The end of 2002 World Championship tournament Elo ratings for 284 players based on a total of 4718 tournament games played since 1999 can be found below the following notes.
Report:
Matt Peek TGM (NZL) retained his World Championship and still has a 60+ point lead at the top of the established list. Péter Petrecz (HUN) looked like finishing the year in 2nd place but lost out to Shaun Cooper (NZL) by just one point after a disastrous Hungarian Masters in December. Péter did do enough in the WTC to finally qualify for an ITM title though.
Pan-Am Champion Brad Swanlund (USA) had a slightly disappointing WTC but it was not nearly as bad as his disastrous WTC in 2001, so he gained 14 points and 5 places to move up to 7th, the biggest rise in the top 10. European Champion Steven Trezise (GBR) had the biggest fall in the top 10, dropping 19 points, and it would have been a lot more had his opponents not gone on to do well against other strong players too.
Other big gainers in the top 20 were WTC runner-up Dave Dyer (USA), up 52 points and 14 places to 13th and WTC QF-ist & Hungarian Masters winner János Püspöki, up 104 points and 23 places to 20th.
Some other players made huge strides in the provisional list and are likely to be high in the seedings by the time of the next WTC. Others caused big surprises in the WTC and looked like they might make big gains but lost their chance when they suffered big losses or failed to put away weaker players convincingly later in the tournament.
Plate winner Kathy Upton (AUS) had looked likely to fall in the table after disastrous losses in WTC R1 and the Plate PR left her with a WTC-only rating of about 1700. A 'lucky loser' place in Plate R1 gave her a chance to recoup those losses though, and by going on to win the Plate with an average Elo of well over 1900 for Plate R2 onwards, she more than did that, turning the potentially big points loss into a 19 point increase overall. Plate runner-up Zoltán Németh (HUN) and other players who did well in the Plate also saw their ratings increase significantly, and although some of them are still in the provisional list, they too seem likely to figure prominently in the seeding list by the time of the next WTC if their good form continues.
Established and Provisional lists:
Under a classical Elo rating system like this, you do not start with an arbitrary rating like 500 and slowly move away from it. Instead, the system calculates a best estimate of your rating based on all the available evidence, i.e. all the tournament games you have played, even if you have not played very many. This makes the system fairer on your future opponents and gives a better idea of your true strength quickly. However, 'best estimate' does not mean necessarily mean 'likely to be maintained'. Ratings based on a low number of games have a large confidence interval around them and this is taken into account when converting Elo ratings into seeding points.
For example, if you played your first tournament game against Matt Peek and beat him by a big margin, your rating for that game might be 2500. That is the level appropriate to someone who beats Matt Peek by a big margin in every game, and you have beaten him by a big margin in every (one!) game so far, but that is clearly not sustainable over 35 more games - compare some of the extreme WTTC ratings for established players with their overall ratings for more proof of this.
So, there are three rating lists below - "established", "provisional" and "very provisional" - these terms are explained at the top of each list. The criteria for appearing in the "established" list are continuing to gradually become more stringent as the average number of tournaments games played per player increases, as flagged on this page previously.
For full details of the criteria for being "estabkished" and the new T' column, and to best compare those on the official and unofficial rating lists, please look at the Tournament Seeding List. This starts off with the Elo ratings then deducts points to allow for the greater confidence interval around non-established ratings.
Reading the lists:
To find yourself in the lists quickly (if using Internet Explorer), do Ctrl-F, type your lobby nickname and then click Find Next, then page up to the green text above the part of the list you are in to see how reliable your rating is. Please note that Elo ratings depend on who you play and how heavily you win or lose against them rather than directly on your final position in each tournament, so sometimes the order of ratings for a tournament differs slightly from the final positions, eg. someone who wins their qualification group by a long way and beats a few top players but does not win the tournament might get a higher rating than someone who wins the tournament after just scraping through qualification or having an easy draw. For more info on the ratings or to check that they have not been 'fixed', please see the links at the end of this page or email me.
TANTRIX TOURNAMENT ELO RATINGS as at 31 Dec
2002
(the rating lists can be found below these
links)
2002 SINGLE TOURNAMENT FIGURES
Click here for the current Tournament Seeding
list
Single tournament ratings (and numbers of
games) are shown below
Tournament Master (TGM & ITM) norms
awarded so far
together with the contribution to the overall
rating of performances
