2009 illustrated the trend most predicted when Garry Kasparov retired in 2005. There's no dominant figure, no clear favorite before every tournament, and outsiders had plenty of chances. That the three top players of recent years -- Anand, Kramnik, and Topalov -- were relatively inactive for long stretches in 09 made it even more of a Wild West.
Of the strongest tournaments of the year, no one dominated. Karjakin won Corus, if with the heaviest of the heavyweights absent, ahead of Aronian, Radjabov, Carlsen and the surprising Movsesian (who was soon to drop 50 rating points as quickly as he had gained them). Grischuk took Linares on tiebreaks over Ivanchuk with world champ Anand on an even score. Aronian took Amber just ahead of Anand, Kramnik, and Carlsen. Aronian won the Nalchik Grand Prix in April ahead of Leko and Akopian. Shirov, nearly forgotten as a top contender, turned in a clutch final-round performance to take the MTel ahead of Carlsen and Topalov. The overlooked Bazna tournament reached a 2729 average and Ivanchuk won a powerful +4 victory ahead of Gelfand. Kramnik finally came back to the board at Dortmund and cruised yet another title at "his" tournament ahead of Leko, Carlsen, and Jakovenko.
Aronian dominated the brief sprint of the 2nd Grand Slam Masters with considerable flair. Then it was showtime in Nanjing. You've noticed Carlsen's name appearing regularly in the runner-up lists above. Just days after it was revealed he'd been training with Kasparov, in the 2nd Pearl Spring tournament Carlsen kicked into a gear we'd all imagined he had. His +6 left him 2.5 points ahead of Topalov and a 3000 performance rating. He slowed a little at the Tal Memorial, which again saw Kramnik with brilliant sharp preparation and a +3 victory ahead of Ivanchuk and Carlsen. Anand was back on +1. Carlsen stepped up with a phenomenal performance in the World Blitz.
Carlsen and Kramnik had been the most notable performers of the year, so it was appropriate they would meet one final time in 2009, in London. In the very first round Carlsen played a beautiful game to defeat the former world champ in what would turn out to be the decisive game of the tournament. Carlsen's three wins also enabled him to take the #1 ranking away from Topalov on the January 2010 list. The 41-year-old Gelfand took the World Cup.
Have we now entered the Carlsen era? Who will play the Karpov, or Karpovs, to his young Kasparov? Veselin Topalov showed, and still shows on occasion, the combination of ability and energy to dominate, but he has lacked the consistency event to event. He had a spectacular second half of 2008 and then barely played in the first half of 2009. With that in mind, his +2 at MTel (behind Shirov) and +1 at Pearl Spring (way behind Carlsen) weren't bad at all. Now he's skipping Corus and will play in Linares before tackling Anand in their April world championship match. Rating and ranking be damned, if Topalov beats Anand in Sofia, Carlsen the heir apparent fades into the background at least for a while. And, by the way, this is as it should be, with the emphasis on the world title where it belongs. But only if FIDE can maintain a respectable cycle will this continue to be the case. Otherwise, the rating list will continue to get the over-emphasis it has received since Kasparov split from FIDE way back in 1993.
If Anand wins, a la Petrosian turning back Spassky the first time with the specter of Fischer already looming on the horizon, we will preserve the feeling of a fading status quo until the candidates finally begin, probably not until 2011. And then... Carlsen, ready already? Kramnik, revitalized? Aronian, ready to put away childish things? Topalov, dangerous as ever? Heck, Gelfand seems to be getting better by the month, so all the delays would seem to be to his advantage. Or are we waiting for someone else to step up?
"Heck, Gelfand seems to be getting better by the month, so all the delays would seem to be to his advantage."
:)
Go Boris!
Concise outline again, Mig. Well done.
I think there are exciting times ahead with several super-GMs, of nearly equal strength, jostling for the top position. Hopefully there'll be more exciting chess as the styles of the dominant players are varied. And it may well be that the spotlight around Carlsen will attract even more 'clean' corporate sponsorship to the WCC cycle and make transparency in FIDE inevitable. The good times could well be on the way back.
"are we waiting for someone else to step up?"
As far as the next WCh cycle is concerned: Gashimov?
Carlsen is overrated. Don't expect him to become World Champ until 2013. Topalov, Anand and Kramnik are not done yet. Carlsen is no Fischer nor is he a Kasparov.
Overrated? Nobody is overrated. If his 2810 is a fluctuation or reflects a higher lever achieved by the young Norwegian, only time will show. Sure Blackeye, he is not a Fisher or a Kasparov. He is more of a Capablanca or a Karpov accroding to Kasparov. In addition, Fisher and Kasparov was not #1 at the age of 19, so indeed, he is not like them, so far. If he will becom a World Champion in classic games, not only Blitz, only time will show. Dont' try to be a Nostradamus.
Carlsen participated in the three strongest tournaments ever rating wise. Second in Bilbao and Tal Memorial, and first with a 3000+ TPR in Nanjing. This when he was 17-18 years old, and even if rating comparisons are misleading these events were very strong if one looks at world rankings. And it was only in Nanjing Carlsen was totally healthy.
The four strongest tournaments Fischer played in his career were slightly weaker world ranking wise: Candidates 1959, 5-6th, 16½ years old, Bled 1961, 2nd (18½), Curacao 1962, 4th (19), Santa Monica 1966, 2nd (23). So Carlsen's results this far are worthy of much respect, even if it doesn't mean that he will score better results ten years from now than Fischer did when he was 28-29. But few chess players have done that...
Carlsen overated? I do not think this is true and I am pretty sure Kasparov doesn't think this is true either. His chess is exciting and he continues to improve. As he gets the positions he likes out of the openings you will see more fantastic games from him. And we are priviledged to see this.
The immediate future of chess looks bright; plenty of good and entertaining games for chess fans! The period of time from 2010 thru 2012 looks to be very interesting. I hope that there will be a simplified format for determining World Champion challengers after this cycle. The "old" system of the 60's - 80's seems sound after the mess of the last two decades.
On a side note, computer programs and databases continue to help the many strong players enhance the quality of their play. Tablebases are helpful but do not yet approach "spoiling" status, at least not until we reach the capability of say 16+ pieces. Isn't it interesting that chess seems to be solving "backwards", i.e. being deconstructed? We're simply working our way until we reach the "holy-grail" of the 32-piece tablebase.
The growth of tablebases is exponential and certainly wouldnt be feasible beyond an extra couple of pieces from now.
I once did a computation of what a 32 piece tablebase would look like. Assuming compression as aggressive as shredderbase, it would require a storage system such that each atom of the moon stored a byte of data. Quantum computing could 'theoretically' reduce the eons to compute it, but would have no impact on the storage requirement. Not gonna happen.
So what you're saying is, it'll be on the market for Xmas. Excellent.
A victory for Topalov will be the best thing for the Chess World Championship (his manager issues aside). Although I prefer Anand, I hope Topa wins.
After the mess Kasparov left the World Championship in, the three players since 1995 who deserve to have their names as World Champs are Kramnik, Anand and Topalov. All have been ranked #1 and all won a worthy championship match / tournament, although Topalov's isn't really recognised given Kramnik was still an alternative champ. But Topalov's victory was the most impressive of all, so maybe if he finaly wins this year, we can start to count the WC from 2006 (after a decade in limbo) with all three represented and handily Kramnik being the first to the crown (since he got it from the King himself, although of course he had no right contesting that final).
A situation with Kramnik as the 14th Champ, Anand the 15th and Topalov the 16th is the only way a historical view of the Championship can get back on a linear track and keep everyone fairly happy. We can easily discount the wins for all of the disputed Fide champs save Anand (who now has a genuine win) and Topalov (who swept everyone aside for his title). So for that reason alone, I hope Topalov does it and we can all go on and retrospectively rewrite the history of the World Championship as outlined above.
Mr. Blackeye just talk to make atention for him self that is all. No body shuold care if he becomes so silly. I do not care I only look out window. Next thing he may say Topalov is not good. Sorry Mr. Blackeye.
While he may take some time to win the world championship, i think that the era of Carlsen has began. He is focused without being fanatical, and he seems to be putting all the right preparations in place. I greatly look forward to this year.
"A victory for Topalov will be the best thing for the Chess World Championship (his manager issues aside)."
A victory for Jeffrey Dahmer in the "Milwaukee Man of the Century" competition will be the best thing for that city (JD's dietary issues aside).
"A victory for Topalov will be the best thing for the Chess World Championship (his manager issues aside). Although I prefer Anand, I hope Topa wins."
Interesting thought!
It astounds me that anyone "nay-says" scientific accomplishments. Rather than cite the lengthy lists of things the nattering nabobs said "will never happen" that have, I'll just say to vxqtl, you could not be more mistaken! It's only a matter of when. My bet? Without a double prior to the 22nd century, and it would not shock me to see it happen in the next 30 years. You have not a clue as to the technological breakthroughs coming in that period of time, so don't bother responding in the negative again.
IMO Carlsen reaching the number 1 spot in the ratings was helped at least a little bit by FIDE switching to 2 months period between lists rather than 6 months as it was previously. It's interesting that even frogbert, who is always very careful about such details, forgot this.
My professional field of computational complexity theory is exactly about "nay-saying" things we can't accomplish, at least not always. In the same breath I must admit that this field still hasn't proved any marked nay-statements, even ones that would be far more modest than the famous Millennium Prize Problem of proving "P != NP".
In this concrete instance, however, one can make a pretty good nay-case. According to Wikipedia, the "Shannon Number" for the game-tree size of chess is estimated between 10^120 and 10^123. Right in that range is the "Bekenstein Bound" of 10^122 bits for the observable universe, which as explained by a colleague here (http://www.scottaaronson.com/democritus/lec20.html) is a limit on the size of any computation. Put another way, if a 32-piece Nalimov tablebase existed---even if it would be serviced by Rybka clusters in other galaxies---our entire Universe would collapse into a black hole!
I have heard about blck hole and mabey where we all end. Mean while play chess.
"if a 32-piece Nalimov tablebase existed...our entire Universe would collapse into a black hole"
It would be worth it, wouldn't it? :)
That's about the most informative post here on DD in a long while! :)
Thank you for all comment. Be care full of black hole.
Wow. Great post.
Chess is bigger than the universe!
The day we solve chess, the whole world and it's suroundings will dissapear into a black hole!
Now I just have to try to memorize this thing about the "Shannon Number" and the "Bekenstein Bound" of 10^122 bits for the observable universe.
I'll pull it when in need to impress. -When somebody implies that chess is not much more than a game of poker.
Dumb idea, more than likely, but is there a possibility that some way will be found to expand the six man database based on the fact that there are only six different pieces in chess? As in, some way of multiplying the calculations to embrace the rest of the pieces based on information already available, rather than calculating all possibilities of the 32 man set by "brute force". Anyone undestand the gist of what I mean to say? Excuse the fogginess, not a technical person, just curious, indulge me :)
I think this is nto right because there are 32 pieces insted of 6.
Clap, clap.
"According to Wikipedia, the "Shannon Number" for the game-tree size of chess is estimated between 10^120 and 10^123."
As of right now the article says it is 10^111. Go wikipedia! Maybe tomorrow it will say 10^130.
Yes thank you for clapping. Some time I am correct I think most I am wrong very wrong. But I know of 32 so I must only be full of luck to be corect.
There's a popular exposition of this idea in Leonard Susskind's _The Cosmic Landscape_.
With the current knowledge and technology we have, we may not reach 32-pieces tablebase. But what about when we have more knowledge and more advanced technology?
12 songs on an LP decades ago vs. virtually hundreds of song in a USB thumb drive. Da Vinci though about flying human centuries ago, before the Wright brothers attemp early last century.
So 32-pieces tablebase in 2410 perhaps?
Enough babbling, I don't think I can memorize perfect play for 8 pieces, let alone 32 pieces.
A situation with Kramnik as the 14th Champ, Anand the 15th and Topalov the 16th is the only way a historical view of the Championship can get back on a linear track and keep everyone fairly happy. We can easily discount the wins for all of the disputed Fide champs save Anand (who now has a genuine win) and Topalov (who swept everyone aside for his title). So for that reason alone, I hope Topalov does it and we can all go on and retrospectively rewrite the history of the World Championship as outlined above.
***
No, sorry, not everyone agrees with perpetrators who try to steal the title via the private title route.
Thus, the line is:
# 1 Steinitz (though he thought Zukertort was champion in their match so perhaps we should start as follows)
***
# 0 Zukertort
# 1 Steinitz
# 2 Lasker
# 3 Capablanca
# 4 Alekhine
# 5 Euwe
# 4 Alekhine
# 6 Botvinnik
# 7 Smyslov
# 6 Botvinnik
# 8 Tal
# 6 Botvinnik
# 9 Petrosian
# 10 Spassky
# 11 Fischer
# 12 Karpov
# 13 Kasparov
# 12 Karpov
# 14 Khalifman
# 15 Anand
# 16 Ponomariov
# 17 Kazimzhinov
# 18 Topalov
# 19 Kramnik
# 15 Anand
Sorry, that is the line and will always be the line of champions. It reflects the continuous run of championships from the start thru 1948 and FIDE (all the federations of the world)taking control of the title process...thru the criminal breakaway period. It is unfortunate that the titleholder had to be forfeited in 1993, but it happened.
And 19 champions is no more cumbersome than 15 champions.
All parties are represented -- including Anand and Kramnik -- though their numbers and title reigns are different under the true line than under the alternate history.
I just trust that the chess world cannot and will not allow a title line of stolen property and outright thievery to be preferred over one based on active play and fair/open qualification methods.
Thus, if Topalov wins...he will REGAIN his place as #18....he will not be # 16.
chesspride, why would you pick such a strange lineage for the WCC? Yeah, yeah, FIDE wants to call the winner of a goofy knockout tournament the "World Champion" for a few years, but why on earth...
While I agree with the spirit of your post, I must differ with the details. Game tree size is not the relevant number for storing a tablebase (though it is certainly relevant to the computational cost of building the tablebase). The tablebase need only encode a small amount of information per distinct legal chess position. This is also a huge number, but *much* smaller that the game tree size.
Further, one may envision techniques to use an incomplete tablebase as if it were complete, without sacrificing the 'full solution' offered by the tablebase. For example, you can skip 90% of the final positions if any skipped position is at most one move from a stored position. Then, using the tablebase, if a position is not present, generate and search for all one move away positions, and analyze. With numbers this big, sparse + compression techniques don't change the big picture, but they do explain my belief that the moon's atoms worth of bytes might be sufficient to *store* a usable 32 bit tablebase representation.
chesspride, why would you pick such a strange lineage for the WCC? Yeah, yeah, FIDE wants to call the winner of a goofy knockout tournament the "World Champion" for a few years, but why on earth...
****
Strange? After the forfeit of Kasparov and Short in 1993...that *is* the line of champions.
FIDE isn't just some private group -- it is the 100+ national chess federations of the world coming together to run the title cycle. Thus, everyone has a stake in the correct line.
And in 1993...when the breakaway occcurred...it was an attempt to take what rightfully belonged to the chess federations of the world i.e. you and me.
And hundreds of players played in those cycles that you seem to discount from 1993 to 2008 (unification). The federations of the world spent quite a large amount of money to support hundreds of players in the attempt to win those titles.
FIDE came into existence in 1924. FIDE had no real authority over the World Chess Champion title until 1948, when a successor to the dead Alekhine had to be determined in a classical tournament. Except for this tournament, the real lineage of world champions has depended on each of them winning the title in a match of classical chess. Because of that, numbers 12 through 18 on your list cannot be considered World Champions in accordance with historical precedent.
"IMO Carlsen reaching the number 1 spot in the ratings was helped at least a little bit by FIDE switching to 2 months period between lists rather than 6 months as it was previously. It's interesting that even frogbert, who is always very careful about such details, forgot this."
This makes no sense. If they were published at 6 months, they would still be published in January and Carlsen would still be number 1. The frequency of FIDE (or the live ratings) publishing lists does not affect the calculations at all. FIDE and the live ratings are calculated the same way. Many posters think the live ratings are only based on recent games. This makes no sense, because when Anand or Topalov took months off they would not appear in the live ratings and they're obviously in the top 3.
British fan, i get what Jean is saying but he is not right.
Basically he says because the ratings are calculated more frequently, it helps someone rise much faster.
But that is not true, more so in the case of Magnus.
With more frequent lists, his rating kept rising less because he was becoming more higher than his opponents. If the six months period were used, with the number of games he played he'd gain a lot more points with a "lower" rating and actually be higher than he is now.
I'm not sure what people mean when they say "Carlsen is overrated."
If they mean that his strength is overrated, then that doesn't make any sense because the same rating formula is used for everyone. (I really don't think any sane person would mean this, but then again, I have doubts about the sanity of some of the posters on this blog.)
If they mean that he is overhyped, then that's OK. Saying that Carlsen is overhyped is an opinion. And we know what they say about opinions and assholes: everyone has one.
If they mean that Carlsen's staying power is overrated, then that to me is the most interesting issue. What I think separates Kasparov and Karpov from the rest of the field is the length of time they dominated. Based on what we have seen so far, if he wants to, Carlsen can dominate chess like Kasparov and Karpov. The question is how badly does he want it...
Your entire post makes no sense, because you don't understand the most basic facts about ratings. Frequency of calculations or publishing means nothing. Math is math. Why can't you guys grasp this simple concept? I honestly don't get it.
While I am also "not a technical person", I would answer "no" to your question ["is there a possibility ...?"].
Approaching the problem from two distinct but obviously related angles:
1) from a computer point of view, the number of positions to be evaluated increases exponentially rather than linearly. Simplistically
- if each side has ten possible moves, Rybka/Frits/Shredder has to consider 10 new positions after one half-move, 100 after a full move, 10,000 after two moves, 1 million after three moves, ... . [ignoring piece captures that simplify the position, only moves by either side, ....]
- if each side has twenty possible moves, the numbers become 20; 400; 160,000; 64 million, ... .
2) from a human point of view: if there are several pieces on the board they do not act on their own but INTERact with each other
A provocative, blunt, rude question: In your own games, do you only strive for maximum activity of each piece? Or is piece coordination also part of your strategy and calculation process? IMO, only in the first case the concept of merging simple databases into a more complex one would be applicable.
A related question to computer specialists: Humans generally (should) know the difference between a middlegame and an endgame, and this might in some ways affect their thinking/calculation process and choice of moves - in practice, the tricky part may be to realize when exactly a middlegame becomes an endgame.
Do computers also know the difference? Clearly only if they are told so by programmers - beforehand based on general principles. For example, would there be a "rule" such as "king safety becomes less important once queens are exchanged"? In certain positions, this might backfire ... .
British Fan,
i disagree...the frequency of the lists can affect how things turn out..and the more frequent the lists, the more accurate your rating.
Say in July 2009 Carlsen met a player rated 2700 and won 2 games against him. He meets same player in Nov and wins 2 games against him. he will still be rated 2700 and he would win the same number of points against him as in July. For 6 months lists.
Scenario two,say for 2month lists, when Carlsen meets the same player again in Nov, Carlsen would be higher rated so even if the 2nd player is still 2700, Carlsen will gain less points in Nov than in July.
That is why i am saying that for someone whose strength is increasing ''rapidly'', the more frequent lists are likely to ''slow'' his rise than if longer period lists were used. The good side is they will be a more accurate representation of his strength.
I hope that makes some sense.
"The tablebase need only encode a small amount of information per distinct legal chess position."
You only need to consider the initial position to realise the fallacy of the above. The initial position needs to include every possible position in its tablebase which therefore includes every possible position. Bear in mind tablebases SOLVE the positions they dont assess at the end of a line that White is a queen ahead so they dont need to analyze further. KWregan is correct
chesspride: "Strange? After the forfeit of Kasparov and Short in 1993...that *is* the line of champions."
Yes, "strange." A rigorous qualification process and a match is a reasonable way to pick a WC that can at least plausibly be argued as the strongest player in the world. The quasi coin-flip knockout format to determine WCC was a failure: Khalifman, Ponomariov, and Kazimzhinov are fine players, but I don't think anyone would seriously argue that they were the best players in the world when they won their titles. The 1998-2005 FIDE title has no more relevance to the WC than, say, 1998-2005 Linares or Corus winners. No amount of money that FIDE spends can change that.
No, vxqtl is right. A tablebase doesn't "solve" anything, it just provides you with the solution already generated.
Fully Agree.
Hi chesshire cat,
I forgot to ask if you would mind contesting the Cochrane gambit sometime. I'd like you to demonstrate it's unsoundness to me.
"Fully Agree."
Fully??? Also regarding Topalov's "manager issues"?
Chortle. Well I suspect the outcome would depend more on our relative chess strength than on the intrinsic merits of the opening, and I am unaware of your approx rating, Hardy-mine is 1900 give or take. If you indicate your handle on any free server or chessbase or any other means of allowing the battle to take place, I'll be glad to cross swords with you, though the Petroff is an opening I use for variety rather than my main weapon.
Beautifully written!
--h
2 + 2 = 4 regardless of how often I calculate or publish my results. I can't make my point any more basic so you can understand it. My 8 year old son gets this. I hope this makes sense to you.
FIDE came into existence in 1924. FIDE had no real authority over the World Chess Champion title until 1948, when a successor to the dead Alekhine had to be determined in a classical tournament. Except for this tournament, the real lineage of world champions has depended on each of them winning the title in a match of classical chess. Because of that, numbers 12 through 18 on your list cannot be considered World Champions in accordance with historical precedent.
****
Jim from Sudbury spits in the face of the 100+ national chess federations who make up and fund FIDE...he obviously cares little for his own national federation's efforts to support and fund players who actually competed in the world title cycles that he derides.
Hundreds of players...and millions of dollars...were spent on those cycles.
Instead, you opt to worship those who take from you -- take your property -- the title of world champion and the title cycle the belongs to FIDE (and by extension, to every one of us).
The players who were forfeited in 1993 had earned their status in the FIDE cycle. They could not just grab the goodies and walk away, any more than the Pittsburgh Steelers could grab the Lombardi trophy and leave the NFL.
Their status in 1993 depended solely on their achievements under FIDE. The fact that the nations of the world ran the cycle only after Alekhine died is irrelevant. They ran it during Kasparov's ascendency in 1984-85...they ran it during his reign 1985-1993, and they ran it during and after his default. Same for Short -- his short-lived candidacy came to a grinding halt after he conspired to take the property of the national chess federations of the world.
I'm sorry that readers here seem more concerned about the property rights of a few than they do for the property rights of the many...but that's not how history works.
British Fan,
i disagree...the frequency of the lists can affect how things turn out..and the more frequent the lists, the more accurate your rating.
Say in July 2009 Carlsen met a player rated 2700 and won 2 games against him. He meets same player in Nov and wins 2 games against him. he will still be rated 2700 and he would win the same number of points against him as in July. For 6 months lists.
Scenario two,say for 2month lists, when Carlsen meets the same player again in Nov, Carlsen would be higher rated so even if the 2nd player is still 2700, Carlsen will gain less points in Nov than in July.
That is why i am saying that for someone whose strength is increasing ''rapidly'', the more frequent lists are likely to ''slow'' his rise than if longer period lists were used. The good side is they will be a more accurate representation of his strength.
I hope that makes some sense.
***
Frequency of lists matters because FIDE uses (or used to use) the same initial rating for a player for the entire rating period.
Example: Start at 2700 and play 10 tournaments. You are 2700 for each of the events, regardless of outcome...and you get a post-rating period rating.
USCF uses an event-by-event model...where you would start at 2700 and then go up or down based on results in those 10 events.
A faster or slower FIDE cycle certainly would impact the speed of rating increase (or decrease).
A faster or slower USCF-style cycle wouldn't matter as it is a tourney by tourney calculation.
chesspride: "Strange? After the forfeit of Kasparov and Short in 1993...that *is* the line of champions."
Yes, "strange." A rigorous qualification process and a match is a reasonable way to pick a WC that can at least plausibly be argued as the strongest player in the world. The quasi coin-flip knockout format to determine WCC was a failure: Khalifman, Ponomariov, and Kazimzhinov are fine players, but I don't think anyone would seriously argue that they were the best players in the world when they won their titles. The 1998-2005 FIDE title has no more relevance to the WC than, say, 1998-2005 Linares or Corus winners. No amount of money that FIDE spends can change that.
***
You cannot pick and choose outcomes you don't like...
Anand vs. Shirov ring a bell? Anand is one of the winners you deride.
Ponomariov vs. Ivanchuk? There seem to be plenty of Chucky fans on here and he was poised to win before having a melt-down.
This isn't about FIDE. This is about the fact that the chess federations of the world adopted this model. FIDE isn't separate -- FIDE is all of us.
British fan, unless I misunderstand your posts, you have failed to grasp the following, as explained above:
Ratings may be published, for example, once or twice a year. This will lead to different rating results. If rapidly improving player x is 2750 at the beginning of the year and remains so till end of December, his calculated rating by year end will be different than if from say July it is calculated from a base of 2790. It is NOT 2+2=4.
Chesspride, good to see you persist with your views on the cycle/titles. I agree with most of your points - though I wouldn't blame any of the players (other than Kasparov/short who started the breakaway) who decided to participate in the cycle after 93 for conspiracy. After all chess is their livelihood. Also there was a theoretical possibility at certain points that someone (Anand/Kamsky) could become a champion under both cycles and the alternate cycle could be dissolved peacefully without the unification mess.
Also interesting to see british fan's comments "I can't make my point any more basic so you can understand it. My 8 year old son gets this" -- while he doesn't give a thought that "may be" he is the one who is wrong. British fan -- try searching for GM Macieja's extensive articles on topics related to rating.. may be a strong GM's work (instd of these patzers' comments) on this will convince you that "may be" you are wrong.
Wrong. Whether ratings are calculated and published daily or yearly, it's doesn't matter. That's why the day FIDE publishes, the live ratings match up perfectly. The frequency of calculations and publishing does not change the laws of mathematics. I've discussed this with a math professor and an IM and they both agreed. Your preconceptions are cancelling out your common sense. My radical opinion is that 2 + 2 does indeed equal 4. ;)
Sigh. Rather stubborn, this British guy.
Who said anything about mathematical laws changing? Perhaps you might explain the rating process you espouse in detail and enlighten us.
chesspride: "Anand is one of the winners you deride."
Deride?! No. Saying that the FIDE knockouts do not have the prestige of a WC match does not deride Anand or anyone else.
And it's not about picking and choosing outcomes that I like...it's about picking and choosing a process that I like. The FIDE knockouts process does not have enough credibility to confer a meaningful WC title. Evidence? Khalifman, Kasimdzhanov, and Ponomariov were clearly not the strongest players when they won--not even close. Kasimdzhanov was not even in the top 50, according to rating! The process was not a strong enough test, and protestations that "It's official because FIDE says so!" are evidence of excessive faith in a fraud.
"That's why the day FIDE publishes, the live ratings match up perfectly."
Yes, but the live rating list uses the official published ratings as basis for calculation (as it should do, obviously). You still don't get the point, but you still like heckling all of us who do.
>> it's about picking and choosing a process that I like..
the winner of the candidates gets thrown out due to difficulty finding a sponsor, and the guy who he beat "qualifies" instead. You call that a "process"? Cheers!
"Who said anything about mathematical laws changing? Perhaps you might explain the rating process you espouse in detail and enlighten us."
Love to. It's the same one FIDE uses every 2 months and same one that Hans Arild Runde uses for the live ratings. Since they are calculated and published at different intervals, yet come up with the exact same numbers once every 2 months they are obviously the same method and you "Interval" guys are obviously wrong. What you "Interval" guys are saying is FIDE, Hans Arild Runde, and everyone else in the world is wrong. Perhaps you might want to look in the mirror.
"1. Yes, but the live rating list uses the official published ratings as basis for calculation (as it should do, obviously). 2. You still don't get the point, but you still like heckling all of us who do."
1. You agree with and have proven me correct. 2. Since we're discussing the official ratings, your point only exists in some fantasy ratings world only you can imagine, therefore in reality you have no point.
Actually, "vxqtl" may be correct after all. I thought I was covering this base by saying "Nalimov", since Nalimov bases identify optimal moves (fastest mate for winning positions, any drawing move for drawing ones). However, this is tantamount only to building a "global optimal strategy tree", not the entire game tree. There are actually 4 different concepts here:
1. Position oracle: says win/lose/draw for every (legal) position but doesn't give moves.
2. Optimal strategy tree: gives optimal moves, but can ignore all positions that are not part of the strategy.
3. Global strategy tree: gives optimal moves from every (legal) position.
4. The entire game tree, blunders and all.
A "bitbase" (e.g. Shredderbase or Scorpio) is 1. only. There the size for chess is "only" about 10^46, again according to Shannon. I believe the exhaustive solution to 8x8 Checkers is 2. only---which is why there was discussion of whether one could still beat Chinnook + that database by making a losing move then outplaying it. The distinction between 2. and 3. makes a big difference---much more generally than "just games"---in two professional papers of mine (the two at http://www.cse.buffalo.edu/~regan/publications.html where "Crasmaru" is a co-author). And 3. can be much smaller than 4., as the algorithm for generating Nalimov bases by working backwards actually shows. Indeed, 3. may not be much bigger than 10^46.
Oh well. Ray Kurzweil's article http://www.kurzweilai.net/articles/art0185.html?printable=1 gives 10^45 as the Bekenstein bound for the volume of a human being. So at least we can say that anyone trying to enter the World Open with a 32-piece bitbase under a trenchcoat would collapse into a black hole!
The article says 10^75 for the earth, 10^86 for the Solar System, 10^106 for the Milky Way... But the real significance relating to the "Cosmological Constant" is the 10^122 number (in the source linked in my original comment) for our observable universe---whether or not it's just a "Hubble Bubble" in a larger multiversal cosmos. For that we'd need Chess on a bigger board, or Go, anyone?
It's entirely true that FIDE and Hans Arild Runde are using the exact same method. He is just letting us see the process "in real time" so to speak. The relevant frequency is obviously not how often ratings are calculated but how often they actually change for the purpose of calculation. That happens every time FIDE publishes a new list. If I am 2700 and perform at 2800 for a year, but the rating list is never updated so I keep gaining points based on my 2700, I will gain much more than if it's updated frequently so that my rating gets closer to my strength quicker. With a lower rating, my statistically expected score will also be lower, making it easier to gain rating. It's very elementary, and it has already been explained to you. Stop being so arrogant and start learning
1) some manners
2) the basics of the rating system.
"Rating and ranking be damned, if Topalov beats Anand in Sofia, Carlsen the heir apparent fades into the background at least for a while. And, by the way, this is as it should be, with the emphasis on the world title where it belongs."
Lighten up, Francis. You need to have a sense of humor. ;) What Mig and I are trying to teach you young ratings/Fritz obsessed guys is that ratings don't matter. The World Championship does. Anand and Kramnik are the 2 best players in the world regardless of ratings.
Uh, the discussion was not about the importance of ratings at all. But I'll remember this trick next time I embarrass myself in a debate..
Wow, he GOT it? Can I drain the blue from my face? Fetch me champagne, caviar, and cancel that order of "Chess Ratings For Dummies"!
I agree that the knockout system carries elements of luck, and isn't the best way to determine a world champion. However:
"Khalifman, Kasimdzhanov, and Ponomariov were clearly not the strongest players when they won--not even close."
Also agreed for the two k's [the capital K is reserved for other players], but Ponomariov was at least a top10 player in 2002 when he became world champion - ELO 2744 meant a bit more back then than now, only Kasparov and Kramnik were much (>20 points) higher-rated at the time. Just after the world championship, he finished second in Linares behind Kasparov.
As Pono was a 19-year old rising star at the time, we could compare him (a bit in jest) to Carlsen today: Carlsen is now #1 and still has to become WCh. Ponomariov was world champion and still had to become #1 (or #2 behind Garry) - it's another story that he didn't manage and even dropped below 2700. Maybe somewhat related to his match against Kasparov that never happened?
"Many posters think the live ratings are only based on recent games."
Just using this as a pretext (obviously British fan completely missed the point everyone else was making)...
...but Jeweller at chesspro.ru does provide rolling performance ratings based only on the last 52 weeks. He recently analysed his "Mix" ratings (including, it seems, all games whether blitz, blindfold or classical) into performances with white and black. The surprising result, perhaps, is that Kramnik actually ended 2009 with the best performance with the black pieces as well.
http://chesspro.ru/guestnew/looknullmessage/?themeid=110&id=1&page=20 (at the very bottom)
White:
1. Kramnik 2856
2. Carlsen 2852
3. Anand 2839
4. Aronian 2811
5. Karjakin 2808
6. Topalov 2799
7. Grischuk 2793
8. Svidler 2792
9. Gelfand 2783
10. Leko 2775
Black:
1. Kramnik 2819
2. Carlsen 2818
3. Anand 2814
4. Aronian 2766
5. Topalov 2765
6. Gelfand 2753
7. Ponomariov 2750
8. Almazi 2749
9. Short 2742
10. Mamedyarov 2735
Short gets an honourable (!?) mention for being the only player to have a better performance with black than with white (-6), while Karjakin was a whopping 78 points stronger with white.
I bet Radjabov has a better TPR with black than white.
That was one of the things Jeweller was looking for, but actually Radjabov's performance was 50 points higher (above average) with white over 2009.
even if there is almost infinite number of legal positions, it is likely you can rule out significant number of them because they can not be reached before opponent could force a mate (or known theoretical win), thus the problem could be solvable in practical means by using computation
So it look like Kramnik is best insted of Carlsen. How come no Kasparov.
Interesting. Good to know. Thanks.
According to statistics on the FIDE rating pages (covering the period since October 2007), Radjabov has a slightly higher winning percentage with black (23%) than white (20%). But he has a significantly higher losing percentage with black (16% vs. 6%).
This is normal for everyone, but particularly if you play the KID!? And some of his losses were against relatively weak players (Van Wely, Sakaev) and cost more rating points ... .
Who said anything about mathematical laws changing? Perhaps you might explain the rating process you espouse in detail and enlighten us."
Love to. It's the same one FIDE uses every 2 months and same one that Hans Arild Runde uses for the live ratings. Since they are calculated and published at different intervals, yet come up with the exact same numbers once every 2 months they are obviously the same method and you "Interval" guys are obviously wrong. What you "Interval" guys are saying is FIDE, Hans Arild Runde, and everyone else in the world is wrong. Perhaps you might want to look in the mirror.
***
Yes, it is the same one FIDE uses...that is the point.
Others have commented that the live ratings are really FIDE ratings computed in real time using the same initial starting ratings.
Can we emphasize that -- THE SAME INITIAL RATINGS over the entire calculation period.
Here is an example of why British fan doesn't see this clearly:
FIDE rating 2700
Plays 5 events
Starting rating for each event 2700
Ending FIDE rating 2720
Live rating shows 2700 2690 2710 2730 2720
FIDE rating shows 2700 2720
Of course they match -- each of those intervening rating calculations was based on a rating of 2700.
Base them on 2700-2690-2710-2730-2720 and you won't match....not for all players.
Now suppose FIDE list came out after the 3rd event.
Now you have
FIDE rating 2710 as the starting rating for this player -- for everyone who plays him (instead of 2700).
Rating calculations will be different!
That is why I compared it with USCF-style ratings that change on an event-by-event basis. You never quite know what the starting point is -- because they use a fluid system that changes by event.
FIDE doesn't. The frequency of their rating release impacts the starting rating calculation.
Thus, a 6-month gap where a 2700 becomes stronger (i.e. 2800 strength) but gains points as a 2700...leads to a whopping large rating gain.
A 2-month gap where that 2700 becomes 2750 for subsequent events...leads to smaller gains afterward.
Such a gain would be much smaller still if the FIDE system rated by event.
However, one advantage for the FIDE model is transparency -- if you have the list and the results, you can match the calculations perfectly.
Transparency was important for a world system...and the number of events at the beginning was fairly modest...so it is not exactly a flaw. Call it a feature.
All correct (to the best of my knowledge), and one thing to add:
FIDE increased the frequency of their rating list for these very two reasons:
- some players are rapidly improving [or declining], so their official rating (initial rating in your post) no longer reflects their current strength if the last rating list was many months ago [up to six months back in the 1990's], and
- the number of events is no longer as "modest" as it was in the beginning
I agree that the knockout system carries elements of luck, and isn't the best way to determine a world champion. However:
"Khalifman, Kasimdzhanov, and Ponomariov were clearly not the strongest players when they won--not even close."
Also agreed for the two k's [the capital K is reserved for other players], but Ponomariov was at least a top10 player in 2002 when he became world champion - ELO 2744 meant a bit more back then than now, only Kasparov and Kramnik were much (>20 points) higher-rated at the time. Just after the world championship, he finished second in Linares behind Kasparov.
As Pono was a 19-year old rising star at the time, we could compare him (a bit in jest) to Carlsen today: Carlsen is now #1 and still has to become WCh. Ponomariov was world champion and still had to become #1 (or #2 behind Garry) - it's another story that he didn't manage and even dropped below 2700. Maybe somewhat related to his match against Kasparov that never happened?
***
Yes, and the major point is that in a world with 10 or 20 equivalent players (i.e. NOW), you willnever devise a system that teases out small differences to find the best one.
What the system does...is CREATE differences by having one player wear the crown (while the others don't).
Botvinnik was not obviously stronger than others from 1948-1963, but his stature is enhanced greatly by having (and retaking) the crown.
The title process confers status -- it doesn't necessarily discover it.
It is nonsensical to attack a cycle...for not finding the person or player that you already know "should" win. Sports has to have an element of uncertainty.
Otherwise, players in the range of # 11 thru # 50 should simply refuse to play in the world cycle. If they win, they will be derided. If they lose, they were expected to do so. Why should they try...if even WINNING does not validate them.
Poor Carl Schlecter. Even if he had won that last game, everyone would have said Lasker was stronger. No wonder he died in a heatless room, penniless.
Poor Bogolyubow. Even if he had won, everyone would still say (drunk) Alekhine was better.
Poor Euwe. Oh wait, that already happened.
Poor Leko. Even if he had held on to win that "exhibition" match against Kramnik...nobody would feel good about saying "world champion Leko".
Because they already know who should win.
Poor Topalov -- even when he beats Anand...Carlson is already stronger and already world champion.
And so it goes.
If you don't respect the title cycle organized by the entire world (i.e. national federations inside FIDE), then you are really devaluing the players as well because you are denying them any chance at all to win the crown. You have already destroyed their sporting aspirations, by making it a coronation, not a competition.
Does world champion ever weare a crown. Is there picture of crown. I think mabey Anand in picture with crown on head. What about other champion. Not the green leafs but a real crown. It shuold be a big crown with gold and dimonds and rubys and shuold cost bilions of eoros.
chesspride: "...the title cycle organized by the entire world..."
"entire world"! Heh heh..That's a good one :)
I make mistake. Not bilions for crown that is too much. Thosands probly but who can say. Must be very big crown so peoples can see it from far away. Also what happen to green leafs do it get throw away in dump or burn it up. If I ever be world champion I want nice big crown and lot of money but I would give money away so it do not go lost or stolen from me and I then need more so I play better to get more but if I rich I do not care so much any more. Some I spend but most give away. Most I want is crown.
IM Stoopid, you are very clever. Cynical, perhaps, but definitely clever.
Hi Cat,
Just a friendly challenge anyway. I'm hardyberger on playchess.com.
Does world champion ever weare a crown. Is there picture of crown. I think mabey Anand in picture with crown on head. What about other champion. Not the green leafs but a real crown. It shuold be a big crown with gold and dimonds and rubys and shuold cost bilions of eoros.
***
I suggest a crown, fur-trimmed cape and sceptor.
Or a large wrestling-style championship belt. With good entrance music.
Topalov could do it.
PS on a serious note, yes there are pictures from 1985 of Kasparov being given a ceremonial wreath/crown of laurel leaves (if memory serves he either wears it or holds something like it).
Ok , not fully.
chesspride, the problem is that there is nothing of importance that distinguishes the FIDE knockouts from other top-level tournaments. You can call it "WC" if you'd like, but the designation carries little weight after 100 years of match tradition.
Recently, the closest analogues to a crown in professional chess may have been this one:
http://www.chessninja.com/images/nakamura-txapela.jpg
Or this rather similar one:
http://www.chessbase.com/news/2009/events/aronian14.jpg
Or for something slightly different:
http://www.chessbase.com/news/2009/fide/nal13-01.jpg
Ok Hardy, will look you up, though due to current bad internet connection (have to get it fixed) an a lot on my plate it might be quite some time before I get around to it, but eventually it'll happen, practise the line :)
I quite like the idea of a crown, and the champ should be forced to wear it to every game, with his matching royal robes, of course. Also a red carpet leading to his seat would be nice. This would put an end to this "first among equals" nonsense pretty quickly. There could also be some kind of corporal punishment for beating the champ.
Uff Da | January 8, 2010 12:53 AM | Reply
chesspride, the problem is that there is nothing of importance that distinguishes the FIDE knockouts from other top-level tournaments. You can call it "WC" if you'd like, but the designation carries little weight after 100 years of match tradition.
****
Funny -- explain how a match between Anand and Topalov is different than any other match played between players of approx. equal strength.
Oh yeah -- the difference is that they are playing for a title.
Same as in the world championship cycles you deride.
Look -- FIDE was forced into that position by the folks who tried to take the title away. State funding for chess matches was non-existent in the 90s. And...with the advent of computers...chess as loooooong games and loooooog matches had a very short shelf-life.
The KO format saved chess and saved the title from the clutches of the private title folks.
Rather than diminishing it, the KO cycle should be praised.
What should be noted, however, is that the KO cycle is appropriate for picking a challenger.
If FIDE had maintained title continuity by having the KO winner play the current titleholder -- in a match of X games (where x is 6 or 8)...then the system would generally have been perfect.
A KO each year to find a challenger. A title match each year...of short duration. Title continuity.
The "problem" you find with the KO is that the title continuity was lacking. Khalifman, Ponomariov and Khazimhinov may not have been the strongest titleholders...but they are quite satisfactory challengers. Compare and contrast vs. N. Short, B. Gelfand, A. Sokolov, G. Kamsky or others who made it to the quarterfinals or semi-finals in prior cycles.
I expressed this very same point to FIDE in a letter to E. Omuku in the 90s.
However, player opinion at the time was very anti-Karpov and the player sentiment was against having the titleholder sit out and await a match opponent.
I think if the players could reconsider that position...they would agree that title continuity was the better choice for the long-term. This is *precisely* why player opinion cannot be the only input for the cycle.
I visited this page first time to get info on people search and found it Very Good Job of acknowledgment and a marvelous source of info .........Thanks Admin! http://www.reverse-phone-look-up.net
Coach Factory recommended you to buy Coach Kristin Bags, the one which can fully display women's elegance and nobleness.coach factory stores embodied both strength of style and features, as well as what else are you able to request only one bag?
I will share this with my friends. Thanks for your work and all. I bookmarked.
http://book-villas.com/ski-holiday/villas-in-austria