Unbelievable rating climb

[cizio]

We talk about caste-system, but what you think about the rating graph of no.3 in female rating list? (from 1988 to 2521 in 1 year)

http://www.iccf-webchess.com/PlayerD....aspx?id=84866

The player, that for sure is a strong player and that probably deserve her rating (2603 at the moment), is still gaining elo for some draw of a 1988 vs a 2300, if she has many games left started in september 2008 probably can reach 2700 without any finished games as a 2600.

Maybe the right way to become a GM is to come back to 1500 and restart?
Maybe there is no caste-system ?
Maybe there was some bug in ICCF rating that now is fixed?
(I want to say again here that I think Gerhard Binder made a wonderful job.)

cheers
Mauro

[capajmc]

Originally Posted by cizio

We talk about caste-system, but what you think about the rating graph of no.3 in female rating list? (from 1988 to 2521 in 1 year)

http://www.iccf-webchess.com/PlayerD....aspx?id=84866

The player, that for sure is a strong player and that probably deserve her rating (2603 at the moment), is still gaining elo for some draw of a 1988 vs a 2300, if she has many games left started in september 2008 probably will reach 2700.

Maybe the right way to become a GM is to come back to 1500 and restart?
Maybe there is no caste-system ?
Maybe there was some bug in ICCF rating that now is fixed?
(I want to say again here that I think Gerhard Binder made a wonderful job.)

cheers
Mauro

Hello Mauro, it seems nothing has change about this problem ( remember the brazilian players like Carlsen and Piccoli ) .... I found a bigger rating climb: http://www.iccf-webchess.com/PlayerD...aspx?id=429154 from 2211 to 2621 in less than a year. This player has more than 140 games simultaneously right now, Yours Sincerely

[cizio]

Originally Posted by capajmc

Hello Mauro, it seems nothing has change about this problem ( remember the brazilian players like Carlsen and Piccoli ) .... I found a bigger rating climb: http://www.iccf-webchess.com/PlayerD...aspx?id=429154 from 2211 to 2621 in less than a year. This player has more than 140 games simultaneously right now, Yours Sincerely

Thanks, your player has a nice downhill also.
How do you check how many games is someone playing?

[capajmc]

Originally Posted by cizio

Thanks, your player has a nice downhill also.
How do you check how many games is someone playing?

With Eloquery, and the f6 function...

[Ron Langeveld]

Originally Posted by capajmc

Hello Mauro, it seems nothing has change about this problem ( remember the brazilian players like Carlsen and Piccoli ) .... I found a bigger rating climb: http://www.iccf-webchess.com/PlayerD...aspx?id=429154 from 2211 to 2621 in less than a year. This player has more than 140 games simultaneously right now, Yours Sincerely

Something has changed. Situations like these are monitored by the Rating commisioner and appropriate measures are taken. This was also reported during congress in September 2009. Players like the one you linked to are clearly trying to sabotage the system and the system reacts, maybe a bit slowly but it does. Adjustments to the rating formula have been made and the RC also has the right to change start ratings when there is due cause.

[Pedro Hegoburu]

Hi Ron,

I am quite surprised to read these players "are trying to sabotage the system" (dixit). They are just playing CC games and their results are being computed by a formula they haven't laid hands on - I don't see any "sabotage" here...

Regards,

Fox Mulder
(I mean, Pedro)

[Uri Blass]

Pedro,
I totally agree with you.

Tritt Maciej could get easily higher rating in case of trying to do it.

1)It seemed that he did not use the DMD strategy by not resigning lost positions

see for example the following game:
http://www.iccf-webchess.com/MakeAMove.aspx?id=222700


2)He accepted draws when draws were not good for his rating and he could simply continue the game in order not to get the draw counted for rating in the existing lists.

see

http://www.iccf-webchess.com/MakeAMove.aspx?id=222683

I believe that he could get rating above 2800 in case of trying to maximize his rating.

Uri

[cizio]

Anyway,
It is weird to earn a 2600 rating (or to go down to 1800) with a 2150 or so performance in the most of your tourneys.

[Gary Ruben]

Originally Posted by cizio

Anyway,
It is weird to earn a 2600 rating (or to go down to 1800) with a 2150 or so performance in the most of your tourneys.

I once gained more than 100 rating points in one period. I recall at the time that was unusual. Computers weren't very good, or at least not as good as today. After I mentioned my accomplishmnent on TCCMB my real name was quickly guessed.

The rating used for the entire event is the one which is entered at the beginning of the event. If a player has a low rating, like 1800, the trick is to start 50 games at one time which means several events. For all 50 games the rating point gain or loss is calculated at the start rating of the event. Even if your rating has risen to 2000 the system uses 1800 for the calculation, or at least it diid. As games end, start new events before the rating gets too high.

Possibly with the strength of the computers these days the solution to the problem, if it is a problem, is to start players with a rating of around 2250.

[Uri Blass]

Starting with high rating is not going to help in the case that a player lose on purpose many games to get low rating first.

I think that if we want to prevent people to get high rating that they do not deserve then we need to do the following steps:

1)Consider tournaments at the time that the player finished all his games and not earlier(it means that players cannot get high rating by using the DMD tactics and have 40 wins,no draws and no losses out of 200 games that they started).

2)Calculate the performance of players and decide that the new rating of a player who improved his rating cannot be higher than his performance and if the calculation make it higher than the performance then reduce the rating
to the performance.

Uri