Alex Holowczak wrote:Give them the chance; the AGM was only 8 days ago...
49 days now, so I suppose it’s time someone answered. You won’t be expecting official answers from me, in the continued absence of a Manager of Grading. For that matter I wasn’t wearing an ECF hat when I said on the SCCU website that halfyear Standard grades would work the same way as Rapid. I just thought I was saying something obvious. But at least I can try to clear up some questions about the way Rapidplay
does work. Let’s hope I get all of it right.
I thought most of it was fairly clear from the online list’s Help page, but maybe it’s not.
The Help page wrote: X - Rapidplay only. Grade based on 30 or more games in the latest 6 months
A - Grade based on 30 or more games in the latest 12 months
B - Grade based on 30 games in the latest 24 months, of which at least 20 were played in the latest 12 months
C - Grade based on 30 games in the latest 36 months, of which at least 10 were played in the latest 12 months...
But I’ve cheated. That’s what it ought to say. If you look it up you will find it talks about “seasons†rather than periods of months, and for Rapidplay that’s not very clear. A 12-month period, for Rapidplay, may start on 1st June or 1st December.
So is Rapidplay based on a 6-month period, a 12-month period, or what? For X and A grades it’s clear. They use 6 and 12 months respectively. For other categories, it’s not calendar-based at all except for the 36-month limit beyond which we do not go. It just uses the 30 most recent games, assuming 30 are available. The category simply gives a rough idea of how numerous and recent the games are.
Roger de Coverly wrote:What grades will be used as the start point for the July 2012 calculations?
Will they be
(a) the July 2011 ones
(b) the July 2011 ones for existing players but the Jan 2012 ones for new players
(c) the July 2011 ones for games prior to the Jan 2012 list and the Jan 2012 ones for games after that.
(c). Each 6-month period is calculated separately, starting from the previous (6-month-old) grade. The points you scored ten months ago won’t change. They belong to a previous 6-month period and have been calculated already.
____________________
How will all this translate to Standardplay? My feeling was, and is, that it would be pretty daft to have one method for Rapid and another for Standard. But, in fairness, there are one or two issues that don’t arise with Rapidplay.
Roger de Coverly wrote:A major difference being that most rapid-play takes place in one-day congresses which are discrete events. It doesn't have to resolve the problems of how and when leagues should report.
Surely it’s understood that Standardplay leagues will need to report in two instalments, as Rapidplay leagues (admittedly far less numerous) already do. The graders have been warned, and none has protested so far. We’re not going to get everything in for the halfyear, but I think there’s a good chance we’ll get most of it. Where we don’t, the whole league will have to go into the second halfyear. I see no real way round that.
The obvious place to put the halfyear divide is, well, at the halfyear. The end of November, as for Rapidplay. But this is nowhere near the halfway point of the league season. League activity will be skewed towards the July list. So have the divide at the end of December, say? And move the entire grading “season†forward a month (Rapidplay included) so it starts on 1st July? You might think the cure is worse than the disease.
There is another issue. Standardplay grading as it stands
is based on 12-month periods. Where games are required from a previous period, we do not take the most recent. We can’t, because league games are not always properly dated. There is a league which plays all its games on 1st September, and another which prefers 31st May. The same thing happens with club internal games, where the dates by and large are genuinely not known. So we take the number of games we need, but they are not the most recent. They are notional games calculated at the average score for the whole 12 months. Not satisfactory, but what can you do? I think it was always meant as a stopgap solution, till the day dawned when we got all the games properly dated. The stopgap looks tricky to apply with a twice-yearly system, so we’ll probably have to abandon it and switch to “most recent†straight away. The good news is, the number of leagues that don’t date their games is down to half a dozen or so, and dwindling. The bad news is, I can’t say the same for club internal. Club internal is now the bigger problem. Reporting in two instalments will help, and maybe adjusting the game dates so they all come in the middle of a period rather than at one end of it. But I doubt if there’s a complete solution.
Roger de Coverly wrote:...the dubious practice of treating juniors as new players...
I don’t like it much myself. Better approaches have been suggested elsewhere.
Christopher Kreuzer wrote:Has there been any thought about revising downwards the 9 games in 3 years bit?
Not that I’ve heard.
Christopher Kreuzer wrote:Would it be too confusing to have an annual system running alongside the half-yearly one? You would get three grades a year. Two "six-month" ones and one annual one.
I don’t know about the enemy, but by God it would confuse me.
Sean Hewitt wrote:[The Rapidplay prototype] isn't a good model at all, and should be binned asap.
I think the ECF would be only too pleased to hear constructive suggestions for improvement. This sounds like the appropriate place for me to stop.
Except to repeat that I am not the ECF and cannot speak for its policies.