Effect of activity on grade changes

General discussions about ratings.
Roger de Coverly
Posts: 21301
Joined: Tue Apr 15, 2008 2:51 pm

Re: Effect of activity on grade changes

Post by Roger de Coverly » Sun Oct 17, 2010 12:50 am

Peter Rhodes wrote:
Why not just calculate it yourself - this is the formula :

=((((10^((DIFF)/400))+1)^-1)+RES-1)*2*KFACTOR

where DIFF is the difference between your grade and your opponents.
RES is the Result (WIN=1, DRAW=0.5, LOSS=0)
KFACTOR - consult FIDE rules, but I think 16 is still used.
.
Be precise though if you want to emulate the FIDE website. The international formula uses a table of look up values derived from this formula, not the direct formula.

Also the K factor is 10, 15 or 30. From memory K=10 if your rating has ever exceeded 2400, K=15 for most people, K=30 for players who haven't played enough games to qualify for K=15.

Ljubica Lazarevic
Posts: 132
Joined: Mon Jul 13, 2009 7:00 pm

Re: Effect of activity on grade changes

Post by Ljubica Lazarevic » Sun Oct 17, 2010 11:57 am

I wonder if there could be a slight tweak to the current ECF grading system for over 30 played games per year? E.g. have some sort of a weighted average system.

Let's say someone has played 90 games over a year, still using the +-50 for a win/loss. Let's use a 'windowing' system of 30 games. The first 30 games weigh 14.3% towards the final grade, the next weigh 28.6%, and the last 30 weigh 57.1%. It's crude, but easy to implement and would perhaps be a fairer reflection on a person's actual performance.

Mike Gunn
Posts: 1022
Joined: Wed Apr 11, 2007 4:45 pm

Re: Effect of activity on grade changes

Post by Mike Gunn » Sun Oct 17, 2010 12:17 pm

I don't agree with this. I find my performance varies during the year according to the events I'm playing in (what varies is the number of juniors, players from other regions etc). This scheme would tend to produce an oscillating grade (well, one which oscillates more than at present!)

Ljubica Lazarevic
Posts: 132
Joined: Mon Jul 13, 2009 7:00 pm

Re: Effect of activity on grade changes

Post by Ljubica Lazarevic » Sun Oct 17, 2010 12:21 pm

I think a lot of that oscillation occurs from the grades only being published once a year. 2-3 times a year would help accelerate junior grades...

Mike Gunn
Posts: 1022
Joined: Wed Apr 11, 2007 4:45 pm

Re: Effect of activity on grade changes

Post by Mike Gunn » Sun Oct 17, 2010 12:33 pm

I think the way to deal with that is rather more regular lists, rather than introducing a weighting function. The key point is that each list would include a year's games giving the same weighting to all leagues/ tournaments that you play on. I don't see the point of introducing a mechanism which would distort the majority of grades (i.e. adult ones) because of the problem with junior ones.

User avatar
Christopher Kreuzer
Posts: 8806
Joined: Fri Aug 06, 2010 2:34 am
Location: London

Re: Effect of activity on grade changes

Post by Christopher Kreuzer » Sun Oct 17, 2010 12:42 pm

Mike Gunn wrote:I don't agree with this. I find my performance varies during the year according to the events I'm playing in (what varies is the number of juniors, players from other regions etc).
This brings us back to a point I raised earlier, where I asked whether people find they perform better at certain times of the season, or in particular events than in other events. Last year, for example, I performed atrociously in county matches, scoring 0/4. But that did lead me to think about how in county matches you are nearly always matched (on the higher boards at least) against someone of about the same grade. Which is different from the situation in league and congress chess, where the grading range of your opponents (in my experience) tends to fluctuate somewhat.

And people often say that they find they perform poorer in evening league chess after work, compared to weekend congresses. But then the type of opposition encountered varies as well. The main factor is stil how well you play and how well your opponent plays, but there do seem to be variations in form between different types of chess, which has been mentioned in the thread on different grading systems.