The battle for the right to challenge Anand for the world championship later this year will run from March 14 until April 1. It is held in London and is a double round robin with a normal scoring system. It has an average rating of 2786.

The entrants: (the ratings are from the March rating list.)

1. Magnus Carlsen 2872
2. Vladimir Kramnik 2810
3. Levon Aronian 2809
4. Teimour Radjabov 2793
5. Alexander Grischuk 2764
6. Vasily Ivanchuk 2757
7. Peter Svidler 2747
8. Boris Gelfand 2740

official site http://www.worldchess.com/candidates/index.html

live streaming http://london2013.fide.com/en/compon...137-live-video

some more information http://www.chessgames.com/perl/chess...3&crosstable=1

This is so awesome. Double round robin = lots of games for us to enjoy. Can't wait for this thing to start.

Here are the current odds.

09:00 World Chess Candidates Tournament 2013
Winner
Win
Carlsen, Magnus 1.65
Kramnik, V 5.50
Aronian, L 6.00
Ivanchuk, V 11.00
Grischuk, A 12.50
Radjabov, T 20.00
Svidler, P 25.00
Gelfand, B 50.00

I think Aronian might be a good bet at 6:1. Svidler at 25:1 and Radjabov at 20:1 might be good value as well.

is it standard scoring or does it have the 3 points for win, 1 point for draw scoring system from bilbao? if its standard i think carlsen is close to a lock. 3 points for a win scoring could get more randomness in the results.

99% sure they are using standard scoring.

Carlsen is definitely favourite, but not close to a lock.

Here are the pairings, Carlsen looks to have a good draw with a lot of bottom half opponents in the final rounds : http://www.worldchess.com/candidates...3_Pairings.pdf

im curious about how the ordering of games has an effect on the expectation of a players score. Basically what I want to know is, what is better: being fresh for your tough games early and having easier games later on when youre drained, or having easy games early and bringing the confidence of having done well early on to the tougher later games. I really have no clue what its like to play a long tournament like this and just how severe issues of fatigue are and if there is any tilt that happens from bad results or the positive side of having great results.

anyone have experiences from long live chess tourneys that can expand on this at all? im sure things like tilt are more prevalent in lower rated tourneys but still very curious about the mental side of things

I played a 10 player round robin some 9 years ago and got a crappy start, played pretty bad throughout the tournament. Several 9 round swisses as well, in those that went well I felt pumped and confident from the first rounds. Would prefer the easy start to build up flow, not sure if it makes much difference when the whole field is a murderers' row.

Also, anyone want to run some simulations? Would be very interesting, might do some myself once I get some more free time starting next week. More out of curiosity than for betting purposes

One year in the Major Open at the British Champs I started with 3/4 but only ended up on 6/11 thanks to a last round win. In a later year I had 3/6 but ended up with 7/11 despite being held when I should probably have won in the last round.

Probably more to do with strength of opposition than being fresh or not.

you mean creating a double round robin tourney with computers that have their ratings (giving magnus a 2900 since hes been steadily rising still) and seeing the results? or something else?

obviously takes the human aspects of tilt, heaters and fatigue out of play but would be pretty cool at least for betting purposes. to add in a human component could possibly substitute better or worse bots in for guys who have been winning or losing. i.e. if someone starts 0 wins 2 draws 3 losses could drop their rating by 50 points since they probably expect to play worse afterwards.

Mostly thought of a Monte Carlo based on ELO expected scores. There's still the problem of how to assign the chance for a draw (if you want draws in your model). A bot tournament would be fun as well though, definitely more scientific than the octopus predicting football wch scores

i have a degree in comp sci that i use about once every 3 years and ive got nothing going on today so maybe ill write a few different programs trying to simulate the results.

how do i calculate % to win/lose/draw based on ratings? googled a bit for formulas but i keep finding the reverse: how to calculate new rating based on results

Grischuk at 12.50 while Radjabov is at 20.00 is pretty outrageous if you ask me..

they realize this is not blitz/rapid right?

I would be interested in running some sims. I would never consider doing so without including draws. Can anyone dig up stats on what percentage of serious games each player has drawn, over, say, the past two years? "Serious games" can mean whatever you think it should mean, but obviously not blitz, and probably nothing against anyone rated below 2600. White/black splits too, please, if someone consistently plays for the draw (and achieves it) as black, and then plays much more aggressively as white, I would want to account for that.

From color-specific draw rates for each player, and ELO ratings, I can put together a formula to estimate the odds of win/lose/draw for any pairing. From there it's just monte carlo time, although I may add a function to the sim that calculates an "effective" ELO rating for later rounds, based on some combination of actual rating and results so far, to model certain players being in good or bad form. If I put any sort of form/momentum factor in, I'll be sure to run the sim both with and without it, to satisfy both those who do believe in momentum, and also those who think each game is a truly discrete event.

Will the tournament have a live stream?

Think the best odds here are actually Carlsen at 1.65:1

cant get c++ to work on my computer. gonna give up. but heres the template i was going to use for my idea. hope this is understandable, might be a bit confusing

i was going to give everyone their given rating except carlsen who i was going to give 2900 since its still rising somewhat steadily and i assume it will asymptotically approach 2900 given his performance ratings over the last few years.

i was going to give the white player a 150 point rating boost each game and use this chart in article 1.48a to figure out the p given rating differences: http://www.fide.com/component/handbo...4&view=article

p is just % of the expected win. i wrote my own formula for win/lose/draw from that.

for p = .5, i assumed 15% player 1 wins, 70% draw, 15% player 2 wins.
p = .51, 16% player 1 wins, 70% draw, 14% player 2 wins.
p = .52, 17% player 1 wins, 70% draw, 13% player 2 wins
...
p = .65, i assumed 30% player 1 wins, 70% draw, 0% player 2 wins.

just gave a linear progression for those p values.

for p = .66, 32% player 1 wins, 68% draw, 0% player 2 wins.
p=.67, 34% p1 wins, 66% draw
...
p=.86, 72% p1 wins, 28% draw


so for instance, carlsen vs gelfand where carlsen is white. i give carlsen a 150 point rating boost to 3050. the rating difference is now 310. look up on the chart that means p=.86 which means carlsen is 72% to win, 28% to draw. whereas if gelfand was white, he gets a 150 rating boost to 2890 for a rating difference of 10. thus carlsen's p is .51 which means hes 16% to win, 70% to draw, 14% to lose. also seems reasonable

so basically i was going to enter a lookup table for p based on rating difference, then figure out how often they are going to win/lose/draw based on that p, randomize it, adjust those players ratings after each game using a simple ELO adjustment formula and then run this for every matchup between every player probably 10k times and see how often everyone wins.

thoughts?

The basic approach seems right, but I'd quibble with some specifics. For one, I don't think white is worth 150 points. That would suggest (from the table you linked) that white scores 87% across all games. I'm pretty sure that's not the case. I've always heard that white is closer to 50 points, which would be about 57%, and might still be too high.

Also, the 70% draw rate sounds less like something based in stats, and more like something you guessed at. The math side of your post seemed right at a glance, but the starting assumptions are a bit off. I'd want to find a representative sample of games (all games between players rated 2700+, over the past two years? Or if that sample is too small, make it 2650+ or expand the timeframe?) and use that sample to empirically determine the expected draw rate and white advantage for this tourney. Also I don't like the idea of boosting Carlsen's rating based on a projection that he's still improving, unless we make a valid effort to project the rating trajectory of everyone else in the field as well, and edit ALL of their ratings based on those projections. Doing it for only Carlsen is unfair. Do you not think 25 year old Radjabov might still be on an upward trajectory as well, and perhaps rated below his true strength?

Agree with your points BJJ (and welcome back ) I might have time to dig those stats for you from chessbase once I get home. My db is slightly outdated (last games are from Feb 2012 I think), but that's more of an annoyance than a true problem I guess.
Chessbase is doing a series of portraits on the players, here are the first two (they have stats too! Draw rate seems to be around that 70 % for Svidler and somewhat smaller for Ivanchuk):
Svidler and Ivanchuk

Yeah, I think draw rate and relative strength with Black vs. White will vary between players.

I think I would take each of the players' lifetime performance vs. 2700+ opposition and use that as a baseline to estimate their strength with White and Black pieces, as well as the draw rate.

-70% just seemed about right given some head to head stats i saw for top guys. looks like ivanchuk's is 64% and svidlers is about 70%. but ivanchuks is much closer to 70 against the lower rated guys and much lower than 70% against aronian/carlsen (although he does have more wins vs aronian than losses). my formula does also correct itself for bigger rating differences. once the rating difference is larger than 110 points the % of draws starts to rapidly decrease. which would explain ivanchuks differing results vs the carlsen and vs the similar rated guys

-150 points probably was a bit too much advantage for white. but 50 points didnt seem enough in view of carlsen being at a disadvantage vs most of these guys with black. i think he expects to lose more often than win with black vs the 2775-2815 players, yet since he has over a 50 point rating advantage on them the formula would give otherwise. and coming up with individual formulas for each player gets pretty messy since there are 56 individual matchups (8 choose 2 times 2 for white/black)

-carlsen has been really rapidly and steadily going up. his performance ratings have been 2840-3020 the last 18 months. most of the other players have leveled off already, but radjabov is the only other one at his peak. i actually thought he had leveled off, but he hasnt played any rated games in a few months. his progress has slowed down a bit since hes been a 2700+ for 8 years, but i think youre right, he probably deserves about 20 point boost as well since he seems to be approaching ~2810 asymptotically. so yea im cool with changing his also

Maybe it would make sense to use something like player's performance rating over last 100 rated games or something like that?

That would capture players who are improving/ regressing and 100 games should be sufficient to smooth out random fluctuations.

seems hard to go through all those games and figure out the average rating for each guy he faces, the records, etc.

couldnt get c++ to work on my mac or running windows 8 in parallel, but downloaded python. its similar enough to c++ where i can still program. got nothing going on today either, will program this in a few different ways and let you know when i get some results later

You people could just use this
http://www.chess-db.com/public/tperf.jsp
Pretty awesome site btw. Finds random information about players you couldn't find otherwise.