Chess Grading Destruction

[MartinCarpenter]

The type of chess probably isn't too hard. Congresses all one (class of) time limit, evening leagues another, 4NCL/County chess a third some oddballs but you'd be accurate enough.

Probably better nowadays to just calculate distinct grades for each of the different classes of time limits on demand?

Much like is already done with rapidplay grades, but available for people to look at via the website if interested. After all, if you're picking a team/board order for evening league time limits you'd want to know who plays well at those time limits

Might amuse people to know who's doing best in each competition each season too.

As well as data, infrastructure is the other big problem to all of this - the grading website is clearly non trivially under resourced right now

[Brian Towers]

Of course, it is correct, getting the organisers to input the data is better than sending it to graders who then input the information. In FIDE this would have the problem that the organisers sometimes deliberately falsify the information. e.g. not submitting data where their own players perform poorly. e.g. inventing games that have never occurred.

Stewart, can you explain how having the extra layer of administration would help eliminate this false information?
If an organizer is prepared to input false results into the system directly what stops him from sending false results to the graders?
Surely the extra layer of bureaucracy helps the falsifier as it inserts somebody else into the chain who can be blamed if they come to light?

[Roger de Coverly]

If an organizer is prepared to input false results into the system directly what stops him from sending false results to the graders?

Absolutely nothing other than the integrity of the grader in querying the false results. That's actually the point, to have someone independent of the organiser giving the result a little scrutiny.

[Stewart Reuben]

Well said Roger.
For international events, where players can get title norms, it is required that the games played be available. It wouls be too great a burden for the rule to apply to all rated games.
Events have to be registered in advance with FIDE and, if then the results not submitted for rating, questions are asked.

I would like there to be an ECF website where players can input their own games with full data of provenance. Everybody would benefit. Of course, that is not a grading matter.

[Brian Towers]

If an organizer is prepared to input false results into the system directly what stops him from sending false results to the graders?

Absolutely nothing other than the integrity of the grader in querying the false results. That's actually the point, to have someone independent of the organiser giving the result a little scrutiny.

The wonderful thing about the world wide web is that it allows a level of scrutiny unprecedented in the paper and pencil world. Scrutiny is performed by everyone looking at the results on the web. In case you hadn't noticed we are already in the 21st century.

[Brian Towers]

I would like there to be an ECF website where players can input their own games with full data of provenance. Everybody would benefit.

LOL!

From the people I've spoken to for whom chess is a competitive business that is the last thing they want. Too many of their previous games are already available for every Tom, Dick and Jonah (or his coach) to download and put through Stockfish in preparation for their next game.

[Roger de Coverly]

Scrutiny is performed by everyone looking at the results on the web.

There are those who express doubt about the reliability of some results coming from countries of the former Soviet Union. I don't believe anyone expressing such doubts has any ability to intervene, even with the checks supposedly carried out under a system of having named International Rating officers. If Federations are prepared to tolerate such activity (and the ECF wouldn't, or at least you expect it wouldn't), there's nothing individuals can do about it.

[Stewart Reuben]

Brian, I think your friends must be a bit confused. If their games are available to their opponents, then so are the opponents' games available to them. Of course they may prefer to play Chess960. but that hasn't really caught on. There would need to be a separate grading list for those games. Just as there should be for games played 3/1/0. But I convinced FIDE there are currently so few of those, that it wasn't worth doing.

Roger, I had the idea that FIDE should send, unannounced, people to do spot checks on events where allegations had been made. Nothing came of that.

There is little doubt that Myanmar manipulated the FIDE Rating System to get their players higher rated. Kevin O'Connell ameliorated that problem, dropping everyone in the federation by 100 points, and the problem has not been publicised since.

[MartinCarpenter]

Ah yes, but taking advantage of that would require a bit of work on their openings That's something an awful lot of UK club players seem to be moderately allergic too The 4NCL expanding to Div3 and making all the games available is starting to force it of course.

[Ian Thompson]

The type of chess probably isn't too hard. Congresses all one (class of) time limit, evening leagues another, 4NCL/County chess a third some oddballs but you'd be accurate enough.

Probably better nowadays to just calculate distinct grades for each of the different classes of time limits on demand?

The trouble with dividing what is currently standard play into subclasses is that you wouldn't have enough games in each subcategory to produce a reliable grade for most players. You'd also likely have widely differing grades in each class for many players because they happened to have a good season in one event and a bad season in another (e.g. my own results so far this season in these categories would be: congress 232; evening league 208; county/4NCL 194. But if I'd scored 1 less point in congresses and 1 more point in county/4NCL they'd all be within a few points of each other.)

[Stewart Reuben]

Martin Carpenter >Ah yes, but taking advantage of that would require a bit of work on their openings That's something an awful lot of UK club players seem to be moderately allergic too<

Indeed. But one of my objectives is to provide an environmnt where players improve their chess. One weapon is grading information, another games information, another prize or status improvement, etc. For me, grading isn't an end in itself. I think one grade, with input for all types, but weighted, may be the best. I would exclude blitz.

[Brian Towers]

Brian, I think your friends must be a bit confused. If their games are available to their opponents, then so are the opponents' games available to them.

No, Stewart, that's not true as you would have realized after a few moments thought.

A few weeks ago I bumped into an old teammate from 40+ years ago. He is still comfortably stronger than me and certainly not confused. He bemoaned the fact that many of his games are easily available on the internet and an up and coming junior (or rather the junior's coach) took advantage of this in a recent tournament and prepared an improvement in one of his pet variations. The junior won. My old friend had no reciprocal possibilities because the junior didn't have games available online.

Don't get me wrong. This is a perfectly legitimate tactic. I've used it myself on occasion but with previous games between me and the other player. It is almost always an asymmetric operation. The weaker player prepares for the stronger player. Almost never the reverse. There are stronger players than me who I play regularly against whom this tactic doesn't work for the simple reason that they either prepare for the game by looking up previous encounters and deliberately playing something different or perhaps they have just changed their repertoire but when it does work it is deeply satisfying. So, while I sympathise with my old friend I'm afraid I identify with that junior. I'm all in favour of a site where the gullible can enter their games but I will be a read-only user.

[Roger de Coverly]

The weaker player prepares for the stronger player. Almost never the reverse.

That's not necessarily true. It can often be a problem to find any games, but looking up the standard of their normal level of opponent gives an insight. One of the things I would screen for is whether they have any pet openings that I would need to decide whether to tackle or avoid. Things like the Morra or the Blackmar Diemar which although giving a pawn away for next to nothing in theory, in practice can come back and bite. Perhaps it depends how arrogant you feel as to whether you want to give them the pleasure of their favourite opening and then attempt crush it, or just avoid positions they might know inside out.

[Stewart Reuben]

Brian >they either prepare for the game by looking up previous encounters and deliberately playing something different or perhaps they have just changed their repertoire<

In other words, they seek to improve their play. Exactly what I am hoping to achieve as a chess administrator.
Generally speaking the stronger player outprepares his opponent. He partly does so by viewing the relevant games of stronger third parties.
I have given these matters more than a few moment's thought.

[MartinCarpenter]

Surely even preparation based on games often helps stronger players vs weaker ones? For one thing there's really quite a big difference in how much damage you can do.

With the 4NCL DIvision 3(N) there's now plenty of ~170's and under with quite a few games on the database. A fair number of them also seem to have incredibly predictable, not brilliantly prepared, sometimes marginally sound, openings.

That really is a very dangerous way to behave. Asking for considerable trouble if/when their opponents get time to prepare for them. Which isn't often, hence one reason (as well as the lack of games of course) they've got away with it historically.

200ish players? Certainly the ones I know round the York team tend to switch their openings around quite a bit which is I think how it has to be nowadays. Also work on them of course, so there's much less damage you can do.