Publishing ratings every month

[matt_ward]

HI all,

Why don't rapidplay and standard play grades get published every month?

Matt.

[Ljubica Lazarevic]

Because that'd be one hell of a job!

[Eoin Devane]

http://www.ecforum.org.uk/viewtopic.php?f=25&t=1906

[matt_ward]

Is it true if you only play 6 rapid grades before the next publication thats what your grade will be?

I heard a rumour if you play 10 games or anything below 30 then they take some from the previous year!

Is this true?

matt.

[Alex Holowczak]

Is it true if you only play 6 rapid grades before the next publication thats what your grade will be?

I heard a rumour if you play 10 games or anything below 30 then they take some from the previous year!

Is this true?

matt.

This will answer all of your grading questions. Or at least, all of the ones you've written above.

For completeness, the answer to your questions is no, and yes if you're not a junior, respectively.

[Christopher Kreuzer]

http://www.ecforum.org.uk/viewtopic.php?f=25&t=1906

I'm trying to post a comment in the thread you linked me to, but every time I click the link, I get logged out! I guess I will have to find the thread the hard way.

[Alexander Hardwick]

HI all,

Why don't rapidplay and standard play grades get published every month?

Matt.

One issue is accuracy. Each grade is based on fewer games. Therefore each grade is less reliable.

Another issue is impracticability. It took a little under 2 months this year for the new grading list to come out (counting from June 1st). Enough said: a grade would already be completely "out of date" by the time it is published.
It would also give some congresses an unreasonably short window for sending in their results. And play havoc with the grade-limited leagues.

I don't pretend to know a lot about the grading system, but I suspect it would be very costly as well.

Out of interest, what are your reasons for such a radical suggestion? I don't think many people, or even many juniors, are improving quickly enough month by month to merit a new grade.

[Carl Hibbard]

http://www.ecforum.org.uk/viewtopic.php?f=25&t=1906

I'm trying to post a comment in the thread you linked me to, but every time I click the link, I get logged out! I guess I will have to find the thread the hard way.

Try that...

[Sean Hewitt]

HI all,

Why don't rapidplay and standard play grades get published every month?

Matt.

One issue is accuracy. Each grade is based on fewer games. Therefore each grade is less reliable.

Another issue is impracticability. It took a little under 2 months this year for the new grading list to come out (counting from June 1st). Enough said: a grade would already be completely "out of date" by the time it is published.
It would also give some congresses an unreasonably short window for sending in their results. And play havoc with the grade-limited leagues.

I don't pretend to know a lot about the grading system, but I suspect it would be very costly as well.

Out of interest, what are your reasons for such a radical suggestion? I don't think many people, or even many juniors, are improving quickly enough month by month to merit a new grade.

FIDE seem manage it every other month without too much trouble.

[Alex Holowczak]

FIDE seem manage it every month without too much trouble.

They do it every 2 months (for now, at least), and they have a team of professionals, probably. I reckon 1 month may be a bit too much, but I don't see why it couldn't be more often than it is.

[James Coleman]

Not only that, and apart from the reasons already stated, if you play one FIDE tournament and do terribly, the resulting change to your rating is not going to be that much, maybe a 5% swing if you really have a nightmare, whereas that would not be the case with ECF grades. Hence the resulting volatility of the ECF grades make the idea impractical.

Incidentally reducing the volatility is the rationale behind counting games from previous years for those players who have not played much.

[Sean Hewitt]

Not only that, and apart from the reasons already stated, if you play one FIDE tournament and do terribly, the resulting change to your rating is not going to be that much, maybe a 5% swing if you really have a nightmare, whereas that would not be the case with ECF grades. Hence the resulting volatility of the ECF grades make the idea impractical.

Incidentally reducing the volatility is the rationale behind counting games from previous years for those players who have not played much.

James, is that right?

If ECF grades were published monthly (for arguments sake) and you scored 0/5 in some event, why do you think that the change of grade is going to make the system volatile?

[Roger de Coverly]

Incidentally reducing the volatility is the rationale behind counting games from previous years for those players who have not played much.

There is a rational way within the ECF/Clarke system of publishing more frequent grades. The outline is as follows:-

(1) Reinstate the "adult" approach for calculating junior grades
(2) Introduce a method whereby "new" players get an established grade after x games (x=9 perhaps)
(3) Persuade leagues and congresses to submit results within days of games being played
(4) Keep the concept of the annual update of the calculation base
(5) Introduce and publish an additional calculation best defined as "what your grade would be at the end of the season if you didn't play any more games" . For example I've already played 20 games this grading season, so my "current grade" would be (published 2010) * 1/3 + (calculated 2010-11) * 2/3 so as to weight to 30 games.

This is similar to what the Scots publish. They give an up-to-date figure but don't change the calculation base. This fiffers from FIDE who now update the calculation base every two months.

Every player would then have two grades which would represent performance last season and performance so far this season.

[Sean Hewitt]

Every player would then have two grades which would represent performance last season and performance so far this season.

Roger - some already suggest that having new grades mid-season would be "confusing". I can't imagine that they would find two grades simultaneously to be simpler!

[Mike Truran]

Sean, I had a dreadful season in the 4NCL (the only FIDE rated games I played) and my ELO rating only went down by around 2% or so. I had a much better season overall (4NCL and Oxfordshire League combined) and my ECF grade went down by around 3%. Of course one example isn't going to prove a lot, but in the past as someone who only plays 15 or so games a season my ECF grade has regularly fluctuated by around 3% up or down.

I suppose, thinking about it, last season I played stronger players in the 4NCL, which I suppose would dampen the adverse FIDE rating effect, while getting better Oxfordshire League results against weaker Oxfordshire Chess League players would have less of a beneficial effect on my ECF grade than if I had got the results against players of a similar strength to those in the 4NCL.

Anyway, I guess that a 3% swing or so doesn't really amount to much in the greater scheme of things.

Not sure what it shows really other than that I don't really understand the mysteries of the FIDE rating and and the ECF grading systems!