Bonus GM Norm

[Jack Maguire]

The recently concluded Groningen Chess Festival makes me think that FIDE should perhaps implement a bonus GM norm provision. It strikes me as highly inequitable that 21-year-old Armenian IM Vahe Baghdasaryan (2368) should receive the same GM norm for his 6.5/9 2620 TPR as that of the tournament winner, 16-year-old German IM Alexander Donchenko (2523) and his scintillating 8/9 2803 TPR. Surely a 2800+ TPR is worth a bonus norm. Groningen was arguably a stronger tournament than the World Junior, for which FIDE awards a full GM title for winning, so why not 2 GM norms for such an incredible performance?

[Rene Preotu]

Re: Bonus GM Norm

The recently concluded Groningen Chess Festival makes me think that FIDE should perhaps implement a bonus GM norm provision. It strikes me as highly inequitable that 21-year-old Armenian IM Vahe Baghdasaryan (2368) should receive the same GM norm for his 6.5/9 2620 TPR as that of the tournament winner, 16-year-old German IM Alexander Donchenko (2523) and his scintillating 8/9 2803 TPR. Surely a 2800+ TPR is worth a bonus norm. Groningen was arguably a stronger tournament than the World Junior, for which FIDE awards a full GM title for winning, so why not 2 GM norms for such an incredible performance?

What about his 2274 performance at the U16 Chess Olympiad? Should FIDE take back the "bonus" norm? ;)

http://chess-results.com/tnr153962.a...&wi=821&snr=59

I don't think you can compare a 13 rounds tournament (World Junior) with any 9 rounds tournament.

[Jack Maguire]

Re: Bonus GM Norm

Arthur B. Bisguier, whom I believe is the world's lowest rated GM, says, "Hell, no, Rene!" (:

http://ratings.fide.com/card.phtml?event=2001098

[Jack Maguire]

Re: Bonus GM Norm

And Nikola Spiridonov seconds Arthur B. Bisguier's sentiments (:

http://ratings.fide.com/card.phtml?event=2900238

[Jack Maguire]

Re: Bonus GM Norm

Why not cover all corners of the world to complete our not-so-holy triumvirate of sub-2200 GMs?

http://ratings.fide.com/card.phtml?event=7100167

[Jack Maguire]

Re: Bonus GM Norm

While not an exhaustive search, fossicking FIDE's top country lists, it appears that Alexander Donchenko will become one the world's 3 IMs rated above 2550 after Groningen. The other 2 are the Ukraine's 21-year-old Stanislav Bogdanovich (2581) and fellow German, 17-year-old Matthias Bluebaum (2565).

[Rene Preotu]

Re: Bonus GM Norm

Look at this player (born 2000) and rated 2517
https://ratings.fide.com/card.phtml?event=13408798

568 points in 3 months (from 1949 to 2295 in 2 months and from 2295 to 2517 in one month) because of the new K 40 factor.
https://ratings.fide.com/id.phtml?event=13408798

Unbelievable, but he only beat an IM (2444) and drew a GM (2577) in the Al Ain classic
I'm wondering if FIDE will do something to change this.

[Tom O'Donnell]

Re: Bonus GM Norm

Arthur B. Bisguier, whom I believe is the world's lowest rated GM, says, "Hell, no, Rene!" (:

http://ratings.fide.com/card.phtml?event=2001098

If I'm lucky enough to reach my mid-80s I hope I can still play chess well enough to maintain a rating as high as GM Bisquier's.

[Digeng Du]

Re: Bonus GM Norm

Look at this player (born 2000) and rated 2517
https://ratings.fide.com/card.phtml?event=13408798

568 points in 3 months (from 1949 to 2295 in 2 months and from 2295 to 2517 in one month) because of the new K 40 factor.
https://ratings.fide.com/id.phtml?event=13408798

Unbelievable, but he only beat an IM (2444) and drew a GM (2577) in the Al Ain classic
I'm wondering if FIDE will do something to change this.

Wow. Certainly, the old k=15 did not allow rating adjustments fast enough for rapidly improving juniors, but changing the k factor to 40 always seemed like a bit too extreme a solution...

[Jack Maguire]

Re: Bonus GM Norm

If I'm lucky enough to reach my mid-80s I hope I can still play chess well enough to maintain a rating as high as GM Bisquier's.

Another oldie but goodie, Tom, is IM Edwin Bhend, now 83, and who once beat Tal. Edwin is playing today against the world's top Junior, 18-year-old Richard Rapport, in R1 action at the Basel Chess Festival 2015. Edwin had certainly been holding his own but has made a couple of inaccuracies in a spot of time trouble, with the game still in progress.

http://www.chessbomb.com/arena/2015-...rd-Bhend_Edwin

[Eric Gedajlovic]

Re: Bonus GM Norm

The Curious Case of Parviz Gasimov

It seems clear that Gasimov's current rating is well above his playing strength.

I believe this happened not only because k= 40, but because FIDE uses published (monthly) ratings rather than live ratings to calculate pre-tournament player ratings. In this respect, Gasimov gained 40.80 points from his performance in the Abu Dhabi Chess Festival which was held in August 2014, but was not rated until January 2015. Another contributing factor is that Gasimov played a very large number games (85+!) between August and the January period - all of which probably used an inaccurate pre-tournament FIDE rating (due to the late rating of the Abu Dhabi tournament as well as the typical lag between live and published ratings for under-rated juniors).

Thus, Gasimov's rating appears to be out of line with his actual playing strength because he played so many games in such a compressed time period during which his published rating was much lower than his live rating. As Victor Plotkin wrote about in an earlier post, this can happen when a junior plays in many games in a month before his/her published rating is adjusted, BUT in this case, the number of games played by Gasimov was truly extreme. Further exacerbating matters is the fact that the Abu Dhabi tournament was not reflected in his published ratings until January.

So, it appears that the recipe for this sort of rating overshoot is for a junior under k= 40 to play in tournaments that are not rated in a timely manner and to play in as many games as possible before they are rated. More generally, as noted by Victor Plotkin in his earlier post, anomalous results can happen when a player’s rating is advantaged by the lag between published and live ratings. This was just a very, very extreme example of the general problem identified earlier by Victor.

One obvious loophole that may be closed is to mandate that tournaments which are FIDE rated be done so within a very circumscribed time period. A more encompassing solution to this problem of rating over shoot would be for FIDE to change to rating tournaments on the basis of live rather than published ratings.

This is my understanding of the situation with the curious case of Parviz Gasimov and his rating gain in the past few months. I apologize in advance if I made any inaccuracies in my analysis.

Look at this player (born 2000) and rated 2517
https://ratings.fide.com/card.phtml?event=13408798

568 points in 3 months (from 1949 to 2295 in 2 months and from 2295 to 2517 in one month) because of the new K 40 factor.
https://ratings.fide.com/id.phtml?event=13408798

Unbelievable, but he only beat an IM (2444) and drew a GM (2577) in the Al Ain classic
I'm wondering if FIDE will do something to change this.

[Mathieu Cloutier]

Re: Bonus GM Norm

Using K=40 is a bad idea for the FIDE, especially as they have tournaments rated after some delays. So you can pile up 2 or 3 good performances and your rating will skyrocket.

But then again, k=10 for established players is also a little too conservative.

If I remember correctly, Jeff Sonas once ran some statistical tests and K=24 was a very good compromise for ratings to move fast enough without creating anomalies.

[Roger Patterson]

Re: Bonus GM Norm

......

If I remember correctly, Jeff Sonas once ran some statistical tests and K=24 was a very good compromise for ratings to move fast enough without creating anomalies.

http://en.chessbase.com/post/the-son...tter-than-elo-

But I would bet that he has updated those calculations and that he was consulted on the changes regarding the K factor.

[Rene Preotu]

Re: Bonus GM Norm

This was posted on Facebook by Emil Sutovsky, ACP president



Dear all,

The ACP has received a number of requests to step in and to do our utmost in order to prevent a gigantic rating jumps, like we witnessed in the last Elo list. I have already addressed FIDE on that issue, and we will follow the development of the situation closely, with the clear aim to disallow to the K-factor or to other reasons to ruin the whole rating system.
ACP - with the chess players and for the chess players.

[Jack Maguire]

Re: Bonus GM Norm

While not an exhaustive search, fossicking FIDE's top country lists, it appears that Alexander Donchenko will become one the world's 3 IMs rated above 2550 after Groningen. The other 2 are the Ukraine's 21-year-old Stanislav Bogdanovich (2581) and fellow German, 17-year-old Matthias Bluebaum (2565).

Alexander Donchenko may well be the highest rated IM on the planet after following up Groningen with a joint 1st, 5.5/7 (4 wins and 3 draws), at Basel 2015.