't need a national rating system and the national system will disappear after a short overlap).

[Ian Thompson]

Accuracy is down to the number of games played (whatever the system) so a FIDE rating based on few games will be just as inaccurate as a national rating.

Yes, but that misses the point. Someone who's been playing in a local league for years will have an ECF rating based on a reasonably large number of games. If the games now become FIDE rated, they'll get a FIDE rating based on a small number of games and it may take years before the FIDE rating becomes as reliable as the ECF rating.

(There is an unstated assumption in my proposal: if we FIDE rate everything then we don't need a national rating system and the national system will disappear after a short overlap).

What happens for players rated under 1000 at the moment and possibly under 1400 from next year?

[Brian Valentine]

This would have the advantage of unblocking England's current problem with inaccurate FIDE ratings for most of our players because they simply don't get enough FIDE rated games.

Would it? Isn't there a risk that players who currently only play in local leagues and therefore don't have a FIDE rating will get a rating based on a small number of games which is therefore not going to be reliable?

It's difficult to appraise Mike's suggestion without some idea of what the "current problem" is and what "inaccurate" mean. It might be tht the FIDE consultation could fix one or both of these without ECF funding.

While drafting I note Mike's later post. I'm all for rating more results for FIDE and (albeit over time as Ian alludes the problems of a sudden influx of new players) firmly believes that rating more results improves the system. His proposal fails to address what to do with players rated below 1000 (now 1400 soon). Over 40% of those ECF rated for Rapid Chess are below 1000.

[Mike Gunn]

The problem is that English juniors and beginners have inaccurate FIDE ratings because they don't get enough FIDE rated games. The idea is to give them more games.

As far as the accuracy issue is concerned, I would move back to a Clarke type system where for active players the rating is based on games played over the last 12 months, but I realise that boat has sailed. I haven't done any proper analysis but it is my observation that the current ECF system is too slow to react to an early overestimate or underestimate of strength. If you try and remove the lag by increasing the K factor and you just get volatility. As a County captain I didn't get much variation in grades year on year (apart from what I expected from players' results over the year) but now players' ratings bounce up and down like yoyos (OK, a slight exaggeration) and it's difficult to know whether a particular player (who has played in my team for 20 years) will be eligible next year or not.

Ian T makes a reasonable point above that the adoption of FIDE ratings would take some time to shake down (in terms of accuracy).

Perhaps the role for national ratings would be to do something better than FIDE for juniors and other new players, until they achieve a stable FIDE rating?

[Roger de Coverly]

The problem is that English juniors and beginners have inaccurate FIDE ratings because they don't get enough FIDE rated games.

I believe the problem is that apart from the k=40 rule which only really works for the most active and consistent players, there is no mechanism in the current FIDE system either to start new players at a level which assumes future improvement, or to have a means of ignoring their earliest rating and treat them as new players once they have improved.

k=40 is not only a ladder taking sucessful players to higher ratings but also a snake whivh dumps them back where they started when they have a poor tournament.

[Matt Mackenzie]

The plan in the last ECF budget was to grow membership income from 235k to 306k by 2025.

How would they plan to do that without putting prices up and perhaps abolishing the freebies? Are there signs of a 1972 style boom or even a 1993 one?

Anecdotally, some clubs to seem to have had an increase of membership post-Covid (due in large part to an influx of online players)

Though even so, this does appear fairly uneven in its effects.

[NickFaulks]

the more parallel games that are FIDE rated, the more the gap can widen.

If the gap widens for one of the players, then it will narrow for the other. It may narrow for both but I don't see how it will ever widen for both.

[Joey Stewart]

The plan in the last ECF budget was to grow membership income from 235k to 306k by 2025.

How would they plan to do that without putting prices up and perhaps abolishing the freebies? Are there signs of a 1972 style boom or even a 1993 one?

Anecdotally, some clubs to seem to have had an increase of membership post-Covid (due in large part to an influx of online players)

Though even so, this does appear fairly uneven in its effects.

It's more then just anecdotal in my area - every single club has had multiple new players joining over the last year, it's definitely a bit of a boom time for chess again.

[Roger de Coverly]

If the gap widens for one of the players, then it will narrow for the other.

In the small print of the ECF's implementation of Elo's methods which is otherwise similar to FIDE's, they have intruduced a tweak to the k=40 rule for juniors. It only applies if the performance over a month is positive. If they lose points over a month, k=20 is applied. Also junior players with ratings too low take points off established players and if they only have short playing lifetimes, never return them.