Which rating should we use?

[Larry Bevand]

The Festival St-Denis was organized by the Chess'n Math Association. We do not usually organize open chess events, as our niche is scholastic chess. However, because the Quebec Open moved to Quebec City this year...and with an August date instead of the traditional July dates, there was a void on the Montreal scene so we felt it would be good to provide a 9-round...one game a day event in the city.

I would appreciate the opinion of folks on the following question: What rating should an organizer use when determining sections and prizes that a player is eligible to play?

This is what we used:

the highest of your permanent rating between FQE, CFC and FIDE.

I looked around at what others were doing (after the fact of course :) ) and this is what I found:

2017 Calgary International:

Based on current FIDE rating (CFC ratings will not be used)

2017 Canadian Open

The following ratings (as of July 1st, 2017) will be used to determine player section eligibility:
For Canadian residents outside Quebec: CFC rating
Residents of Quebec: FQE rating
US Residents: Higher of CFC and USCF ratings
All other foreign players: Highest of CFC, FIDE, or other available rating
For any player in the Master section, FIDE will be used for eligibility if it is higher than other available ratings.
Unrated or provisionally rated players may not win more than C$1000 except in open section.

2017 Canadian Amateur: Eligibility: Players MUST play in the highest section they are eligible for, based on current rating. FIDE or FQE Ratings taken at par

2017 Montreal Open: Rating used (Pairings & Sections) FQE for Quebec residents / highest of FQE or CFC for the Rest of Canada. If you have had a rating in the past that is higher then you can not play in a section that is less than 100 points from your highest rating.


2017 Quebec Open - For the A section, FIDE ratings will be used for players who have them, FQE for Quebec players, CFC for Canadians from other provinces, and National ratings for players from other countries.

2017 Toronto Open - I could not find the criteria...can some one help me here please.


2017 Varennes International ($16,000 Guranteed Prize Fund...I could not find the criteria...can some one help me here please.

2018 Trois-Rivieres Open ($11,300 Guaranteed Prize Fund) Ratings: For Section A, FIDE ratings will be used. if a player does not have a FIDE rating, the highest valid rating between that of the FQÉ and CFC will be used.
For all other sections, the highest valid rating between that of the FQÉ and CFC will be used. If you have had a rating in the past that is higher then you can not play in a section that is less than 100 points from your highest rating.


In hindsight I would have done it differently for our event. Is there an ideal solution? Is there a pôlicy that our National Federation has adopted that I know not? Would it not be nice to have a standard approach to this...or have I been out of the loop so long that I know nothing. (I like the rules for the Canadian Open but should we distinguish between permanant and provisional ratings when determining which section a player can participate?)

Also, I have noticed another rating system (why not...if CMA can have one afterall :) ) the URS System...

URS (Universal Rating System) Ratings
(Updated every month, following the updates on universalrating.com)

The Universal Rating System™ (URS™ for short) is a new sport’s rating system designed to assess the relative strength of participants. It was developed as a result of a collaborative research project funded by the Grand Chess Tour, the Kasparov Chess Foundation and the Chess Club of Saint Louis. Unlike Elo rating, the URS calculates a single overall rating for every chess player which is optimized to represent their strength at classical chess based upon their game results across all time controls (classic, rapid and blitz). The research claims to indicate that incorporating both slow and fast-play results provides better information regarding a player’s overall chess ability, with the enlarged pool of data aiding the overall accuracy of the rating system and improving the ability to estimate the strength of a player at classical time limits. Read more at the official website universalrating.com.

The ratings are updated (as of today) monthly. Chess-db.com is displaying those ratings on the players' profiles and will be offering various statistical information on this and other pages.

Should this also be taken into consideration?

Larry

[Paul Leblanc]

Re: Which rating should we use?

In BC tournaments we emphasize CFC ratings as long as they are current. Here is Victoria Chess's policy:
Ratings:
The player has an obligation to disclose all previous ratings in other countries and regions.
The TD has the absolute right to use any rating that he deems appropriate although in general he will follow the terms below.
The primary rating used is the CFC regular rating provided that the player has an established rating and it is current.
If the player does not have a current, established CFC regular rating but does have a current, established rating in another 'official' system (excluding CFC active ratings), then that rating will be used. If in this situation, a player has multiple current established ratings in 'official' systems, FIDE will be used as first choice, the player's 'home' system will be used as 2nd choice, and otherwise the TD will use his discretion. The TD will decide if a given rating is 'current'. In general, we consider 10 games in the last year in that system to be current.
If the player does have a rating(s) in an official system(s) but they are not current or established, then the TD will use his discretion but will be biased towards the player's home system and towards more recent activity.
If the player does not have any official ratings but does have a rating in an unofficial system (e.g. Victoria Chess Club rating, Vancouver Active rating) the TD may, at his discretion, use that rating. Internet ratings will typically not be used.
Conversion of other ratings to CFC equivalent will be at the TD's discretion but will typically be done at par.
This rating will be used for both pairing and prize distribution. Pre-registration lists, wall charts, etc. will use this rating (perhaps with a notation to indicate source, e.g. FIDE)
For the purposes of determining if the player needs to pay a "play up a section" fee, any reasonable rating meeting the section rating requirements will be sufficient for us to waive that fee (although we may use a different rating as determined above for pairing and prizes).
Unrated players are not eligible for class prizes, only the Premier section place prizes and the unrated prize (top score by an unrated in any section). Unrated players may play in any section but are subject to the "play up a section" fee. It is strongly recommended that unrated players with no previous tournament experience start off in the lowest section.

[Fred McKim]

Re: Which rating should we use?

The Festival St-Denis was organized by the Chess'n Math Association. We do not usually organize open chess events, as our niche is scholastic chess. However, because the Quebec Open moved to Quebec City this year...and with an August date instead of the traditional July dates, there was a void on the Montreal scene so we felt it would be good to provide a 9-round...one game a day event in the city.

I would appreciate the opinion of folks on the following question: What rating should an organizer use when determining sections and prizes that a player is eligible to play?

This is what we used:

the highest of your permanent rating between FQE, CFC and FIDE.

I looked around at what others were doing (after the fact of course :) ) and this is what I found:

2017 Calgary International:

Based on current FIDE rating (CFC ratings will not be used)

2017 Canadian Open

The following ratings (as of July 1st, 2017) will be used to determine player section eligibility:
For Canadian residents outside Quebec: CFC rating
Residents of Quebec: FQE rating
US Residents: Higher of CFC and USCF ratings
All other foreign players: Highest of CFC, FIDE, or other available rating
For any player in the Master section, FIDE will be used for eligibility if it is higher than other available ratings.
Unrated or provisionally rated players may not win more than C$1000 except in open section.

2017 Canadian Amateur: Eligibility: Players MUST play in the highest section they are eligible for, based on current rating. FIDE or FQE Ratings taken at par

2017 Montreal Open: Rating used (Pairings & Sections) FQE for Quebec residents / highest of FQE or CFC for the Rest of Canada. If you have had a rating in the past that is higher then you can not play in a section that is less than 100 points from your highest rating.


2017 Quebec Open - For the A section, FIDE ratings will be used for players who have them, FQE for Quebec players, CFC for Canadians from other provinces, and National ratings for players from other countries.

2017 Toronto Open - I could not find the criteria...can some one help me here please.


2017 Varennes International ($16,000 Guranteed Prize Fund...I could not find the criteria...can some one help me here please.

2018 Trois-Rivieres Open ($11,300 Guaranteed Prize Fund) Ratings: For Section A, FIDE ratings will be used. if a player does not have a FIDE rating, the highest valid rating between that of the FQÉ and CFC will be used.
For all other sections, the highest valid rating between that of the FQÉ and CFC will be used. If you have had a rating in the past that is higher then you can not play in a section that is less than 100 points from your highest rating.


In hindsight I would have done it differently for our event. Is there an ideal solution? Is there a pôlicy that our National Federation has adopted that I know not? Would it not be nice to have a standard approach to this...or have I been out of the loop so long that I know nothing. (I like the rules for the Canadian Open but should we distinguish between permanant and provisional ratings when determining which section a player can participate?)

Also, I have noticed another rating system (why not...if CMA can have one afterall :) ) the URS System...

URS (Universal Rating System) Ratings
(Updated every month, following the updates on universalrating.com)

The Universal Rating System™ (URS™ for short) is a new sport’s rating system designed to assess the relative strength of participants. It was developed as a result of a collaborative research project funded by the Grand Chess Tour, the Kasparov Chess Foundation and the Chess Club of Saint Louis. Unlike Elo rating, the URS calculates a single overall rating for every chess player which is optimized to represent their strength at classical chess based upon their game results across all time controls (classic, rapid and blitz). The research claims to indicate that incorporating both slow and fast-play results provides better information regarding a player’s overall chess ability, with the enlarged pool of data aiding the overall accuracy of the rating system and improving the ability to estimate the strength of a player at classical time limits. Read more at the official website universalrating.com.

The ratings are updated (as of today) monthly. Chess-db.com is displaying those ratings on the players' profiles and will be offering various statistical information on this and other pages.

Should this also be taken into consideration?

Larry

I would think it would be several years before this system could possibly overtake FIDE as the "Gold" standard.

[Roger Patterson]

Re: Which rating should we use?

I see Paul has already posted the policies that Victoria Chess uses. But some points to consider:

1) Taking the highest of all the ratings is what statisticians call a biased estimator. That is, it produces a number that is expected to be higher than the true strength (if all ratings are on average correct but randomly up or down, taking the highest produces a number that is higher than the average).

2) Many people have ratings in systems that they are not active in or play only once a year. A rating that represents a large number of recent games is more useful than one that represents only a few games in the recent past.

3) Then of course there is the problem that there is an implied comparison of one rating system to another. FQE used to be thought to be lower than CFC, ditto FIDE. USCF - unclear. Not sure what the current consensus on this is but it's difficult to know what the truth is.

So, my approach is to pick your main system (CFC for us) but allow for exceptions for those who are not particulary active in that system.

[Kevin Pacey]

Re: Which rating should we use?

I don't like the idea of a Universal rating system being the standard. IMO Regular, Active and Blitz time control ratings should be seperate as these are rather like being for seperate variants of chess... however it's at least amusing to see what your Universal rating would be (I'd find it relevant only if a player really excelled or sucked at slow AND fast time controls).

On a somewhat related note, I still rather like having a USCF-style rating floor for a given rating system - e.g. it's easier to avoid the eternal problem of sandbagging. The given rating system might become a bit inflated, but in the case of the CFC at least, that may make for more members who are a little more contented than otherwise might be the case.

[Kevin Pacey]

Re: Which rating should we use?

For an example of something like a Universal rating system as it would work for run-of-the mill players, here's such a ratings list for playing all kinds of chess variants on Chess Variant Pages' Game Courier (which is play-by-mail); I'm currently rated 4th on the list of hundreds of players (many not active currently), but so far I've played chess variants which are only very much like standard chess. If I were to branch out and play all sorts of weird variants, I suppose my listed rating would drop like a stone (be aware that the page takes some seconds to fully load, and to present a sidebar to scroll with):

http://play.chessvariants.com/pbm/ratings.php

FYI, The one player other than myself on the list that I know has a FIDE rating is Francis Fahys of France (3rd on the list currently), who apparently has a FIDE rating of only 1400+ (based on FIDE's rating list), though from what I can tell he at the least seems to play a mean game of Chinese Chess, or Shogi (Japanese Chess). A bit farther down the Game Courier rating list, Jose Carrillo is in Ajax Ontario (he quit playing chess regularly long ago, due to a dislike of opening theory). Carlos Cetina is in Mexico and seems to be a 2000+ strength chessplayer, and good at all kinds of variants. Fergus Duniho is the Chess Variant Pages' webmaster, in NY State (closer to Ottawa than his State Capital). Much further down the list, there's Charles Gilman, residing in the United Kingdom, who's invented 200+ chess variants. I've also seen a posted comment by a player years ago who claimed to be 2400+ USCF, though I doubt he's anywhere on the modern day version of the Game Courier rating list. Standard chess is one of the 'variants' that can be played on Game Courier, fwiw. Also, note that its rating list comes with some explanation of how the rating system works.

[Kevin Pacey]

Re: Which rating should we use?

FYI, I've added a paragraph to my last post.

[Larry Bevand]

Re: Which rating should we use?

First I would like to thank those who did respond to this thread. I think it is important. But perhaps I am wrong. Enlighten me please!

Secondly, I am disappointed that many of those in the know could not be bothered...although they use the board all the time to advertise their events (at no cost)...ahh the human beast :).

Sooo...I will tackle the problem in another way. We had a gentleman from India taking part in our under 1800 section. Obviously he had a number of different ratings....FIDE, USCF India URS and perhaps others.

Which rating would you have taken?

Larry

[Garland Best]

Re: Which rating should we use?

Ok, I'll take a stab at this,

1) India - would not use.
2) URS - would not use.
3) CFC, FIDE, USCF - I would look at which system he has significant activity on over the past 2 years. Anything that has 10 or more games played, I would use consider valid. I would go with the highest of the valid ratings.

[Victor Plotkin]

Re: Which rating should we use?

Larry,

If you really need my advice...

1. Use CFC rating as a base.

2. If a certain player from QC doesn't have CFC rating, (or his CFC rating is not reliable) use FQE+50 or FQE+100.

3. If a certain player from USA doesn't have FIDE rating, but has USCF rating - use USCF+100

4. For any foreign player use FIDE + 150 (or FIDE + 100, but I recommend 150). The average gap between FIDE and CFC is around 130 points.

[Mathieu Cloutier]

Re: Which rating should we use?

3. If a certain player from USA doesn't have FIDE rating, but has USCF rating - use USCF+100

You mean USCF - 100?

[Victor Plotkin]

Re: Which rating should we use?

You mean USCF - 100?

I mean exactly CFC = USCF + 100.

[Louis Morin]

Re : Re: Which rating should we use?

I mean exactly CFC = USCF + 100.

Are you sure? Here are my ratings:

FQE = 1839
CFC = 1841
FIDE = 1851
USCF = 2008

[Mathieu Cloutier]

Re: Which rating should we use?

I mean exactly CFC = USCF + 100.

This kinda goes against the common knowledge that USCF ratings are inflated compared to other systems. But maybe I'm not up to date on that issue.

[Hugh Brodie]

Re: Which rating should we use?

I have also always assumed US ratings to be higher. Goichberg tournaments (Continental Chess Association) tournaments traditionally added 100 points to FQE and FIDE ratings (and some number to CFC ratings), but they now have a foreign rating conversion web page at http://www.chesstour.com/foreignratings.htm
FIDE ratings now have a minimum of 50 points added - with more added the lower the rating. FQE ratings have a similar sliding scale. CFC ratings are taken at par.

Maybe the CFC should look at these tables - possibly an item to be brought up at the AGM, or some work for the Rating Auditor?

My ratings:

USCF: 2030
CFC: 1851
FQE: 1859
FIDE: 1909