7. Rating Auditor Report (Paul Leblanc)

[Ken Jensen]

...what if it was explained to the kids and parents that the active rating (or rename it junior rating) was for junior only tournament and regular rating was for when they played in adult (mixed) tournaments. That should work, no?
...
Except for the stronger players, I don't think they care which rating it is, CFC CMA, Mississauga, whatever.

No, kids will not buy CFC active ratings. In BC I know there is constant pressure, or competition for event ratings. Our players have a choice between CFC, CMA NWSRS and various house ratings.
The tug of war between CFC and CMA is ongoing. So far CFC has won a large market share on the argument that it is The National Rating. As soon as you go away from regular to active the balance shifts away from CFC.

In BC most active events go unrated or house rated. I can guarantee that any current CFC rated event that you strip of it's regular rating will be lost to the CFC. Is that the plan? It may help the under rated juniors problem.

There are a lot of active events in BC. Some are CMA rated. Some are house rated, many are not rated. NONE are CFC Active rated. That's not entirely true, since 2006 there have been 5 CFC Active rated tournaments in BC. That is why I say that if your solution to the high performing juniors "problem" is to deprive them of a rating until they are adults the CFC will not have any juniors.

That said I believe you are on the right track by looking at the ratings calculations applied to juniors and adjusting as necessary so Junior Ratings and related skill match the equivalent adult rating.

The old rules that applied regular ratings to junior games as short as 30 minutes were made on the realization that for most juniors that is a "long" game. The strategy, like the USCF's was to get juniors inetgrated into the system early and develop them into adults.

Getting rid of the 30 minute rule was the first shot at the under rated junior "problem". It is along the same thought lines of "deprive them of a rating until they play only adults". Unfortunately it may have had the opposite effect. With fewer rated games the kid's ratings are even slower to adjust, and more likely to be out of date.

We will never be able to compensate completely for the effect of intensive study and progress at the chess board which young kids can demonstrate. No more than we can stop them from growing three sizes in a year.

Ken Jensen

[Paul Leblanc]

We looked at all players with 25 to 50 rated games from 1 April 2007 to the present. There were 9,100 data points. We were looking to see if an outstanding performance in one event was an indicator that we had a fast-rising player on our hands worthy of bonus points.
We set the criteria at a 50 point gain in a single event as well as the attainment of a personal high rating and followed those players to their 75th rated game. It turned out that 83% of them did indeed continue to a higher rating and the average gain was a further 71 points.
The control group, all players with between 25 and 50 rated games, regardless of whether they had an outstanding single event, also gained further rating points. 71% of them went upwards and the average gain was 50 points.
This is against a background where the average rating of all players in the CFC remained static.
So, the tentative observation is that this group is under-rated. I have much more work to do on this before translating it into a recommendation to change the bonus point system.

[Ken Jensen]

It sounds to me like you have the target in sight.
You know the numbers.

Extrapolating backwards through your data set you can determine with confidence that these players were 100 points low, or what ever number the data supports. Then through the magic of hindsight you can fiddle with the math to find a formula that produces a more appropriate rating. Be it bonus points, provisional ratings or new math, this solution will not only retain juniors, it will encourage and excite them.

Ken Jensen

[Bob Gillanders]

Is it a case that the ratings calculator is a zero net sum operation? In other words does every junior rating point have to come at the expense of an existing adult player?

Yes, that is the basic problem.
Ratings are a zero sum game. The winner gains rating points, and the loser loses an equal number of rating points.
This causes a natural rating deflation over time. Most players gain rating points over the years, then eventually stop playing, removing a larger number of rating points from the pool than when they entered. Bonus points and participation points are used to combat this deflation.

[Fred McKim]

Ken: The performance ratings shown in the Crosstable listings do not include the "protection" for having played opponents > 400 points away. So this performance rating is only a raw estimate. Players winning all of their games against players averaged 500 points lower will have a performance rating 100 points below their pre-rating, but will still gain a handful of points.

[Fred McKim]

I have performed all sorts of informal surveys and studies over the years. Generally speaking juniors who play frequently are generally 100 points underrated, while those who are among the top 10-20% are probably 200 points underrated.

Obviously when junior plays junior, we are only going to see relative changes in the their rating, it's when they get playing the more static adult population we see these numbers coming into play and corrections of some sort should be made.

[Ken Jensen]

I agree with your observations Fred. The problem of pegging a junior's skill level with a rating is challenged by the the moving target aspects. It's hard enough to get an accurate rating of a static player. As you describe active juniors typically change 100 points a year or more. If they do not play enough rated games in that time there will be major jumps in rating when they do. We hold regular adult/junior mixed events to help that transition and make sure the juniors get lots of rated games.


For what it's worth I have seen the impatient juniors waiting for their rating to catch up. They have solved a problem in their game and are playing 200 points above their rating, but the rating only moves 50 points a trournament. So it takes them 3 or 4 tournaments to get adjusted. The problem of course comes in when they improve further before they have found 4 CFC rated events to play in.

Ken Jensen

[Halldor P. Palsson]

We never lose an opportunity to drive folks away from the CFC. The CFC Active rating system has been a farce for a long time that nobody takes seriously.

The 30 minute all junior regular rating is a great way to get juniors into regular organized chess. It is also fast enough that a chess school can have some instruction and a tournament game in one session for their students.

Faster games for learn to play juniors is a great idea. My introduction into chess 45 years ago at the junior leval was 7 min a side, next level to 15 min and then 45 min. The main objective is to get kids to have fun playing lots of chess and learning the game. After winning events at the 45 min level or getting 75% in two we got to play with adults at regular controls entering at the lowest category.

What is missing in the junior rating discussion is that we have to put a floor of 800 or 1000 on the CFC system. Some European national systems are run this way - no ratings lower than 1000. That way the entrants come in over rated by 300-400 points and do not deflate the system as is the case now because they start from 600 and exit 1400-1600.

[Pierre Dénommée]

I have moved a motion to abolish the Active rating fee and to create a new category of membership allowed to play only Rapid Games and Regular tournaments specifically allowed by the Executive. The motion has been defeated despite the fact that it has been done elsewhere. I still see the Active rating as a promotional tool that is to be used to attract players to Regular games.

We never lose an opportunity to drive folks away from the CFC. The CFC Active rating system has been a farce for a long time that nobody takes seriously.

[Vladimir Drkulec]

We never lose an opportunity to drive folks away from the CFC. The CFC Active rating system has been a farce for a long time that nobody takes seriously.

Nobody takes it seriously and that may be a good thing. For one thing I am more likely to play in an active tournament where I might have to play people six hundred rating points below me because I don't take my active rating seriously. I will only do that locally in regular chess and even then I won't do it often.

The 30 minute all junior regular rating is a great way to get juniors into regular organized chess. It is also fast enough that a chess school can have some instruction and a tournament game in one session for their students.

So it is okay to corrupt our regular rating system for the sake of convenience because some juniors who do not pay a membership fee and pay a significantly reduced rating fee do not care about active ratings?

When John Coleman broached the idea of rating the games played by the kids in our Friday night classes they overwhelmingly voted against it even though pretty much all of the kids are members of the CFC. The Friday night classes are very healthy and we are currently operating at room capacity with others trying to join the class. We are getting decent turnouts from the kids at the available tournaments especially the one day events.

I am very active in local children's chess as a coach and mentor and while I do see these kids as the future it is wrong to corrupt our rating system to pander to the sensibilities of certain parents. The kids just want to play and have fun. They really aren't as fixated on ratings as adults are.

If the kids' parents want the kids to play for regular ratings, insisting that they play according to the rules and play games of at least one hour sudden death time control hardly seems onerous to me.