New ECF ratings codes

[Neill Cooper]

Ratings codes have changed from A to F to:

Category A: A player who has 30 or more rated results in the previous 12 calendar months

Category H: For OTB, players who inherited a rating on 1st July 2020, but who has less than 9 rated results in the 60 calendar months prior to 1st July 2020 and has not played since; For Online a player with a rating inherited on 1st June 2021 and has no subsequent rated results. When a player with an H rating subsequently plays, their rating will convert to category K.

Category K: All ratings of players not qualifying as A or H, with full active ratings that will be updated by the K-method

Category P: All ratings of players with active ratings that will be updated by the P-method. Such ratings are published but because of limited data; player's strength here should be assessed with wider evidence where available.

Unrated: Any K or P rating where the player has no rated result in the previous 36 calendar months, but see below. An unrated player, on return, will be treated as a new player with one dummy game assuming a draw against an opponent with the players last full rating.

These categories became effective on 1st June 2022. On 1st Deceember 2022, any player still with an H rating will be converted to unrated (but will not become unrated before then).

(See https://www.ecfrating.org.uk/v2/help/help_profile.php )

[John Swain]

This new system is very ill-judged, at least in terms of the lists of top players by age.

Yesterday, I received a query from a parent of a very promising local junior who could not understand the new system nor why his son was not in the list of top players. Thirty games in the last calendar year are now required to be listed; the junior in question fell four short since resuming OTB play late last year. I would have thought that a lower threshold, perhaps the equivalent of three congresses (15 games) might have been a more appropriate target.

An unfortunate further consequence is that fairly weak players are now listed in the Top Hundred; the main criterion seems to be quantity (30+ games) and not performance.

[Keith Arkell]

Hopefully loads of us will get one more game added to our June ratings when the omitted 4NCL round 9 games, played back on May 2nd, get included.

[Paul Cooksey]

It does seem surprising to exclude K ratings from the top player lists. I had understood that an advantage of Elo was that ratings could be accurate without 30 games being needed.

In passing, starting the rapid rating calculation of adults with a dummy 1700 rather than their standard rating seems more likely to produce inaccurate ratings.

[Brian Valentine]

The Top Players list will be fixed. We were expecting to put in an improved version for month end, but it failed testing. The result is that the lists are still looking for A-E rather than the new categories.

[Paul Bielby]

As I turn 86 later this month, I was very surprised to find myself at the top of the 'Top players' list for Over-80s. I know at least half a dozen, probably more, octogenarian players much better than I am. Then the list implies that there are only 16 rated players over-80. Have the missing 84 players from the May 'Top 100' all passed away in the last month? I doubt it.

More importantly why do these 'Top player' lists stop at 80+? When Senior players lists start at the ridiculously young age of 50+, 55+, 60+, etc. shouldn't we have a a list for the over-85s, over-90s? FIDE can manage it for their ratings indeed in even more detail (81+, 82+, 83+, etc.)

[Roger Lancaster]

This new system is very ill-judged, at least in terms of the lists of top players by age.

Yesterday, I received a query from a parent of a very promising local junior who could not understand the new system nor why his son was not in the list of top players. Thirty games in the last calendar year are now required to be listed; the junior in question fell four short since resuming OTB play late last year. I would have thought that a lower threshold, perhaps the equivalent of three congresses (15 games) might have been a more appropriate target.

An unfortunate further consequence is that fairly weak players are now listed in the Top Hundred; the main criterion seems to be quantity (30+ games) and not performance.

I'd echo this.

[NickFaulks]

I was very surprised to find myself at the top of the 'Top players' list for Over-80s.

I was unaware of these age lists, and am a little surprised not to find my name on any of them. Perhaps this is because my federation is not ENG.

[Ian Thompson]

I was very surprised to find myself at the top of the 'Top players' list for Over-80s.

I was unaware of these age lists, and am a little surprised not to find my name on any of them. Perhaps this is because my federation is not ENG.

You're number 11 on the 70+ list, but only if you change the selection criteria from English players to all players.

The reliability of these lists is questionable anyway, particularly with the senior age groups, because the ECF doesn't have everyone's ages.

[NickFaulks]

You're number 11 on the 70+ list, but only if you change the selection criteria from English players to all players.

Thanks, I'm still learning!

[Brian Valentine]

This afternoon we have put in a rating system upgrade.

The first change is topical in that the Top Players section has been revamped. This takes into various feedback over the last two years and includes:
• Changing the age definitions to now fit academic year and FIDE ages correctly
• Qualifying categories become A, H and K.
• Adding Online tables
• Adding junior exact year tables
• Adding a senior 85+ category
• Adding a members only alternative
• Improvement and activity tables are based on latest month rather than from July 2020.

The player profile page now includes a field for titles, where those with FIDE and English National master titles are recognised.

One point of explanation may help people like Nick Faulks who were on a Top Player list but have disappeared with this change. The change in categories has changed the treatment of active/lapsed ratings. During covid we extended the active rating period out to six years; apart from category H cases who have a stay of execution, ratings now expire after 3 years. However to have an A-E category one had to have played in the previous two years. Under the new approach this is extended to three years. The Top Players lists now include players who last played between two and three years ago.

The seniors lists have thrown up one incorrect date of birth.

Brian Valentine
Manager ECF Rating

[John Swain]

Many thanks, Brian, for responding so speedily with your rating system upgrade. I'll pass on the good news to the junior and his parent that he's now back in the top players' list!

Many thanks too, as an ECF official, for engaging with the EC Forum!

[Paul Bielby]

Well, I had my brief moment (48 hours) of glory at the top of a top players list. But it all looks a lot more reasonable now.

Many thanks too, Brian, for the 85+ list and for including the NM titles. Would anybody qualify for a 90+ list?

[Arnav Sud]

This new system is very ill-judged, at least in terms of the lists of top players by age.

Yesterday, I received a query from a parent of a very promising local junior who could not understand the new system nor why his son was not in the list of top players. Thirty games in the last calendar year are now required to be listed; the junior in question fell four short since resuming OTB play late last year. I would have thought that a lower threshold, perhaps the equivalent of three congresses (15 games) might have been a more appropriate target.

An unfortunate further consequence is that fairly weak players are now listed in the Top Hundred; the main criterion seems to be quantity (30+ games) and not performance.

I kind of agree. With full time work and family commitments most people can squeeze in a Club Season and (say) two to three congresses. That is about 21 games (Club Season + 2 congresses) or 27 games (Club Season and 3 congresses). That is a serious annual commitment to expect from amateur players!

[Mike Gunn]

The number 30 has a special significance in chess rating theory. Richard Clarke identified it as the minimum number of games needed to establish s reasonably accurate rating. If you are a "serious" amateur player then you should be playing at least 30 games a year in the absence of Covid (in my humble opinion).