GRADING

BCF News SCCU home page

updated 28.7.03
BCF GRADING
GRADING LIST 2003
rjh 28.7.03
The List appeared in Edinburgh on the 25th July. Cover is much the nicest colour we've had in recent years (but be warned it's not as washable as it looks). The other thing you notice is that it's a lot bigger than last year's, with a bigger price to match. £15 as opposed to £13, with 263 pages of grades instead of 194. This does not mean thousands more players - there are actually about 300 more - but a return to the old practice of printing up to six clubs per player.
It's not for me to review the List, since I was heavily involved in its production. Let's just observe that it comes with a big Amendments list inside, to repair the accidental omission of the Thames Valley League. Pity, that.
I can, at least, statisticulate. Have some facts.

Players published
Juniors published
Average age, all juniors

2002
12250
2304
11.7

2003
12468
2720
11.3

STANDARD
Players published
Juniors published
A grades
B grades
C grades
D grades
E grades
Mean grade A-E
Median grade A-E

2002
10616
  1060
  1823 (av 130)
  1599 (123)
  2775 (115)
  2672 (106)
  1747 (102)
          114.6
          113

2003
10436
  1129
  1804 (131)
  1628 (121)
  2572 (115)
  2667 (107)
  1765 (100)
          114.0
          112

RAPID
Players published
Juniors published
A grades
B grades
C grades
D grades
E grades
Mean grade A-E
Median grade A-E

2002
3457
1787
  518 (98)
  249 (96)
  485 (102)
1297 (87)
  908 (77)
          88.8
          83

2003
4015
2214
  590 (90)
  318 (93)
  513 (98)
1469 (82)
1125 (65)
          81.3
          76

All the 2002 figures are, in fact, revised December ones after the insertion of some late events. The number of published players in July was about 75 less than stated. This year late results will not be accepted, but nothing is known to have missed the deadline anyway.
     The total number of halfgames graded this year was 293,760 (17,917 players). Last year, 280,151 (16,950). Standard chess graded has declined slightly, and Rapid has gone up rather a lot (21%, in terms of game results). You'll notice that most of the Rapid influx seems to have been Junior, and that its mean grade has gone down nearly as much as its activity has gone up. There will be a connection there, but the new system may also have something to do with it. It is said to lack the old one's tendency to overgrade new players.

So what is new? The central program itself, for a start. It is Howard Grist's completely re-written version of the original 1998 Chris Howell system. In 2000 we moved from this to a temporary "hybrid" system which was never satisfactory, because it started afresh for new players with every single event they played in. (The same player might count as 100 in one event and 50 in another, depending on his results in the event.) This year, in addition to other modifications, calculations have been done centrally in one go, taking all of a new player's results together.
     There have also been administrative changes. One of them has been to pay someone, for the first time, to manage the system on a day-to-day basis. That's where I come in. Processing of files has also been a paid job as it was last year. But otherwise the Grading team is unchanged, and it's not so much a question of changed personnel as changes in the general approach.
     Partly it's been about new players. Isn't it always. The software has always, since 1998, been capable of accepting results for unknown players and generating a reference number ("BCF Code") for them. But there were always bound to be duplicate and triplicate entries, which had to be weeded out by hand, and this was taking so much time at the end of the season that when they introduced the hybrid they also took away the grader's right to enter results for new players. Instead he had to request codes for them in advance. The remedy proved no better than the disease, because of the time taken to provide the codes. Most events have new players in them, and the result was that everything piled up at the end of the season as always. Hence the revised list in December, and if anything December was early by previous years' standards.
     So this year it was decided to use the system the way it was meant to be used, and let graders enter new players ad lib, and avoid the end-of-year logjam by asking them to send their January congresses in January or February instead of waiting till June. (Seems rather obvious, really!) In addition, with more paid time available, the BCF has been able to communicate more readily with graders, to respond more quickly to their submissions, and to pick up problems as they arose. The new software has helped in this process. In short, I would say the new approach has worked. Results have come in very much more quickly. This has more or less eliminated the late panic and - I hope - made the coverage more complete. What it has done for accuracy remains to be seen. There will always be mistakes, at the grader's end no less than the BCF's, and some have emerged already. But the Federation, with no late results to insert, will at least attain its objective of having the corrections sorted out far earlier than December. (In fact next year's grading can start tomorrow if it wants.)
     I think the new central program has passed the test. Perhaps I should say that it did not wipe the Thames Valley League. A human did. The event was in, and processed, with time to spare. And then got wiped at some point, and no one noticed until the List was with the printer. If the program was at fault, it was only in not ringing loud enough alarm bells. I think it will be ready for this one next year, if anyone tries it again.
     Issues for the future
There's one in particular that is causing concern on the Grading team. Negative grades. When the scale was fixed, I don't know how many years ago, it wouldn't have occurred to anyone to grade U8 tournaments, or even U80, and the idea of a negative grade would have seemed laughable. It isn't now. Don't look for them in the printed list. If you work out at minus 70, and people do, your grade is printed as 1. But it will count as minus 70 in calculations, as the program stands currently. Is this honest? What's the solution? Print it as minus 70? Print it as 1 and count it as 1? Add 200 to everyone's grade? Switch to Elo? (But be careful, you don't have to go much below minus 70 before you're negative in Elo as well.) The problem has been around for years, but it has come to a head and we're past the point where we can ignore it and hope it will go away. It will only get worse, if the increase in junior grading continues. The new calculation routine will even accelerate the plunge, because it's more honest than the old one about very low grades. My solution? Add 200 points, or 300, or 500, to everyone's grade. And do it in the Grading List 2004. It will be an upheaval, but postponing it another year won't help. Opinions invited.
     Note. Minus 70 was an extreme example, as A-E grades go. But it happens. There are 166 Rapidplay A-E grades below zero, and 99 of them in double figures. There are a handful in three figures, if you count players who don't qualify for an A-E grade.

Earlier material is in the Archive.