ELO vs ECF

General discussions about ratings.
andrew martin

ELO vs ECF

Post by andrew martin » Sun Apr 06, 2008 8:54 am

In my opinion the ECF system of grading should be replaced by the ELO system, immediately. What do others think ?

This argument is on the same level as the adjudication/adjournment debate. To modernize English Chess we have to scrap all adjournments and adjudications immediately.

We all talk of the decline in English Chess. Here are two solutions to part of the problem. We must stop even debating the above issues and just accept that they are going to happen. Why delay ?

Andrew

Ben Hague
Posts: 34
Joined: Thu Apr 05, 2007 6:59 pm

Re: ELO vs ECF

Post by Ben Hague » Sun Apr 06, 2008 11:05 am

I agree that the Elo system is better than the ECF one, but as far as I can see there isn't that much difference. For me the fundamental problem is that you're taking a small number of games played under all sorts of conditions, with widely varying time controls, involving isolated groups and trying to come up with a number that accurately represents playing strength. It works better at the top international level because you have much more standardisation and mixing of players, but for English chess I can't see the improvement being big enough to get worked up about. I'm interested to know why you see it as important?

andrew martin

Re: ELO vs ECF

Post by andrew martin » Sun Apr 06, 2008 6:52 pm

Why is the ECF system different to virtually every other country in the world? To me, this makes no sense.

You said it yourself-ELO is superior, so we should go with it.

Andrew

User avatar
Greg Breed
Posts: 723
Joined: Thu Apr 05, 2007 8:30 am
Location: Aylesbury, Bucks, UK
Contact:

Re: ELO vs ECF

Post by Greg Breed » Mon Apr 07, 2008 2:52 pm

Do all the other countries that use the ELO system run leagues? The emphasis in these forums seems to suggest that chess is declining in Britain (and/or England) and yet we probably have more local, county and national(?) leagues than any other country. I don't know for sure, but I guess most other places usually have mostly tournaments. This changes things:

These days you can't adjourn or adjudicate in a tournament (except maybe that one in Southend) however, this happens frequently in League matches which are played in the evenings after work. They simply can't all be finished on the night. A lot of players don't like blitz or rapid-play chess and therefore do not like quick-play finishes either. Their point is that it reduces chess to bashing moves out without thought. I completely understand this viewpoint even though I'm an advocate of finishing on the night.

It would make life a lot more difficult for all the league controllers and graders out there (who do it all for free in their spare time) to have to break the season into four different rating periods. I don't know if it works in other countries but one or two adjourned games means that the match result is incomplete and the game cannot be submitted for grading until it is finished. This escalates every time there is an adjournment as the controller now has more to keep track of. They would now also need to keep track of which games have been submitted for grading as well as which games are outstanding for results.

I think the ELO system would only be feasible if a better system of submitting results for grading were introduced for the leagues. It is simply too time consuming at the moment for a volunteer who has their normal life to maintain as well playing in leagues and tournaments and running a league and other activities.
Hatch End A Captain (Hillingdon League)
Controller (Hillingdon League)

Chris Majer
Posts: 85
Joined: Wed Jan 24, 2007 11:29 pm

Re: ELO vs ECF

Post by Chris Majer » Tue Apr 08, 2008 11:57 pm

Andrew

Why change to a system that isn't working either? I refer you to Jeff Sonas: http://www.chessbase.com/newsroom2.asp?id=562

He seems to recommend that Elo should be relaced by a linear system - maybe the ECF has got it right!
Chris Majer
ECF Chief Executive

Paul Stimpson
Posts: 109
Joined: Mon Feb 25, 2008 10:52 pm
Location: Essex
Contact:

Re: ELO vs ECF

Post by Paul Stimpson » Wed Apr 09, 2008 3:24 pm

I have to agree with Andrew that we should move to ELO ratings as soon as possible. If FIDE relax the playing session time for FIDE ratings the ECF could will end up with a redundant product. FIDE have lowered the entry level to 1400 and more and more events are popping up with FIDE rating opportunities. To my mind the writing is on the wall for ECF grades and about time to.

andrew martin

Re: ELO vs ECF

Post by andrew martin » Thu Apr 10, 2008 12:36 pm

Much has been made of the decline in English chess over the past 20 years. Hardly any tournaments for our players, very poor support from the government refusing to recognise chess as a sport, good young players fading away simply due to lack of opportunities,paltry prize money in local events and a federation which may or may not have the support of the average chess-playing man in the street.

My own view,despite the criticism that has been levelled at the ECF, is that everyone WANTS to have a strong federation. OK, the ECF therefore has to show some teeth

The first thing we must do is to get our own house in order. We have to modernize. The grading and adjournments issues are almost non-debatable. The Federation has to be strong and replace the grading system with ELO ratings and insist all games in all leagues and events must finish in one session. Time controls can be managed to provide an interesting session. There may be a howl of protest; well those are the people who are REALLY responsible for holding chess back in England, not the ECF.

Frankly,you either understand this or you don't.

Andrew
Andrew

Ian Kingston
Posts: 1071
Joined: Sat Apr 14, 2007 3:16 pm
Location: Sutton Coldfield
Contact:

Re: ELO vs ECF

Post by Ian Kingston » Thu Apr 10, 2008 2:28 pm

I fully agree that we should switch to Elo ratings. Apart from bringing England into line with the rest of the world, which would be sensible in principle, there is at least one practical advantage. An increasing number of English players take part in foreign competitions (and vice versa), and it makes sense to have everyone on the same type of system.

We would still have the problem of some players having a national rating and a (different) FIDE rating, but that seems not to have troubled other countries.

I'd like to know what would actually be involved in making the switch. As far as I can tell, we would need to:

1. Convert all ECF grades to their Elo equivalent
2. Collect results as at present
3. Process results using a different computer program

Under point 1, a little statistical work would be needed to find an accurate conversion. Neither of the two formulas currently in use is accurate enough.

One objection that might carry some weight is that players can calculate ECF grades and performances in their heads or on the back of an envelope. To calculate an Elo rating you really do need a computer. I don't see that as a major problem though – a rating calculator is not a difficult thing to put on a web site.

There are other issues (frequency of publication rating lists, for instance), but those don't seem to me to be relevant to the question of whether or not we should make the switch in principle.

So what else might I be missing? What possible major stumbling block could there be?

User avatar
Charles W. Wood
Posts: 554
Joined: Wed Feb 13, 2008 11:50 pm
Location: Bradford, West Yorkshire
Contact:

Re: ELO vs ECF

Post by Charles W. Wood » Thu Apr 10, 2008 5:15 pm

I'm a bit stuck on this one, I am very proud of the ECF grading system, but the ELO one looks good too. Join the world or not. :twisted: When I think of the Chess for Schools project unfolding over the next year subjects like this should be addressed, if we have something that matches the world system all the work that is about to go into developing schools across England will last so much longer. E.G.: little Johnny and little Janet from Cumbria are in Spain on Holiday with their parents and thy spot a chess tournament during their stay, if they have an ELO rating they can take part, but if they have an ECF grading they will be put off.

I also believe the greater number involved will create MORE competition between juniors as the numbers are more finite.

We as a nation either want to be part of the greater world or we do not. So from my point of view we should move towards ELO and get some of our clever graders working towards a better world system. I am quite surprised we haven't done it already. :roll:
Charles W. Wood
Captain of Legion

User avatar
Nigel Wright
Posts: 246
Joined: Tue Apr 10, 2007 8:49 pm
Location: Derbyshire, England
Contact:

Re: ELO vs ECF

Post by Nigel Wright » Fri Apr 11, 2008 2:11 am

I like being different from the rest of the world, but I'm all for ELO. It seems stupid someone having both types of grade, as it is not a very good indication of their current playing strength, especially when the FIDE ratings are updated more often than ECF...
To Drink or not to Drink, that is the question.

I Drink therefore I am.

I'm not as think as you drunk I am.

User avatar
Greg Breed
Posts: 723
Joined: Thu Apr 05, 2007 8:30 am
Location: Aylesbury, Bucks, UK
Contact:

Re: ELO vs ECF

Post by Greg Breed » Mon Apr 14, 2008 12:12 pm

I agree in principle that we as a nation should adopt a standard that is used by the rest of the world (although, strictly speaking, the USA is not even included in that), but that standard should be correct. The article that Chris Majer pointed out that was written by Jeff Sonas is clearly aimed at improving the correctness of the ratings formula. How would this affect us as a nation?
For now, to keep things simple, I want to consider only four classifications of time controls. The "Classical" time control, of course, refers to the traditional time controls of two hours for 40 moves, one hour for 20 moves, and then half an hour for the rest of the game. "Modern" (FIDE) controls are at least 90 minutes per player per game, up to the Classical level. "Blitz" controls are always five-minute games with no increments, and "Rapid" has a maximum of 30 minutes per player per game (or 25 minutes if increments are used).
I have never in my life played the "Classical" time control and I'm fairly certain that most of the chess players in this country haven't either! This means that all our games would receive the "Modern" classification.
Give Classical games an importance of 100%, whereas Modern games are 83%, Rapid games are 29%, and Blitz games are 18%.
Effectively this would mean that most of us under 200 ECF (who generally play the "Modern" time controls) would only ever receive 83% of the grade outcome as a maximum!

Clearly this system is aimed mostly at improving the rating system that favours using "Classical" time controls.
...Although later on Jeff describes a system that follows a formula based on time control for (an average) 40 moves (game), it would still mean that most of the chess playing populace will never likely receive 100% for their games.
Hatch End A Captain (Hillingdon League)
Controller (Hillingdon League)

User avatar
John Upham
Posts: 7179
Joined: Wed Apr 04, 2007 10:29 am
Location: Cove, Hampshire, England.
Contact:

Re: ELO vs ECF

Post by John Upham » Thu Apr 17, 2008 3:49 pm

Andrews' views are clearly and eruditely put.

I too, find that the ECF is still defying the ( spirit at least ) FIDE Laws of Chess by continuing to rate games decided by adjudication thus :

1.1 The game of chess is played between two opponents who move their pieces alternately on a square board called a `chessboard`.

A game of chess is NOT between two opponents plus third parties, their mates and / or various computers.

In a similar vein, adjournments should not be rated as third parties can be involved and should not IMHO.

I'm also a strong believer in aligning our rating system with FIDE and ditching the Clarke System. It was nice whilst it lasted?!

To avoid further decline, we do not have choices here.

I look forward to the point when the ECF has fallen behind so much that we go cloth cap in hand to the Turkish Federation for support and guidance. We will then get the same level of international respect as English tourists on a Club 18-30 holiday.

Did anyone notice that Brian Eley has been invited to participate in the 2008 Championships by virtue of qualification?
Will Interpol be providing his travel bursary?

I wonder how many dead people have also been invited?

John
British Chess News : britishchessnews.com
Twitter: @BritishChess
Facebook: facebook.com/groups/britishchess :D

Dr Andrew Cula

Re: ELO vs ECF

Post by Dr Andrew Cula » Thu Apr 17, 2008 5:46 pm

Ian Kingston wrote: I'd like to know what would actually be involved in making the switch. As far as I can tell, we would need to:

1. Convert all ECF grades to their Elo equivalent
2. Collect results as at present
3. Process results using a different computer program
I'd imagine the BCF stores grading data for the last 'x' years, and so could (for example) assign ELO ratings as of BCF ratings in 2000 then process the next 8 years of data to fix the grades.

User avatar
John Upham
Posts: 7179
Joined: Wed Apr 04, 2007 10:29 am
Location: Cove, Hampshire, England.
Contact:

Re: ELO vs ECF

Post by John Upham » Thu Apr 17, 2008 6:02 pm

I'd suggest:

1. First task would be to deal with the alleged rating deflation and produce a clean set of ratings.
2. Implement the most reliable version of the rating calculator.
3. Ask /tell players who already have an FIDE Elo if the ECF can tandem their existent FIDE Elo into the ECF system.
4. Do a one time conversion from corrected ECF into Elo.
5. Turn off the old system and put in Hastings Museum.
6. Publish new ratings using new system.
7. Update every three months.
8. Join the modern chess world at last.


Am I dreaming or would Sir Alan Michael Sugar fire me?

John, The Apprentice
British Chess News : britishchessnews.com
Twitter: @BritishChess
Facebook: facebook.com/groups/britishchess :D

Chris Majer
Posts: 85
Joined: Wed Jan 24, 2007 11:29 pm

Re: ELO vs ECF

Post by Chris Majer » Fri Apr 18, 2008 11:18 pm

John wrote
Am I dreaming or would Sir Alan Michael Sugar fire me?
Yes, you're fired because you've ignnored what your Customer has already told you! A year ago, ECF Council decided that the ECF didn't want standard play lists more than once a year.
Implement the most reliable version of the rating calculator.
Why do you think that's the FIDE methodology? Perhaps the ECF have it right?
We could of course use ECF methodology, but generate a four figure rating. The Welsh Chess Federation have done this in the past and that may be the way they are still calculating ratings.
Ask /tell players who already have an FIDE Elo if the ECF can tandem their existent FIDE Elo into the ECF system.
You are assuming that the FIDE system is working better than the ECF. I wouldn't concur with that for Elo ratimgs close to/below 2000.

Why do you think that having a four figure rating than appears to be equivalent to Elo but isn't, is better than a three figure grading that you know you have to convert?
Chris Majer
ECF Chief Executive

Post Reply