English Chess Forum

Hi folks

This is the 40 point rule from the ECF website:

"There is a proviso that if your opponent's Grade differs from yours by more than 40 points it is assumed to be exactly 40 above (or below) yours. This is to prevent a player increasing his Grade by losing to a much stronger player, or decreasing his Grade by beating a much weaker player."

However, as stated the rule has the apparently unintended consequence that if you beat someone >40 points higher than you, your grade is capped as if they were only 40 points higher than you. So if an improving junior graded 100 has a season consisting of beating me 30 times, his grade will be 190 only. Yet clearly his grade should be higher than mine, i.e. over 200.

Or am I missing something?

_________________
http://streathambrixtonchess.blogspot.com/

Yes; juniors are treated as ungraded.

A junior graded 100 will actually be U when considering their next year's grade. They then get given an estimated grade by calculating their grading performance based on their results this season. You then grade the lot of the games, with your junior opponents' grades being taken as the new estimate, not their previous season's grade.

If an adult does the same, this doesn't happen. However, I don't think someone graded 100 beating someone graded 200 thirty times in a row, and playing no other graded games whatsoever, is a scenario worth worrying about on a practical level.

_________________
February 16-17: 2013 BUCA Team Championship
April 20: 2013 British Blitz Championship
July 27-August 10: 100th British Chess Championships

That seems correct to me

There is also another inherant problem, my grade current grade is 193, therefore every player I play below 153 is graded at that grade and thus only get 203 points for a win, unfortunately my current performance is actually above this (around 208), therefore playing "weak" players is actually detrimental to my overall performance, even though I am winning these games

This one of the problems with the basic average grading calculation, as in the FIDE system you can never lose points for winning a game

In the mark one version of the grading system back in the 1950s, no such rule existed. With the growth in weekend tournaments towards the end of the 1960s, mismatches became more commonplace and as you state, you would lose points by beating someone more than 50 points below you. Moving it to 40 points was a quick hack to the system which had the effect of removing the worst of the anomalies.

The anomalies are still there of course and put a brake on the grade change for anyone rapidly improving or rapidly declining. It also applies the other way. If you have a grade of 100, get paired and lose against a player of 180, then you score 90. If you have a grade of 130, your survival chances aren't much better, but you get a performance of 120.
That said, for the most active player, the ECF system has less inertia (memory of previous ratings) than Elo style systems.

The international Elo system used to cut off at 350, but it's now been moved to 400.

The 40 point rule is then biased against lower graded players who are not juniors and improving. For example a 100 could draw with a 180 and only get 140, etc.

True and it's always been so. Apart from juniors, players improving by 40 points are very much a minority. This can be tested by comparing last year's grade to this year's.

Isn't somebody performing over 10 points above their current grade always affected by the 40 point rule, as I have pointed out.

It is not a major problem as it used to be due to the recalculation of ECF grades, but is a problem which I think we are stuck with unless the grading system is totally converted into the FIDE system, where this problem will never happen, due to the fact that your points gain are calculated on a game by game basis rather than an annual average

Seems so, although this isn't just a problem of the 40 point rule. E.g. as a 163 if I beat a 133 then I get the full 50 points - 183 rating performance. I'll lose out, though, if I'm actually playing at 190 or 200 for the season (obviously this is very much a hypothetical example in my case). The problem seems not so much the 40 point rule as playing folk graded quite a bit below you.



Need we be stuck with it though? Why not say: 40 points max difference if you lose to a higher graded person. If you *beat* the higher graded opponent, though, you get the full wack. Admittedly this wouldn't solve your scenario, but it would improve things a bit.

_________________
http://www.streathambrixtonchess.blogspot.com

"Of all the stupid questions so far, This one is particularly annoying, and irrelevant and when I meet you, I propose to write it down on cardboard and nail it to your face."
- Jerry Sadowitz

Thanks for the replies. It seems to me to be crazy for grading gains from wins (or draws) to be *penalized* by the strength of your opponents!

_________________
http://streathambrixtonchess.blogspot.com/

I suspect only if you play a lot of much lower or much higher graded players. If you were beating 190s every game then with your grade you would get 240 regardless of being 203 or 208 etc

I'm sure this has been covered before, but I thought I'd ask anyway - Why does the ECF prefer it's grading system to that of Elo?

_________________
The FullPonty
Nidum Chess Club

Over the years I seem to have scored more -90s than +90s so I'm all in favour of the rule as it is.

The ECF system predates the existance of the Elo system. Our grading system goes back to the 1950s, FIDE only had ratings established in the 1970s (late 1960s?). We've just never switched to Elo. Personally, I would be in favour of switching to Elo.

_________________
February 16-17: 2013 BUCA Team Championship
April 20: 2013 British Blitz Championship
July 27-August 10: 100th British Chess Championships

The ECF grading system pre-dates the Elo system. It is designed to deal with the situation where most of the results arrive on an annual basis (i.e. leagues), not to deal with more frequently updated grades. As Prof Elo points out in his book the ECF and Elo systems are basically equivalent, considering the grades/ratings that players will end up with, if they are fed the same data.

I have never had an explanation of why the 40 point rule applies to drawing with or beating much stronger players (or drawing/losing to much weaker)

Anyone??

_________________
"Set up your attacks so that when the fire is out, it isn't out!" (H N Pillsbury)