A highly informative overview of current FIDE ratings may be found here: http://www.chessbase.com/newsdetail.asp?newsid=6027. Basically, this is a system by which chess grandmasters get rated based on their tournament play over a lengthy period.
The top chess players of recent years (post-Kasparov) are the trio of Kramnik (Russia), Topalov (Bulgaria), and Anand (India), with Anand being the current world champion and with Topalov scheduled to play him for the title in the very near future. Each of these guys has been at the top of his field for at least a decade and Magnus Carlsen has done nothing as yet to displace them from the very top. Even when it comes to the FIDE ratings, all these players are neck and neck (Carlsen: 2810; Topalov: 2801; Anand: 2790; Kramnik: 2788).
What is amazing here is not the absolute ratings but the prodigious nature of Carlsen's ascendancy to the highest rating in the world, which can be seen by comparing the following graphs showing the ratings progress from 2001-2010 for each of these players:
As the graphs show, Topalov, Anand, and Kramnik have each had a 2700 or greater ranking (more or less without fail) for each of the years from 2000 through 2010. They are the old guard who have dominated the game for years. They have significant variations in their play (for example, Kramnik is super-methodical and cautious, preferring to grind his opponents slowly while opting for drawish variations in his manner of play while Topolav is a tremendous risk-taker who frequently likes to bust up the positions with wild variations and super-aggressive tactics) but the chess world has lived with this for some time now and there is nothing new here.
In contrast, Carlsen has moved in a virtually unbroken upward line from about a 2000 FIDE rating in year 2001 (when he was 10 years old) to his current 2810 rating in year 2010 (he is now 19). Thus, he represents the child prodigy coming into maturity as a breath of fresh air to the chess world - though he is not the world champion, for him, it is all potential future greatness and this has stirred tremendous excitement. It doesn't hurt that Carlsen is likable and is being mentored by Kasparov, who is perhaps the greatest grandmaster of modern times.
It is not slightly misleading, it is very misleading. To become champion, he must win the world championship, which involves winning the candidates tournament, and then beating the current champion in a match. Which is very different then being the top-ranked player.
... is that nobody has ever become world champion at a younger age than their twenties.
Being a chess idiot, I don't really know why this has been true, but I assume it's simply a matter of practice: that it takes time to get good enough at chess to usefully train against grandmasters, and then it takes time with the grandmasters before you can dominate at grandmaster-level chess, which my chessmaster friends have told me is almost like a different game, one which they can follow with great excitement but not necessarily understand well enough to play.
The book "Talent Is Overrated" cites some interesting studies on chess players that support the idea that this is probably the main factor. Basically a combination of improved training methods, long hours of deliberate practice, and starting at as young an age as possible. There isn't much in the article that would lead me to think otherwise. The comparison to Mozart is interesting as the book goes into much detail citing evidence how much of Mozart's "talent" was due to his father's efforts to deliberately train him to become a great musician.
I agree that the title is misleading. But keep in mind that every player who has ever been ranked #1 (Fischer, Karpov, Kasparov, Kramnik, Topalov, Anand) has gone on to become World Champion. It's just a matter of time before Carlsen gets his shot at a title match.
If you're a purist, you probably believe that one can only become world champion by defeating the title holder in a head-to-head match consisting of of many (12-24 or more) games. That's how it was done from 1886-1993, and I think most chess players (myself included) prefer that format. But Topalov is technically recognized by FIDE as a World Champion.
Any professional go player can remember their own games as matter of course. They also replay hundreds of pro games for study, thereby memorizing them. I'd be surprised if it were any different for chess players.
In fact, even amateur low-dan -- advanced but far from professional level -- players can replay games they have played in the same day, and can memorize a few games with some practice.
Go games often last more than 200 moves but players memorize sequences, not isolated moves.
Pros can routinely read 50+ moves ahead but, as tsipiteri mentioned, the hardest part is evaluating the positions.
While there is no single "world" pro championship, a couple of Korean teenagers have won prestigious international titles. A 19-year-old who's "merely" a contender wouldn't be called a prodigy in the go world. I guess Asian go players just train more and are more competitive than chess players. Unlike Carlsen, they don't usually care about a life besides their game.
Somehow go always makes its way into online conversations about chess.
The Wikipedia article on computer go suggests that humans are better at calculating long sequences in go than in chess because the pieces in go stay where they are unless they're captured. This seems reasonable to me, though I'm no better than a beginner at go (I'm probably around the equivalent to an "amateur low dan" at chess, and can do what you suggested a go player of that level should be able to do in replaying and memorizing games.)
Also, it's not really fair to compare international titles in go, of which several are contested every year, to the world championship in chess, which is only contested every other year. A better comparison might be some of the major international tournaments like Wijk aan Zee, Linares...or perhaps this year's Nanjing Pearl Spring tournament, in which Carlsen decisively defeated a field consisting of five other top-20 players: Veselin Topalov (currently world #2, and the challenger for the world championship to be held later this year), Wang Yue (#9), Peter Leko (#12), Teimour Radjabov (#16), and Dmitry Jakovenko (#19). Carlsen scored 8/10 (six wins, four draws) -- no one else scored above 5.5/10.
"His early coach Simen Agdestein successfully juggled being Norway's chess No. 1 and a national team soccer player..."
It would be fascinating to hear if he was able to leverage anything from one of those realms to the other. Did he see football as a more fluid kind of chess?
Any other examples of someone achieving that level of excellence at both a physical endeavor and an intellectual one? Has there ever been another national chess champion who was simultaneously a professional athlete?
I'm highly inspired by Josh Waitzkin, who inspired the book and movie "in search for Bobby Fischer" and then went to win 21 national championships and 2 world championships in Tai Chi Push Hands, a competitive kind of tai chi.
He explores how he's linked chess and tai chi with overall 'learning' in his book 'the art of learning' which I highly recommend.
He's alse been training BJJ and said he expects to be world champion by year 2010.
Lord Dunsany comes to mind. Note quite the same, but he was a national chess and pistol-shooting champion in Ireland. Also, he was a really good fantasy author. He's from the pre-Tolkien era, which is somewhat rare and very distinctive.
I hear that chess champions have to have a surprising amount of physical stamina to play chess for so long.
It's really not possible (or practical) to plot 20 moves ahead in a chess game. Every different piece moved by your opponent would represent a new 'line' that you would have to compute 20 moves for.
> It's really not possible (or practical) to plot 20 moves ahead in a chess game.
20 'best' moves ahead, according to the observer/player/chess-engine; assuming opponent played the best moves (according to the observer) in reply.
Remembering a game played 6 years ago or being able to "think" 20 moves ahead may seem an amazing feat to regular chess players. But for professional players of GM or IM status - I don't think it is something amazing. Professional chess players memorize 100s (if not 1000s) of chess opening moves and their variations for many years. Remembering a chess game is not very impressive in that context.
nanijoe is still pretty much technically correct (the best kind of correct). It's just that when chess players talk about plotting "n moves ahead", they mean something different than a programmer talking about a program computing "n moves ahead". Top chess players evaluate a handful of possible lines of play very very deeply. But it's impossible for the human mind to evaluate every response and counter-response to any reasonable depth, which is why computer players dominate so much in tricky tactical positions.
I'd imagine at the top level being able to see say the 20 best moves ahead would regularly occur similarly as predicted by the player and if things deviate it could be there there advantage anyway.
Yes, and in fact that's how openings work. Each opening is a sequence in which both players are making an optimal response to the previous move. If a player deviates from the opening, then by definition he just made a suboptimal move.
There are a relatively small number of equally optimal moves (compared to the total number of possible moves) for a given opening, after white has opened in a certain way. The vast majority of moves other than the ones covered in the opening and/or its known variations are indeed sub-optimal, and quite often if someone deviates from one of the great many standard openings, this can be exploited by the opponent.
The truth of the matter is that chess players seldom play out a single opening..Since most openings are pretty well known (along with their optimal moves), the beginning part of a chess game is usually spent trying to maneuver your opponent into an opening of your liking (or trick them into thinking you are playing one thing, then switching on them).
What starts out looking like the Slav , may end up Queens Indian (as an example), and this makes the 20 moves ahead claim even more ludicrous.
There are whole books on (slaying) the dragon variation of the Sicilian defense and the Sicilian is just one option that can result from the first move being e4..of course your opponent can also transpose to a different formation as the game progresses.
In the end game where there are relatively few pieces on the board you may be able to plan ahead with more precision, but other than that trying to think 20 moves ahead is not just impractical, it may well be dumb.
This is what I meant. Openings aren't traps, and openings can branch as players choose different variations, but to stray from the opening book entirely is seldom a wise move. (That's one reason I never got too far into chess; I realized that competitive play would require a lot of opening knowledge.)
I'm not a chess player but I'm pretty sure that's not how it works.
20 is only the depth of the search, the width is variable. Good players can prune the game tree aggressively -- i.e. they select only a few candidate moves at each step -- while still getting a good approximation of the best move.
You're correct. The branching factor in a typical chess middlegame is 35-36, but top players usually only consider 1-3 moves per position. This is why the best humans were able to beat computers at chess for such a long time. (In the last 5-10 years the computers have pulled ahead, for good.)
In the variant called suicide chess it's routinely possible for humans to recognize wins 30 moves in the future. I've seen strong players do this in fractions of a second during lightning games. One factor contributing to this is a comparatively low branching factor in terms of raw move legality. But I think it's similar (maybe slightly lower) to the effective branching factor a grandmaster considers in regular chess.
Another factor is that it's not necessary to see the actual moves. Humans can immediately evaluate patterns that have a known outcome that many moves in the future without needing to enumerate the exact moves of the principle variation in their mind. You might argue that this isn't really what is meant by "x moves ahead". But I think this more "fuzzy" definition may very well be more accurate.
In Gödel Escher Bach, Hofstadter has three sections on "chunking." Chunking is distinguished from the type of branching calculation of future moves. Rather than evaluating linear move-countermove-move branches, humans can also evaluate in terms of previously recognized patterns or configurations of pieces [essentially what you're talking about, mightybyte, if I understand you correctly]. This kind of pattern recognition or pattern seeking is a substantial shortcut, reducing the amount of mental computation necessary in branching approaches. It's also an area in which human still potentially retain an efficiency advantage over computer opponents like Deep Blue.
The top chess players of recent years (post-Kasparov) are the trio of Kramnik (Russia), Topalov (Bulgaria), and Anand (India), with Anand being the current world champion and with Topalov scheduled to play him for the title in the very near future. Each of these guys has been at the top of his field for at least a decade and Magnus Carlsen has done nothing as yet to displace them from the very top. Even when it comes to the FIDE ratings, all these players are neck and neck (Carlsen: 2810; Topalov: 2801; Anand: 2790; Kramnik: 2788).
What is amazing here is not the absolute ratings but the prodigious nature of Carlsen's ascendancy to the highest rating in the world, which can be seen by comparing the following graphs showing the ratings progress from 2001-2010 for each of these players:
Carlsen: http://ratings.fide.com/id.phtml?event=1503014; Topalov: http://ratings.fide.com/id.phtml?event=2900084; Anand: http://ratings.fide.com/id.phtml?event=5000017; Kramnik: http://ratings.fide.com/id.phtml?event=4101588.
As the graphs show, Topalov, Anand, and Kramnik have each had a 2700 or greater ranking (more or less without fail) for each of the years from 2000 through 2010. They are the old guard who have dominated the game for years. They have significant variations in their play (for example, Kramnik is super-methodical and cautious, preferring to grind his opponents slowly while opting for drawish variations in his manner of play while Topolav is a tremendous risk-taker who frequently likes to bust up the positions with wild variations and super-aggressive tactics) but the chess world has lived with this for some time now and there is nothing new here.
In contrast, Carlsen has moved in a virtually unbroken upward line from about a 2000 FIDE rating in year 2001 (when he was 10 years old) to his current 2810 rating in year 2010 (he is now 19). Thus, he represents the child prodigy coming into maturity as a breath of fresh air to the chess world - though he is not the world champion, for him, it is all potential future greatness and this has stirred tremendous excitement. It doesn't hurt that Carlsen is likable and is being mentored by Kasparov, who is perhaps the greatest grandmaster of modern times.