ECF to scrap its grading system

[Nick Grey]

Ok it seems London is out on a limb. Not the first time. I seem to remember we voted to remain. I also do not believe the leagues have got it wrong on adjournments. Nor the leagues which have folded for lack of players.

We have done some strategic thinking as well as action at our club. There is a demand for social chess and matches at the club. Less it seems for playing away matches more than an hour away. Many not wanting to play weekends. And many wanting to play more than once a week & some hoping to play at most once or twice a year.

I'm happy with 4NCL apart from defaults and organisers do their best to give players a game if they want it.
The dual ECF/Fide calculations give some people a good natter/moan too. As do the traffic & weather conditions & teas & coffees & water.

And it would not bother me if there was no grading system at all.

[Alex Holowczak]

I also do not believe the leagues have got it wrong on adjournments. Nor the leagues which have folded for lack of players.

Hmmm.

[Roger Lancaster]

As I understand it, opponents of adjournments can come up with two alternatives - guillotine finishes, which in turn have many opponents due to mad time-scrambles, or increments. One difficulty with increments, when used in evening leagues, can be eloquently set out by reporting below the relevant rule (quoted verbatim) from one such local league - For games played using the Fischer Timing rate of play (all moves in 70 mins plus 10 secs) clocks are stopped immediately by the player having the move when “time” is called and the game shall be sent for adjudication. Before a hail of invective commences, I should add that I'm no great fan of adjournments but feel that the alternatives can also have snags.

[Ian Thompson]

One difficulty with increments, when used in evening leagues, can be eloquently set out by reporting below the relevant rule (quoted verbatim) from one such local league - For games played using the Fischer Timing rate of play (all moves in 70 mins plus 10 secs) clocks are stopped immediately by the player having the move when “time” is called and the game shall be sent for adjudication.

If the playing session is 3 hours I wouldn't have a problem with such a rule. It means that at least 120 moves would have been played before adjudication occurred. Looking at my games, that means about once every 3000 games. Looking at a large database, a game longer than that happens about once every 2000 games.

[NickFaulks]

If the playing session is 3 hours I wouldn't have a problem with such a rule. It means that at least 120 moves would have been played before adjudication occurred. Looking at my games, that means about once every 3000 games. Looking at a large database, a game longer than that happens about once every 2000 games.

Precisely. What happens far more often is that a game gets delayed by ten minutes ( time for thirty more moves ) because of a dispute, or because the analogue clock has been running slow and the players didn't notice. That is of course considered to be part of the charm of the traditional game.

[Roger Lancaster]

One difficulty with increments, when used in evening leagues, can be eloquently set out by reporting below the relevant rule (quoted verbatim) from one such local league - For games played using the Fischer Timing rate of play (all moves in 70 mins plus 10 secs) clocks are stopped immediately by the player having the move when “time” is called and the game shall be sent for adjudication.

If the playing session is 3 hours I wouldn't have a problem with such a rule. It means that at least 120 moves would have been played before adjudication occurred. Looking at my games, that means about once every 3000 games. Looking at a large database, a game longer than that happens about once every 2000 games.

In the league in question, games nominally start at 7.45 (although there's sometimes a delayed start when - for example - traffic conditions leave both sides short of players) and typically finish at 10.30 so, if I've got my arithmetic right, even a prompt start could result in as few as 75 moves. A 10-minute delay, which is far from uncommon, gets one down to just 45.

[While I was typing this, Nick appears to have made much the same point].

[Roger de Coverly]

If the playing session is 3 hours I wouldn't have a problem with such a rule.

I think it's the Thames Valley and Border Leagues that have these rules or something similar. Has the move 90 adjudication ever been applied in practice?

One other thing that bothered me about increments was how to deal with draw claims under the 50 move rule. Suppose for example someone decided to play out KRB v KR.
More recent models of DGT clocks do however always count, even when they don't do anything with it. You only need to know the move number at which the ending was entered, no need for a score after that if less than 5 minutes remained provided checking the count on the clock is allowed.

My objection to adjudication was the gamesmanship it permitted where perhaps in a two and a half hour session, both players reached move 30 taking about an hour each, whereupon one of them would sit for 30 minutes. If there's an end of session adjudication option, that makes it a dangerous practice, but only if the player with the time gain is willing to travel.

You could have the "adjudication at move 90" option even without increments. You would need the clock to tell you that move 90 had been reached as scoring that far is unlikely. It's like a quickplay finish except "unable" claims are ruled out until move 90.

[David Sedgwick]

In the league in question, games nominally start at 7.45 (although there's sometimes a delayed start when - for example - traffic conditions leave both sides short of players) and typically finish at 10.30 so, if I've got my arithmetic right, even a prompt start could result in as few as 75 moves. A 10-minute delay, which is far from uncommon, gets one down to just 45.

In that case, the time limit of G/70 + 10s seems to me to be plain daft.

G/60 + 10s, allowing time for 105 moves even on your worst case scenario, would be far more sensible.

[Reg Clucas]

What happens far more often is that a game gets delayed by ten minutes ( time for thirty more moves ) because of a dispute, or because the analogue clock has been running slow and the players didn't notice.

Or even more often, the home team hasn't exercised its prerogative to start the clocks on time.

[Nick Grey]

Not having long enough to seal a move is a problem. Especially a match not starting on time on an overrunning junior club. There are exceptions of course on starts.
I think ECF will have a problem in implementing a new grading system with instructing leagues on how they can play their games.
Quite aware that there are compromises though most of the time it is about distance of travel.
Sorry but I'm one of those encouraged by the 70s chess explosion and in those days games were adjourned at the top level. So good enough for a world championship it was and still is good enough for me.

[Brian Valentine]

Not having long enough to seal a move is a problem. Especially a match not starting on time on an overrunning junior club. There are exceptions of course on starts.
I think ECF will have a problem in implementing a new grading system with instructing leagues on how they can play their games.
Quite aware that there are compromises though most of the time it is about distance of travel.
Sorry but I'm one of those encouraged by the 70s chess explosion and in those days games were adjourned at the top level. So good enough for a world championship it was and still is good enough for me.

I don't expect to be "instructing" leagues. Given the recent council decision to continue with adjudications, I expect to take the date on which we are told the result is decided as the relevant date to decide when it is included for grading.

Brian Valentine
Manager of ECF Grading

[Andy McCulloch]

Some improvement in the speedy reporting of congress results will certainly be needed for any monthly grading.
I have taken part in seven events in England so far this season. Results for only four of these congresses are presently available on the grading site, and three of these were at the British in July.
The Scottish elo system has been updated on a weekly basis for some years, and congress results are posted within a day or two, often before entrants get home, and graded within a fortnight at most..
My local Glasgow League results are processed very rapidly as well, often on the league website within hours of matches being completed.
I appreciate that the Scottish system is smaller by a factor of 10, but computers, and the internet, are available to us all.
The annual grade obtained at the end of the season is used to determine board order in league matches for the entire season, no using the higher of July or January list.
The 'live' grade is more of personal interest to players, and, in my own case, it is disappointingly decreasing.

[Alex Holowczak]

Some improvement in the speedy reporting of congress results will certainly be needed for any monthly grading.
I have taken part in seven events in England so far this season. Results for only four of these congresses are presently available on the grading site, and three of these were at the British in July.

I regularly find myself taking ages to submit grading files, whereas I do rating files almost instantly. Because why bother getting them in quickly? I have to for rating files, so I do. No reason why I couldn't do grading files as quickly, but with no need to do so I just do it as and when.

[Roger de Coverly]

No reason why I couldn't do grading files as quickly, but with no need to do so I just do it as and when.

It sounds as if the ECF have more than one problem to resolve. Even if they can sort out the theory of what a monthly grade is measuring and get the necessary calculation and reporting mechanisms in place, there's still a need to change a culture of lax reporting and for leagues, that's down to club officials and their speed of result reporting.

[Nick Grey]

A match over 10 boards could have 9 results in the month of the match & the final result in the next list (or list after). What reason could there be to change the current convention.

It seems some are also getting bogged down in what the monthly grading calculation would be when not even designed the new system.