English Chess Forum

The system used to calculate junior grades for August 2009 was seriously flawed, creating some unrealistic grades and leaving opponents unable to calculate their new grades. Most juniors are not rapidly improving, but a few are.
Hopefully the ECF will be more open about how it will calculate Junior Grades this coming season, and the method which will be used in August will be published and fixed by June 30th (if not earlier). Here are a few suggested alternatives, based on 2 themes:

1) Only juniors who play more than 30 games should be deemed to be new players, as they are the ones who are most likely to be rapidly improving; or
2) Juniors whose grade increases by more than 10 (or some value) grading points are deemed to be new players; or
3) Players (Adult or Junior) whose grade increases by more than 10 (or some value) grading points are deemed to be new players ; or
4) Players (either 'Junior' or 'Adult and Junior') who play more than x (e.g. 30) games and whose grade increases by more than y (e.g. 10) grading points are deemed to be new players.

Neil,

Option 3) was my original suggestion back in 2006.

Out of interest, how are you defining "rapidly improving"?
Do you know (statistically speaking) that "most juniors are not rapidly improving" (whatever definition you attach) or is this a guess?

Sean Hewitt wrote:Out of interest, how are you defining "rapidly improving"?
Do you know (statistically speaking) that "most juniors are not rapidly improving" (whatever definition you attach) or is this a guess?

Rapidly improving is, say, + 20 points in a year.

The figures are anecdotal based on the 100+ juniors I know quite well, many of whom are at secondary school and have long play grades. Of those who play less than 20 games a year only a few, until last year, were going up by more than 10 points a year. (I think actual I have also done the analysis and from 2007 to 2008 only a small fraction (about 10%) of juniors had a grade change of +20 or higher.)

I still believe "we" should publish these magic numbers but I seem to be in the minority

I agree that the grading system needs fixing. However any change does give rise to side effects which I hope the graders will consider fully.

The dangers with these proposals are:
1) They almost certainly will be inflationary
2) The base of last year's grade will be used even less in the calculations, hence any calculation system based on previous gradings is likely to be less stable.

These issues might be minor in comparison to advantages gained but they need due consideration.