Author: Uri Blass
Date: 09:00:57 07/24/04
Go up one level in this thread
On July 24, 2004 at 11:39:44, Albert Silver wrote: >On July 23, 2004 at 21:44:41, Nick Rowe wrote: > >> >>> >>>-------------------------------------------------- >>>Testing results on intermediate stages of training >>> >>>CT-ART 3.0 Start ELO Resulting ELO >>>Yevelev V. 2220 2433 >>>Kurenkov N. 2210 2401 >>>Gabrielian A. 2330 2447 >>>-------------------------------------------------- >>I looked at these peoples ratings history and it mostly seemed that they were >>already at this strength practically when she claims to have started training >>them. i'm skeptical > >I don't have access to the earlier rating lists, but what I did see was fairly >consistent with what she reported, so I suspect you're misreading. Ex: > >"This system was tested for the first time on Vladimir Evelev (born in 1983). I >became acquainted with him in the beginning of 1998. He had an ELO rating of >2220 then (AS - I don't have the 1998 rating list to check), and I commented to >him during our first meeting that he had much greater potential." > >(...) > >"The dynamics of his changing ELO rating during two and half years was as >following: 2220-2280-2327-2346-2352-2452." > >And sure enough FIDE has him listed at 2352 in Jan 2000 and 2452 in July 2000 >(http://www.fide.com/ratings/id.phtml?event=4126017&moder=4) > >Artur Gabrielian is reported at 2330 at the start, and as much as 2482. The >numbers aren't 100% in accordance with the FIDE rating list, but are close >enough. > >"Similar tasks were put to Arthur Gabrielian (born 1982, now an IM). Taking into >account his age, personal characteristics and features of temperament, we may >say that the speed of his growth and the intensity of his studies were somewhat >raised. >During two years of training with chess software Arthur's rating shot up to >2482! Training become more interesting for the chess players who entered the >club later, it also became more sophisticated as Convekta Ltd started flooding >the market with new software. By now the number of chess programs to pick from >grew to around 20." > >Gabrielian's rating was around 2330 twice on the FIDE list, and Mikhailova says >her report covers results obtained between 1997 and 2004 so it's hard to >determine exactly when. If we take the furthest in January 2000 at 2327 we can >see that about 2.5 years later, much like Vladimir Evelev, the results are >impressive indeed, peaking at 2476 in July 2002. >(http://www.fide.com/ratings/id.phtml?event=4131002&moder=4) > >One can bicker about the 3-6 Elo points difference, but it still corroborates >the results reported. > > Albert I think that the main problem is that we do not know if they could not get similiar progress without the special training that they got. The question is if a player with stable rating of near 2200 in the last 3 years can earn 200 elo in 2 years. If they want to test the system they need to use it not for somebody that they believe has a great potential but for 2200 players that failed to improve in the last 3 years and are not considered to be players with a great potential relative to other players. Uri
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.