chess.com and lichess players are not unrated

[Erik Malmsten]

During the past two years many Canadians have been playing (blitz) online and have got a rating on chess.com and lichess. They are experienced adult players who have learned openings from youtube videos and, I argue, should not be treated as beginners in CFC tournaments. In the past, when we got a European entrant we accepted their rating estimate and let them play in the top section. I know a "2000" and "1300" player who want to play CFC chess and they're told that they have to play in under 1000 section, that they can't play up. That's ridiculous. The Annex Chess Club has an Internet player in the over 1800 section and he is playing well, he deserves to be in that section. And we have to accept that CFC ratings are largely 2 years out of date, so online ratings should be looked at if they want to play up a section.

[Bob Gillanders]

Good point Erik. Thanks, you bring up an important issue.

Of course, this issue is not new. We have always had cases were unrated players who maybe very strong, what section do they play in?

So, some years ago now, we decided to initial a new players rating based on their FIDE, FQE, or USCF ratings, upon approval.
It has worked well and helped get new members to the appropriate section quickly. But it didn't cover everyone, no system ever can.

Now we have a situation, where large numbers of people have ratings "out of date" due to the pandemic. There is no easy solution to "correct" everyone rating to where it should be!

But let me point out that TD's/ Organizers have always had the discretion to determine the eligibility of who plays in what section. If TD's / Organizers feel that an online rating better represents a players strength, then they can use them. I can see some TD's following CFC ratings closely, and some getting very creative. Not everyone is going to be happy with every decision. But please everyone be sympathetic to the challenge posed.

[Erik Malmsten]

Good point Erik. Thanks, you bring up an important issue.

Of course, this issue is not new. We have always had cases were unrated players who maybe very strong, what section do they play in?

So, some years ago now, we decided to initial a new players rating based on their FIDE, FQE, or USCF ratings, upon approval.
It has worked well and helped get new members to the appropriate section quickly. But it didn't cover everyone, no system ever can.

Now we have a situation, where large numbers of people have ratings "out of date" due to the pandemic. There is no easy solution to "correct" everyone rating to where it should be!

But let me point out that TD's/ Organizers have always had the discretion to determine the eligibility of who plays in what section. If TD's / Organizers feel that an online rating better represents a players strength, then they can use them. I can see some TD's following CFC ratings closely, and some getting very creative. Not everyone is going to be happy with every decision. But please everyone be sympathetic to the challenge posed.

So i hope you can talk with the Hamilton organizers about allowing a player who has a plus score against me to not have to play players rated only 500. Not even in the under 1600 section where most players are around 1000.

[Bob Gillanders]

So i hope you can talk with the Hamilton organizers about allowing a player who has a plus score against me to not have to play players rated only 500. Not even in the under 1600 section where most players are around 1000.

Will do. I expect to see Ken at the Burlington club tonight.
But I make no promises. The Hamilton Club Executive is driving the bus. I am just the TD.

[Erik Malmsten]

Will do. I expect to see Ken at the Burlington club tonight.
But I make no promises. The Hamilton Club Executive is driving the bus. I am just the TD.

Thanks.

Watching Eric Hansen (in Toronto) in the Charity Cup (Unicef). Pretty tense for the players on the edge of qualification.

[Larry Bevand]

Good point Erik. Thanks, you bring up an important issue.

Of course, this issue is not new. We have always had cases were unrated players who maybe very strong, what section do they play in?

So, some years ago now, we decided to initial a new players rating based on their FIDE, FQE, or USCF ratings, upon approval.
It has worked well and helped get new members to the appropriate section quickly. But it didn't cover everyone, no system ever can.

Hi Bob,

If a youngster has a CMA (Chess'n Math Association) rating... the CFC does not recognize it (yes more kids in Canada have CMA ratings than FIDE CFC and FQE combined)....

I have finally convinced the FQE to use CMA ratings for pairing purposes when a youngster does not have an FQE rating... in the events that they run.

For some reason, they do not publish this policy on their site...

Of course I understand that it is politics.

But honestly, if a player has been playing under another system...and has a rating..why would you simply pair them as unrated...Goichberg (Continental Chess) adds points to FQE ratings when Quebec players take part in his events.. he uses whatever info he has to put the person in the right section. So if the CFC thinks CMA ratings are too low...you can do what Goichberg does with FQE ratings...no?

Larry

[Erik Malmsten]

Hi Bob,

If a youngster has a CMA (Chess'n Math Association) rating... the CFC does not recognize it (yes more kids in Canada have CMA ratings than FIDE CFC and FQE combined)....

I have finally convinced the FQE to use CMA ratings for pairing purposes when a youngster does not have an FQE rating... in the events that they run.

For some reason, they do not publish this policy on their site...

Of course I understand that it is politics.

But honestly, if a player has been playing under another system...and has a rating..why would you simply pair them as unrated...Goichberg (Continental Chess) adds points to FQE ratings when Quebec players take part in his events.. he uses whatever info he has to put the person in the right section. So if the CFC thinks CMA ratings are too low...you can do what Goichberg does with FQE ratings...no?

Larry

I would expect that the majority of the 40 entrants below 1600 have a cma rating. Is there a formula for converting to CFC equivalent? Has there been many online cma tournaments during Covid?

But, they are also playing on chess.com and lichess. Should those ratings also be looked at?

[Larry Bevand]

Hi Eric,

The last online event organized by CMA attracted 288 youngsters (Feb 27) : https://echecs.org/cotes/appariements/14831


Larry

[Frank Dixon]

Great discussion, and necessary.

I met a player new to Kingston, last night, at the Kingston Chess Club, where I played my first OTB chess in two years. He is from overseas, a graduate student at the university, has a FIDE rating about 2000 from many years ago, based on only a few games, has only a few CFC games, from a CFC tournament which is not yet rated (2022 University Teams), and played very well to defeat me (a rusty CFC 1864, but I have been as high as CFC 2090). As an organizer, I have the right to estimate his rating, based on all available information (as I have done in the past in several cases), and I think that a fair rating for him is 2000. But, if he wishes to play in the big Kingston event coming up, will he receive the same consideration there? That is a fairly easy case.

From the standpoint of a professional scientist / mathematician, I state with firm assurance that CMA ratings are much lower than CFC / FQE ratings for the same players, on a comparison basis, because the CMA system is not set up to include general population rated players. It deals with only less experienced youth players. It is inherently deflationary at source, since its operational method DOES NOT EQUATE with what professor Arpad Elo used when he set up the FIDE rating system, and before that the USCF rating system. To be accurate, a rating system needs to be following the so-called 'Normal Distribution', or Bell Curve, from a mathematical statistical standpoint, of the players in its system. Everything flows from that. The larger the population, the more accurate it will be. Rating calculations for events use the Normal Distribution for changes; this applies to CFC, FQE, CMA, USCF, and FIDE. In other words, CMA methods over its population do not fit with the assumptions inherent in Elo's formulations. Now, it is true that there are some Master-rated players in CMA, but they DID NOT EARN those ratings by playing exclusively in CMA events; rather, their CMA ratings were simply brought over from the CFC system.

So, Larry, don't hate me for stating this, since it simply the truth. Any mathematician familiar with chess rating systems will say the same thing.

FQE ratings are lower, compared to CFC, because the FQE did not adjust them in 1980, as the CFC did, for deflation. That was about a 100 point boost, then. Also, the FQE population is smaller than the CFC's, so system-based assumptions are less valid. Not to say that they are NOT valid; I would not go that far.

[Erik Malmsten]

Found this article on chess.com that states that a chess.com D-class blitz rating of 1300 is equal to 1420 FIDE. A B-class 1700 blitz rating is about the same as a fide rating of 1690, but a master on chess.com 2200 is only rated as an expert fide of 2105.

And comparing chess.com blitz to lichess blitz rating. 1300 chess.com equals 1630 lichess. 1700 chess.com equals 1930 lichess. But 2200 chess.com equals 2265 lichess.

so 1345 lichess is equal to 895 uscf. 1720 lichess to 1470 fide. 2210 lichess to 2015 fide.

https://chessgoals.com/rating-comparison/

[John Brown]

One solution is if you have a CMA rating under 1000 then you'll get a pairing rating of 500 -700 points higher for the CFC or FQE tournament you enter for the first time. If you do well you'll keep the 500-700 points and that will be your CFC / FQE Provisional Rating for your next tournament. If you do poorly you'll get a rating of 350 added to your CMA rating for you next rated tournament.
Both players will still keep getting point increases until their CMA rating is above 1000/

[Roger Patterson]

Found this article on chess.com that states that a chess.com D-class blitz rating of 1300 is equal to 1420 FIDE. A B-class 1700 blitz rating is about the same as a fide rating of 1690, but a master on chess.com 2200 is only rated as an expert fide of 2105.

And comparing chess.com blitz to lichess blitz rating. 1300 chess.com equals 1630 lichess. 1700 chess.com equals 1930 lichess. But 2200 chess.com equals 2265 lichess.

so 1345 lichess is equal to 895 uscf. 1720 lichess to 1470 fide. 2210 lichess to 2015 fide.

https://chessgoals.com/rating-comparison/

I use that site. Some comments:

1) the comparison is for people who have ratings in both an OTB system and a an online system. So, it systematically excludes those who are only rated online. (the very people you are interested in). My belief in that having never played OTB means you should subtract another 100-150 points off what ever comparison you arrive at from those numbers.

2) Note the huge standard deviation. For the Jack Taylor Memorial this year, I estimated rating for everyone for pairing and prize purposes (as I did not have an unrated prize). Post event, for those I estimated ratings for, the average was within 50 points of the final CFC rating so not too bad but the standard deviation was something like 400 points so not terribly reliable.

3) I normally take CMA at par for pairing / prize / section purposes. But all 3 digit rating are kind of bogus. It takes so little to improve that any such rating is instantly out of date for any kid who is making any effort.

4) The general point that these adults with 1500 or whatever chess.com ratings are not beginners and do not belong in a U1000 or similar section is totally valid. U1600 or similar is totally fine, U2000 is pushing it, even with high chess.com or lichess ratings and Premier/Open is not advised.

[Bob Gillanders]

Will do. I expect to see Ken at the Burlington club tonight.
But I make no promises. The Hamilton Club Executive is driving the bus. I am just the TD.

We did discuss the online ratings question amongst a number of weekend tournament issues challenging the new team.
Online ratings will not be considered for this weekend. Sorry about that. The preference is to keep it simple and not change the rules this late in the day.
We did agree to address the issue well in advance of the next weekender.

[Frank Dixon]

To clarify:
CMA ratings, and its system, are OK for the purposes they are being used for, WITHIN CMA. It is where CMA ratings are compared with ratings from other systems that the problems arise.

[Alex Starks]

CAM rating = CFC rating only for players over 2200

Conversion formula is here

https://chess-math.org/cma-rating-conversion