ECF Junior Selection Policy September 2025

[Mick Norris]

revised

The objectives of the ECF Junior Selection Policy are to choose the strongest players to represent England in international individual and team events, to produce the best possible results for England players and teams, and to assist talented juniors in gaining valuable international experience without impeding their general academic achievement.

Achieving these aims is a significant challenge, given that government funding for chess is currently less in the UK than in many other countries. Fundraising to support England Juniors playing in the most important international junior events is a priority for the ECF, and a key part of the role of ECF Director for Junior Development.

The ECF’s ethos is to select juniors to represent England purely on merit, rather than their ability to self-fund their participation in international junior events.

[Ian Thompson]

revised

It mentions "the ECF International Series of FIDE-rated Swisses (from 2026 onwards)" as a qualification route for juniors. That sounds like something that might be of interest to adult players as well.

[Roger de Coverly]

It mentions "the ECF International Series of FIDE-rated Swisses (from 2026 onwards)" as a qualification route for juniors. That sounds like something that might be of interest to adult players as well.

They might just mean the existing Cambridge and Coventry events plus the English Championship.

[John Swain]

It seems to me that this new policy devised by Tim Wall is a welcome step in the right direction. Focusing on ECF standard play ratings rather than paying undue attention to rapidplay events is sensible, given the nature of many international junior events. Having a smaller, more select ECF group at each tournament, ideally accompanied by a coach, makes good sense too.

[Ian Thompson]

It mentions "the ECF International Series of FIDE-rated Swisses (from 2026 onwards)" as a qualification route for juniors. That sounds like something that might be of interest to adult players as well.

They might just mean the existing Cambridge and Coventry events plus the English Championship.

It's probably related to this, which I noticed while reading the July ChessMoves this afternoon -

"The Cambridge International was very well received, and players enquired whether or not the ECF could organise more of these events on a similar basis, perhaps in other parts of England. I [Alex Holowczak] have been in contact with some organisers already to explore this possibility. Should the demand for these events be sufficient, the long-term plan will be to turn this into an annual event series, combined with existing events of a similar nature already held in England such as the Hastings Masters, the London Chess Classic Masters, the Isle of Wight Masters and the Northumbria Masters."

[Brian Beattie]

It seems to me that this new policy devised by Tim Wall is a welcome step in the right direction. Focusing on ECF standard play ratings rather than paying undue attention to rapidplay events is sensible, given the nature of many international junior events. Having a smaller, more select ECF group at each tournament, ideally accompanied by a coach, makes good sense too.

I agree that the revised policy looks to be a step in the right direction.

However, the reality remains that prospective juniors are best served then by eschewing Classical ECF rated junior events (British Junior, EJCOA, LJCC, etc.) and deliberately targeting 60+ (probably closer to 80) ECF Classical games per year otherwise they will likely either lose too much rating or not acquire rating fast enough to put them in contention. At least for those in the U12s and below. Whether this a major bug or feature I guess depends on one's perspective.

[Roger de Coverly]

and deliberately targeting 60+ (probably closer to 80) ECF Classical games per year otherwise they will likely either lose too much rating or not acquire rating fast enough to put them in contention. At least for those in the U12s and below. Whether this a major bug or feature I guess depends on one's perspective.

Doesn't that depend on whether they are targeting ECF or FIDE ratings? The point being that the domestic ECF system has incorporated more measures than the FIDE one to revalue rapidly improving players to a rating which reflects their playing strength.

[Brian Beattie]

and deliberately targeting 60+ (probably closer to 80) ECF Classical games per year otherwise they will likely either lose too much rating or not acquire rating fast enough to put them in contention. At least for those in the U12s and below. Whether this a major bug or feature I guess depends on one's perspective.

Doesn't that depend on whether they are targeting ECF or FIDE ratings? The point being that the domestic ECF system has incorporated more measures than the FIDE one to revalue rapidly improving players to a rating which reflects their playing strength.

I am welcome to be wrong but I haven't seen any measures that ECF has incorporated that does this. If the measure(s) involved revolves around K factors (possibly the ECF allows a factor of 40 where FIDE might revert to 20 or lower sooner?) it's still ineffective as the numbers above are based on K=40

Edit: Ah, I see here it's 20 when falling. Not sure it changes the situation all that much though.

[Roger de Coverly]

Edit: Ah, I see here it's 20 when falling. Not sure it changes the situation all that much though.

It must help though. It's quite common for improving players to go through a poor patch when they get bad results. Having an asymmetric calculation means they gain points even when on a longer time average they just broke even.

Whether it's enough to comabat the deflation seen in FIDE ratings has yet to be formally reviewed. FIDE took a drastic step of shifting the minimum rating from 1000 to 1400 and revaluing all the ratings between 1400 and 2000. Apart from relatively inactive players with original ratings from many years ago, I think it still the case that ECF ratings are higher than FIDE ones.

Events with FIDE rating are popular among younger players. It can be noticeable that if there's an Open FIDE rated and an U2000 that isn't, that the U2000 lacks ambitious juniors.

This year's Kidlington seemed a case in point.
Open
https://rating.englishchess.org.uk/even ... ction_no=1
U2000
https://rating.englishchess.org.uk/even ... ction_no=1

[Brian Valentine]

I think the 60+ suggestion can be improved.

The effectiveness of the strategy depends on the intensity of activity. Assuming they are improving and playing more chess does not diminish performance then ideally the aspirant should play 18 games in the month before the quailifying list date. In that event their next rating will based wholly on that month's performance (the other 42 results being irrelevant). If they were to play 10 games a month for six months, then about 1/8th of that month's rating will be at the level of the start rating. If 18 games in a month is too intense they can get significant benefit by some intermediate trade-off.

[NickFaulks]

Edit: Ah, I see here it's 20 when falling. Not sure it changes the situation all that much though.

Yes it certainly does. By doing nothing other than playing six games per month against each other, juniors can expect to increase their rating by an average of around 100 points over a year ( plus / minus a random element depending upon the pattern of those results across months ).

[Brian Beattie]

I think the 60+ suggestion can be improved.

The effectiveness of the strategy depends on the intensity of activity. Assuming they are improving and playing more chess does not diminish performance then ideally the aspirant should play 18 games in the month before the quailifying list date. In that event their next rating will based wholly on that month's performance (the other 42 results being irrelevant). If they were to play 10 games a month for six months, then about 1/8th of that month's rating will be at the level of the start rating. If 18 games in a month is too intense they can get significant benefit by some intermediate trade-off.

Clearly, it can be optimized. But most kids are probably going to get their first ECF classical rating courtesy in part to playing something like the LJCC U8 against likely a bunch of underrated and low rated kids. At which point they are going to have to play a lot of games to get an accurate rating.

But, since I have the right person here. If we have a child who is 200 points underrated and who is improving at 200 points of rating every year how many games per year are they going to have to play to stablize around their actual rating performance. And, how does this number change assuming they also want to play the British Junior every year where they on average face players 200 points underated.

Edit: with tournaments distributed evenly throughout the year

[Brian Beattie]

Edit: Ah, I see here it's 20 when falling. Not sure it changes the situation all that much though.

Yes it certainly does. By doing nothing other than playing six games per month against each other, juniors can expect to increase their rating by an average of around 100 points over a year ( plus / minus a random element depending upon the pattern of those results across months ).

Well, we can look at a more practical example. My children have a combined 23 periods of Classical data. Of those 20 of them are rising and 3 of them are falling. And of the falling months one was for a loss of 1 point and the another was for a loss of 4 points. So, not really compelling. And there is a decent probability the third falling period may of never happened but you can't bail out of a team competition just because you are ill. All three falling months also involved junior events which shouldn't come as a surprise.

So, I am not too convinced that rapidly improving juniors (U13 AND U2000) benefit much from a falling K=20. Now, of course, once juniors get enough games per year that their rating approaches some proximity of their actual performance obviously these players will start experiencing more falling periods. So, ECF and FIDE probably vary quite little in closing the big gaps. But, maybe once you are within 50 points the ECF system will narrow that gap at quicker rate.

[NickFaulks]

But, maybe once you are within 50 points the ECF system will narrow that gap at quicker rate.

Do you mean the difference between published rating and "true playing strength"?

If so, you are asking for the unattainable. With k=40, you cannot even measure playing strength with such accuracy, because random fluctuations are bigger than that.

[Brian Beattie]

But, maybe once you are within 50 points the ECF system will narrow that gap at quicker rate.

Do you mean the difference between published rating and "true playing strength"?

If so, you are asking for the unattainable. With k=40, you cannot even measure playing strength with such accuracy, because random fluctuations are bigger than that.

I am not asking for anything. I am simply giving examples. Not any different than the hypothetical case of two juniors playing each other 6 times per month which you led with. But, we can expand on it. Let’s say we have junior who is 1800 and 5 times per month plays a 2000 rated player. The kid is underrated and actually plays at a 2000 level. For simplicity he alternates win and losses. Will use K factor of 40 for both rising/falling

Month 1: 1863
Month 2: 1876
Month 3: 1922
Month 4: 1922 (Could actually be a falling month but whether K40 or K20 would be meaningless)
Month 5: 1958
Month 6: 1949 (We have our first official falling month so in ECF around 1955 now instead)
Month 7: 1979
Month 8: 1965 (For obvious reasons ratings continue to alternate between rising/falling now)
Month 9: 1991
Month 10: 1975
Month 11: 1999 (55 games in and we are about to touch 2000 for the first time)
Month 12: 1981
Month 13: 2003 (don’t really need to continue ultimately we’ll alternate between 1990 and 2010)

I don't think any of this particularly contradicts getting 60+ games per year nor ECFs falling K factor having it's greatest impact within 50 rating points or so. Of course, the example is extra simplistic as the junior isn't improving throughout the year and I probably screwed up something along the way but intuitively it looks about correct.