I thought it was well established that many juniors tend to have sudden jumps in standard of play followed by a plateau before another jump. So I would expect there to be a lot of juniors, who, even though the long-term trend is improvement, are not improving for a period of at least a few months.NickFaulks wrote: ↑Tue Feb 12, 2019 11:06 amIf there is such a creature as a junior who is not improving....
ECF to scrap its grading system
-
- Posts: 3554
- Joined: Wed Jul 02, 2008 4:31 pm
- Location: Awbridge, Hampshire
Re: ECF to scrap its grading system
-
- Posts: 5249
- Joined: Mon Apr 09, 2007 5:56 pm
- Location: Croydon
Re: ECF to scrap its grading system
I stand corrected, although I suspect that in practice the upward "overshoot" is the more likely.NickFaulks wrote: ↑Tue Feb 12, 2019 11:16 amHardly the exact opposite. I have not gone back and checked this case but If I remember it correctly, his k factor was reduced not because his results were good but because he had so many games reported in that period that k=40 whould have caused overshooting. This could have been up or down.David Sedgwick wrote: ↑Tue Feb 12, 2019 11:07 amAm I right in thinking that FIDE do the exact opposite?
-
- Posts: 5249
- Joined: Mon Apr 09, 2007 5:56 pm
- Location: Croydon
Re: ECF to scrap its grading system
I recall my first five grades:Ian Thompson wrote: ↑Tue Feb 12, 2019 11:21 amI thought it was well established that many juniors tend to have sudden jumps in standard of play followed by a plateau before another jump. So I would expect there to be a lot of juniors, who, even though the long-term trend is improvement, are not improving for a period of at least a few months.NickFaulks wrote: ↑Tue Feb 12, 2019 11:06 amIf there is such a creature as a junior who is not improving....
1968 - 143
1969 - 139
1970 - 167
1971 - 164
1972 - 163
-
- Posts: 21313
- Joined: Tue Apr 15, 2008 2:51 pm
Re: ECF to scrap its grading system
It was a long standing hack to the Clarke system from about the 1970s onwards, that junior grade + Something was used in the calculations. "Something" was initially 5 points, then 10, and then later an age related addition. The perception from the 1960s had been that the grades of the then top adult players had been drifting downwards.NickFaulks wrote: ↑Tue Feb 12, 2019 11:06 amIf there is such a creature as a junior who is not improving, their rating will also be given an upward kick. Reasonable tradeoff?
-
- Posts: 4549
- Joined: Tue Apr 03, 2007 11:04 pm
- Location: writer
Re: ECF to scrap its grading system
For those who find have difficulty finding Brian's paper on the ECF website (like me) it is
https://www.englishchess.org.uk/grading-consultation/
For those who don't know: If 700/k is exceeded by the number of games in a period, then the player's rating will be distorted. This does not just hapen with juniors, but also blitz and rapidplay ratings. So k=40 is a mere 18 games in a month. k=20 it is 35. Of course, a player's rating can go down too much a well as rise. For such exteremely acive players,It is better to base the new rating solely on the current month; or to use k=700/no. of games.
In the early days Elo had great problems with players who were new to the list, but of already established playing strength. I know my own initial USCF rating caused concern in 1963. It was about 100 points too high at 2350. Monthly lists, ratings going down to 1000 and rating after just 5 games ameliorate this problem.
Brian, currently I think the lowest grade possible is 0. What will it be when the changes are made?
Roger, in the earliest versions of the Clarke system, players could LOSE gradng points even though they won the game.
https://www.englishchess.org.uk/grading-consultation/
For those who don't know: If 700/k is exceeded by the number of games in a period, then the player's rating will be distorted. This does not just hapen with juniors, but also blitz and rapidplay ratings. So k=40 is a mere 18 games in a month. k=20 it is 35. Of course, a player's rating can go down too much a well as rise. For such exteremely acive players,It is better to base the new rating solely on the current month; or to use k=700/no. of games.
In the early days Elo had great problems with players who were new to the list, but of already established playing strength. I know my own initial USCF rating caused concern in 1963. It was about 100 points too high at 2350. Monthly lists, ratings going down to 1000 and rating after just 5 games ameliorate this problem.
Brian, currently I think the lowest grade possible is 0. What will it be when the changes are made?
Roger, in the earliest versions of the Clarke system, players could LOSE gradng points even though they won the game.
-
- Posts: 21313
- Joined: Tue Apr 15, 2008 2:51 pm
Re: ECF to scrap its grading system
Hence the 40 point cutoff, again which like junior additions was a 1960s development.Stewart Reuben wrote: ↑Tue Feb 12, 2019 12:01 pmRoger, in the earliest versions of the Clarke system, players could LOSE gradng points even though they won the game.
If the proposed system is copying the FIDE one with a 400 point cutoff, that can also act to hold down the ratings of improved players and protect those of established players by restricting the loss and gain when the result goes totally the other way.
-
- Posts: 4549
- Joined: Tue Apr 03, 2007 11:04 pm
- Location: writer
Re: ECF to scrap its grading system
Roger >FIDE one with a 400 point cutoff<
I chose the 400 point cutoff to encourage stronger players to play in Open Swisses. The late Mikko Markulla wanted it to be 500 points.
I chose the 400 point cutoff to encourage stronger players to play in Open Swisses. The late Mikko Markulla wanted it to be 500 points.
-
- Posts: 8462
- Joined: Sat Jan 02, 2010 1:28 pm
Re: ECF to scrap its grading system
Wasn't the original choice 350, later changed to 400?
If you want a picture of the future, imagine a QR code stamped on a human face — forever.
-
- Posts: 798
- Joined: Wed Oct 08, 2008 8:38 pm
- Location: Sevenoaks, Kent, if not in Costa Calida, Spain
Re: ECF to scrap its grading system
No, that was a later development, mid-seventies I'd say. It was 50 points when I started playing in 1967Roger de Coverly wrote: ↑Tue Feb 12, 2019 12:08 pmHence the 40 point cutoff, again which like junior additions was a 1960s development.Stewart Reuben wrote: ↑Tue Feb 12, 2019 12:01 pmRoger, in the earliest versions of the Clarke system, players could LOSE gradng points even though they won the game.
-
- Posts: 5234
- Joined: Tue Mar 31, 2009 11:51 pm
- Location: Millom, Cumbria
Re: ECF Monthly Grading Consultation
IMO there is nothing wrong with this section having a thread on the topic as well.
"Set up your attacks so that when the fire is out, it isn't out!" (H N Pillsbury)
-
- Posts: 5249
- Joined: Mon Apr 09, 2007 5:56 pm
- Location: Croydon
Re: ECF to scrap its grading system
My guess would be 1970.Simon Brown wrote: ↑Tue Feb 12, 2019 1:04 pmNo, that was a later development, mid-seventies I'd say. It was 50 points when I started playing in 1967Roger de Coverly wrote: ↑Tue Feb 12, 2019 12:08 pmHence the 40 point cutoff, again which like junior additions was a 1960s development.Stewart Reuben wrote: ↑Tue Feb 12, 2019 12:01 pmRoger, in the earliest versions of the Clarke system, players could LOSE gradng points even though they won the game.
-
- Posts: 21313
- Joined: Tue Apr 15, 2008 2:51 pm
Re: ECF to scrap its grading system
I think it was in place when Swiss events became large and open. Given that some of them were using random pairings, there would have been a wave of moans when 220 players got pairings against 120 players. As it was the pointlessness of such pairings was made in forums of the time, otherwise known as letters to or reports in BH Wood's magazine.
-
- Posts: 1295
- Joined: Wed Aug 12, 2009 11:12 pm
Re: ECF Monthly Grading Consultation
I agree with Angus. Whilst there is nothing wrong with new threads partially duplicating others, its preferable that for discussion of substantive issues posters can follow the debate in one place rather than two, and are not then tempted or obliged to post the same comments twice over. Merger would enhance the discussion rather than diminish it. And yes I am aware that some posts have been repeating themselves in multiple threads for a number of years often with no real regard to the subject matter -contributing to the richness of the forum
-
- Posts: 4549
- Joined: Tue Apr 03, 2007 11:04 pm
- Location: writer
Re: ECF to scrap its grading system
Nick .Wasn't the original choice 350, later changed to 400?<
Yes, that is correct.
There was indeed a 50 point cutoff in the BCF system. That meant it was possible to win the game, but gain no points against somebody graded 50 points or more below you. But if you drew or lost, it was a catastrophe.
FIDE switched from rating a whole tournament by taking the average of the opponent's ratings to rating each game separately at my suggestion, relayed through Vishy Anand.
Taking the average of a non-linear set of numbers was nonsense, but very convenient as all tournaments were round robins, mostly for title norms, initially.
Yes, that is correct.
There was indeed a 50 point cutoff in the BCF system. That meant it was possible to win the game, but gain no points against somebody graded 50 points or more below you. But if you drew or lost, it was a catastrophe.
FIDE switched from rating a whole tournament by taking the average of the opponent's ratings to rating each game separately at my suggestion, relayed through Vishy Anand.
Taking the average of a non-linear set of numbers was nonsense, but very convenient as all tournaments were round robins, mostly for title norms, initially.
-
- Posts: 9085
- Joined: Sat May 30, 2009 5:18 pm
- Location: Oldbury, Worcestershire
Re: ECF Monthly Grading Consultation
Can we have a new thread discussing whether or not the other two threads should be merged?