Is chess really a young man’s game?

Diagram Spain
Black to play

Kasparov no longer plays and, having passed 50, if he did play he would just be a has-been, at least if you listen to the words coming out of his own mouth. “Chess is becoming younger” is one of the claims he has made, in-between his disrespect for the Anand – Gelfand World Championship match last year and his suggestions that Anand should retire.

It all sounds very plausible when someone like the greatest player in history says this with the authority and conviction he usually produces. But ask yourself: would Kasparov be a top 10 fixture if he was still playing? Do you think he would be that much worse than Aronian, as an example, if he was still as determined to play chess as he used to be?

Insiders all know that Kasparov is “in love” with Magnus Carlsen and has wanted to see him as the World Champion for a long time. He started the talk of an Aronian – Carlsen match as the only legitimate thing; but personally I prefer the current situation where we have a shaky qualification system, when I think of the alternative of the late 1990s, where the World Champion offered a match for two players, and then gave the loser a World Championship match. Actually, the match that Kramnik won was first turned down by Anand, as far as I know.

What about all those teenage grandmasters; is this not a sign that chess is a young man’s game? In my opinion, no. It shows something about the new tools available today and the availability of tournaments. The old grandmasters with a rating of 2300 playing various “factory style” tournaments also make it easier (not to criticise those, they are a reflection of the norm system we have; my point is simply that they have existed for the last 25 years only).

I would also like to see the average age of new grandmasters. You should of course exclude those given the title, like the senior World Champions (or the runner-up, in one bizarre case), Gaddafi and other oddities. Will the average age really be higher than in 1960? I am not sure. My guess is that there will be just as many old new grandmasters as there are young new grandmasters, balancing the average.

When Anand played Gelfand in Moscow, we not only had the highest average age of a World Champion match for a century, we also have the oldest World Champion since Botvinnik. Botvinnik was born in 1911 and lost the title to Smyslov in 1957. If he did not have the benefit of re-matches, which is now gone, we can assume that he would not have regained the title in 1958 or later for that sake. Anand was born in 1969. If he defends the title in the match against Carlsen, he will have been an older World Champion than Botvinnik; if we compare like for like.

I had a look at the average age of top 10 and top 20. It is 31. But include a non-retired Kasparov in the mix. Top ten would be 33, and top 20 the same, as current number 20 is Giri at age 18.

So who is the oldest player over 2700 today? Boris Gelfand of course! And Happy Birthday to Boris, who is 45 today.

Boris has been written off as an old man since around 1998. He got no invitations to Linares and so on, but continued playing and working. There was no World Championship Cycle for him to participate in, with the exception of Dortmund 2002.

His great “comeback” started in 2007. Since then he has achieved the following: 2nd-3rd in the 2007 World Championship. Gold at Board 1 at the Olympiad, winner of the World Cup, Winner of the Candidates tournament. Losing the World Championship Match by a smaller margin than Topalov and Kramnik, and winning the first leg of the current Grand Prix.

In 2013 he did not do too well in London. He seemed tired in the beginning. But in the end he was 5th, which is no scandal. Then he won the Alekhine Memorial together with Aronian and now the Tal Memorial in Moscow, making him one of only two players to win two of the five strongest tournaments in 2013. The other is of course Carlsen…

If anything, Boris looks stronger to me now than he did five or ten years ago. Still people liked to talk about his challenge in 2012 as being undeserved. It is quite disgraceful actually.

What does this mean for you?

The only thing I see as consistent is that training pays off. Boris solves exercises, analyses endings, prepare openings, follows other top players’ ideas and thinking, and so on.

Basic work – not necessarily hard (which sounds unattractive; something Boris would detest it being called), but lots of it and not shying away from pushing against resistance, be it in calculation exercises or solving opening problems.

If you are getting on a bit, don’t think that you cannot improve. Of course you can. It is a question of consistently putting in a lot of work, improving the areas where you are struggling. With age comes often work, wives and children. These eat time and energy, exposing the real struggle: Finding time, not whether or not you can improve. Of course you can. But it will not be easy…

The diagram at the top

I am currently playing a tournament (don’t call it a comeback; it is a holiday) in Sanxenxo in Spain. The tournament is in a fantastic place, the hotel is lovely and the phone in the room next to ours only rings every fifteen minutes. I highly recommend playing this tournament next year, especially if you like good weather and seafood.

I have 2/2 and yesterday broke the World Record for the quickest checkmate with …h3-h2 by four moves, reducing it from 20 to 16.

Peter Teleman – Jacob Aagaard

Sanxenxo, 23.06.2013

1.e4 c5 2.Nf3 Nc6 3.Nc3 e6 4.Bb5 Nd4 5.Nxd4

5.0–0 a6 6.Bd3 Nc6 7.Re1 g5!! with good play for Black is a personal favourite of mine.

5…cxd4 6.Ne2 Nf6 7.Ng3?

This just looks wrong.

7…a6 8.Ba4 h5!

On the way to great things!

9.0–0 h4

9…Ng4?! 10.d3 Qh4 also looked tempting, but I felt it was too complicated. Indeed after 11.h3 Bd6 12.Qe1! White has good play.

10.Ne2 h3

Simple chess. The weaknesses are permanent.

11.g3 Nxe4 12.Nxd4 Qf6?!

12…b5 13.Bb3 Bb7 is simple and better for Black.


13.c3 Bc5?! was my intention, but now 14.d3! or 14.Qg4! would have turned the tables, giving White a slight edge even.

13…b5 14.Bb3 Bb7 15.Ne1?

15.d4 Bd6 looks great for Black, but had to be tried. We have reached the top diagram.

15…Nxg3! 16.hxg3 h2#

Postscript: I would be very happy if someone wanted to do the work of comparing the average age of the top 10 or 20 historically. Please get in touch if this is you. I want to use this information for Thinking Inside the Box, where I will explain why we have the oldest World Champion for half a century. To me it makes perfect sense.


76 thoughts on “Is chess really a young man’s game?”

  1. Computers changed everything. They changed in the 90’s when they started to be tactically strong, and now they are changing chess again because the propose new strategies that are inexplicable but they work. Look at Aronian games, sometimes I don’t understand his moves…they seem clueless…but they work…and work well. If me…ehm…if Garry had to comeback seriously, he should face these strange strategies…and I don’t know if he could be on the top again. Of course he would be in the top ten…I’m sure he’s still at least as strong as Anand or Kramnik…but Carlsen/Aronian/Caruana are in my opinion stronger than him at the moment.

  2. Who knows .. now that Garry has fled Russia and Russian politics, we might return to playing, tho I doubt it.

    I seen Garry say, that he is no longer able to compete on a world class level in chess, do you think otherwise Jacob?

  3. Obviously Kasparov has been gone for 9 years already and would not be at that level anymore. But had he played the last nine years, he would of course have been top 5 – and better than Caruana. Aronian, Kramnik, Carlsen would all have been his competition. No one else, the way they play at the moment.

    Actually the computers are not an advantage in early development. The strongest player in the world who used computers heavily before becoming a GM is Nakamura. All the others did not. In my opinion they are a disadvantage for the budding talent. Using them too early is unhelpful. For most people this is not important, but for future World Champions, it matters.

  4. I think it obvious that Gelfand was really under estimated by everyone except Anand!!!
    Who was very proud of this last title defense and should be because he knew how well Gelfand was playing!!! Also Gelfand is doing something that I find very exciting and that is play for a win with black! What a concept in the top ten, not that other players don’t do this but he is really making it work bringing bqck the Sicilian in a strong way. Gelfand has really stepped up his game and is a serious threat to anyone regardless of age…

    I was just thinking about this age this and Kasparov…One of the things about Kasparov is he had Karpov to keep him having to be at the very top of his game. Karpov is a brilliant player and I think this made Kasparov have to work even harder because Karpov was always playing so well and obviously the second best player at that time no doubt about it. Some of there matches were very close…Which shows not only how strong Kasparov is but how strong Karpov is too.

    My Question is if Carlsen becomes WC then who is his Karpov, is there really someone as strong today as Karpov was back then. And what about Anand? He is Word Champion for a reason, he plays great match chess and is a fantastic player on top of it. I wouldn’t rule him out. I think he will come prepared as usual.

    And I see comments on Kramnik’s play recently saying well not so good, but wait a minute just before Tal Memorial he played some of the best chess of his life, he played so well, and then to just barley lose must have been heart breaking…I think this more than anything has to do with his Tal play. Besides that all the players in the London Candidates must be very tired…That was a seriously stressful action packed stress filled tense tournament.

    And look at Glefand…He come right back and wins the Tal Memorial. This means age is not as much of a problem as everyone thinks and that Gelfand is really much better than he gets credit for, although now I suppose people will be forced to pay attention to his greatness as a player!

    I am also very impressed with Caruana…I really feel like he has some serious potential and is not afraid to do the work, I have read that he really works hard and I love seeing that pay of for him.

    What is interesting to me is people have said that older players fall off or start to decline…To me I see the younger players playing more sloppy that the “Old guys”. Gelfand is a prime example, he come prepared and his opening play is so pristine and crisp, his play in general and he is consitant which I would think that the same people that think his or anyones age is working against him at this point its seems to be the other way around. Kramnik too in the the Candidates, his openings prep and all around play was truly a thing of beauty.

    I am inspired by Gelfand and am a definite fan now…He is one of the top guys that is making chess very exciting to watch right now, I wish him a happy birthday and congrats on winning the Tal Memorial in fine style!!!

  5. Shurlock Ventriloquist

    Q: Is chess really a young man’s game?

    A: No.

    It is a young PERSON’S game as young women can also excel at chess at levels that the more venerable cannot.

  6. I love that Gelfand won the Tal memorial as a player who was around when Tal was still playing!!!

  7. @Jacob Aagaard
    I’d like to agree with you but unfortunately I think computers are an advantage even if used early in the player development. I grow with classical concepts, the weakness of the d5 square, the bishop’s pair, and so on. Now the top player play whatever it works. If it works it works and no concept can overcome this. Some of the beauty of chess has been lost but we can’t stop the future. Just a hint…Karpov’s play was considered flawless, particularly in the ending. If you analyze his games with today’s engines you find many flaws in his play…his moves were logical and his opponents weren’t illogical enough to find resources…but the computer doesn’t know the logic and refutes in few seconds Karpov’s plan. Today Karpov wouldn’t have a chance against Aronian/Caruana/Nakamura and even Kasparov would suffer.

  8. @garryk

    “Karpov’s play was considered flawless, particularly in the ending” and “Today Karpov wouldn’t have a chance against Aronian/Caruana/Nakamura”.

    I would love to see Karpov (in his prime) against Nakamura in an ending. I think he would have more than a chance.

  9. Jesse Gersenson

    Re Postscript: there is a website with a lot of this information, or something similar…shows correlation of age to playing strength. Perhaps by Sonas (sp) who appears at chessbase.

    (webmaster testing: inserting a FEN position…a board should appear…may clash with gray/blue background of this comment area)
    [fen size="small"]r3kb1r/1b1p1pp1/p3pq2/1p6/4n3/1B4Pp/PPPP1P1P/R1BQNRK1 b - - 0 1[/fen] Black to play

  10. @John Shaw

    In the ending I agree Karpov’s play would be better than most current top-GM…at that time there was still the adjournment… but I’m afraid Karpov (even in his prime) wouldn’t survive the middlegame. Do you remember the famous Miles win with 1 … a6? Karpov (as most of his contemporaries) wasn’t used to play irrational position…and today most games are full of irrational moves. Nakamura/Aronian/Caruana are Miles on steroids, they would destroy any 80’s or 90’s player.

    I say these things with sincere regret for the old days…I like Karpov’s play much more than Nakamura’s…

  11. I wrote a big blogarticle end of 2012 which made a comparison between top 100 players over a decade. One of the aspects discussed in it is the age, see:
    Some interesting conclusions:
    – We can see that the average agedifferences depend very much of which percentile of players are chosen.
    – It is possible even as a middle-aged player to make still progression if one plays a lot and shows a certain level of professionalism.

    Much more can be found in the article. I know some people want me to write in English instead of Dutch on my blog but my prime public is the Belgian/Dutch chessplayers and I don’t have the energy/time to write the articles in 2 languages so hope google translate is ok.

  12. Just a quick query for Jacob.

    In the game continuation: 1.e4 c5 2.Nf3 Nc6 3.Nc3 e6 4.Bb5 Nd4 5.Nxd4? [Isn’t this a well known mistake due to: See note to Black’s sixth move]
    5…cxd4 6.Ne2 Nf6 [Why did you avoid the “Book Refutation” of 6…Qg5!]



  13. Jacob Aagaard :
    Actually the computers are not an advantage in early development. The strongest player in the world who used computers heavily before becoming a GM is Nakamura. All the others did not. In my opinion they are a disadvantage for the budding talent. Using them too early is unhelpful. For most people this is not important, but for future World Champions, it matters.

    If you have no access to trainers, books, competition,… then computers can be a very thankful instrument to improve in chess. I was able thanks to mainly playing on tabletcomputers to raise my level from pure beginner to approximately 2200 elo, see my blogarticle
    2200 is no worldchampionmaterial of course and I do agree that other (non-computer)-methods are likely working faster and better. I only want to stress that working with a computer shouldn’t be disregarded as complete nonsense even if it is for early development. In the end we should also realise that very few of us are worldchampionshipmaterial so other targets (e.g. FM, IM) are probably more realistic to go for.

  14. garryk :@John Shaw
    In the ending I agree Karpov’s play would be better than most current top-GM…at that time there was still the adjournment… but I’m afraid Karpov (even in his prime) wouldn’t survive the middlegame. Do you remember the famous Miles win with 1 … a6? Karpov (as most of his contemporaries) wasn’t used to play irrational position…and today most games are full of irrational moves. Nakamura/Aronian/Caruana are Miles on steroids, they would destroy any 80′s or 90′s player.
    I say these things with sincere regret for the old days…I like Karpov’s play much more than Nakamura’s…

    Had there been a Naka, Aronian or Caruana in Karpov’s time, he would have simply adapted to circumstance, so I can’t agree with your hypothesis.

    Moreover, attempting to compare and contrast players of different era’s with the aim of determining who was stronger is tenuous at best. Perhaps a better way to judge the relative chess strength of these different generations, would be how far ahead of their contemporaries they were, or in the case of the current generation, are.

    Food for thought.


  15. The average age of the new grandmaster I didn’t check as it will take too much time but we can easily check when the youngest grandmasters are born:
    No surprise but almost all the youngest grandmasters are coming from very recent times so I believe it is not strange to state that the average age of the new grandmaster went down. This can also be seen in the statistics in my blogarticle mentioned above. Please don’t compare/confuse the average age of new grandmasters with the top 10/ 20 players in the world as these 2 are very different.

  16. Even if you do the calculations you have to be aware of another aspect of the computer era that I haven’t seen mentioned here. It strikes to me that in the past (i.e. precomputer era), team assisted preparation and chess playing experience were worth enormously for few of the same people to stay up there for very long.
    In today’s play, being an expert in few lines could be even a disadvantage paradoxically. These days, the powers of the “sponsored” knowledge (team of seconds, government support, etc) can easily be fought via some work with our personal engines, which of course, are also “their” personal ones too, the very same adviser. In this regard, current chess is immensely more competitive than before, and I agree with Kaidanov about the issue of absolute and relative strength. Presently, he believes that ratings reflect well the strength of the players, and all of those “new” 2700+ deserve their rating, but at the same time they are not much better than lots of their contemporaries.
    All in all, I strongly believe that whatever results you get from this investigation, you have to be particularly aware that the further you go in the past, the less competitive that absolute chess strength was, and thus I wouldn’t be surprise at all if many talented players could just get away with their natural skills, paired with (as of these days standards) mediocre or even nonexistent preparation, and lots of experience playing a role OTB. Just like with another sports; in Football, it is well known that players from the past got away for many years without very serious training and remained in the high level anyway.
    Another little issue: Could you (Jacob) please explain why in your opinion computer chess assisted training and knowledge acquisition could possibly be prejudicial for one’s formation? I recently read a note that there is a “new” type of school in USA (regular elementary school) that intentionally keep children away from any electronical device (including computers) up until they are somewhat 11 or 12. And the same article explains that the (rich) IT CEO´s and many other people working in technology has taken those as preferred to send their children to, so maybe your argument is along these lines, but this far it remains a little empty to me.

  17. There’s also a German brain scientist who currently has a very popular book in Germany (“Digitale Demenz”) that argues that use of digital media is very detrimental to learning, especially for the young.

    I’ve only read some reviews, but apparently the point is that we learn by doing and that digital media promote learning by just taking in a lot of information, or just copy and pasting it without the brain having to do anything at all. Also the act of writing things down by hand helps memorization more than typing does, related to the motor skills needed for writing. Supported by a lot of brain science.

    There’s also the book “The Shallows” which argues similar things, if I recall correctly.

    Anyway all that translates rather directly to figuring stuff out on your own, without help of Houdini, but with a lot of talking and arguing with friends, and while moving pieces on a board.

  18. @ Remco G: Yes, pretty much. Two references for what I said, could be found in French here and in Spanish here as 2 examples, but the media from many places picked this too. I am sure that something of the sort is applicable to chess as well, but I’d like to see Jacob’s view about that, and also to ask about how he knows that Naka in his formation, used more computer than Caruana or Carlsen (both younger than him). Hard to believe and I guess impossible to prove too.

  19. Jacob Aagaard

    There are many benefits and opportunities for older players today, meaning you will find more older grandmasters as well.

    Obviously, we should also remember that when Fischer became a grandmaster, he was already a world class player, while a new grandmaster today is scraping the top 800 or so. Kasparov at 12 was comparatively stronger than Karjakin at 12, would be my claim.

  20. Jacob Aagaard

    I see that there are people out there who know a lot more about this issue than I do, but I still have a few points to make.

    The first one is that becoming a grandmaster early on, no longer guarantees success later on. Clearly, the title is not what it once was, comparatively. There are a lot of arguments about more chess players, does not mean that 2500 is worse than 2500 twenty years ago, only that there are more strong players. It clearly has some validity; but there is also an effect, meaning that you get more opportunities. Kramnik was a FM at 17, as I recall. He lived in a place called the Soviet Union and his rating was 2595. So he was not a young GM. Lack of opportunities.

    The reason why learning with a computer early on could be harmful, is because it means you think in a different way. Somehow you are a bit less active. I believe it could reduce your overall potential.

    The test of this is to check the longevity of the players who are still eager and working hard, compared to those who came with the computer (Nakamura and Karjakin are the prime suspects of computer learning). Then you have to add in all sorts of other aspects as well, and in the end you end with an argument based on beliefs to some extent. However, I don’t think it is a coincidence that we have some very strong players around 40 just now. They got the best from the pre-computer era and the best subsequently.

    Regarding Karpov and Nakamura. This is sort of amusing. Nakamura has worked a lot on his endings the last 1-2 years, I think. You can see it in the way his endgame play has improved massively. Sure, if you want to create an experiment where Karpov has not worked with a computer (which he never learned to do) and Nakamura has, then maybe it is 50-50. But if you take a more relevant experiment, where Karpov gets the computer at 25 and Nakamura has had it since he was 10, then my money is on the Champion of old.

  21. @Jacob Aagaard
    I do follow most of your thoughts but a few things need to be added.
    1) I agree that learning with a computer early on could be harmful but if we have to make the choice between the computer or getting no feedback at all from the played games then I vote for the computer.
    2) Comparing people between different era’s has little to no sense. I remember an interview of Jan Timman on chessvibes: where he made the following very important remark: ‘If I had been 18 today, I would not have become a professional player.’ We are all a product of our time so it is simply wrong to make conclusions between different playing conditions.

  22. @ Jacob Aagaard: Which Olympiad saw Boris Gelfand get the gold medal for board 1? I checked the 2008, 2010 and 2012 Olympiad entries on Wikipedia and Gelfand isn’t listed as gold medal winner for his board.

  23. Jacob Aagaard

    Boris told me he had the most gold medals from Olympiads, but then realised that there were people 40-50 years ago that did at least as well as he did.

    But I remembered 2008 as a gold medal for Boris. I can now see that he was narrowly edged out by Peter Leko. There is a strange thing going on in my head, where I psychologically trust my memory more than the facts on the internet.

    Summa summarum: I might have mixed things together here and overstated his brilliant results.

    What I do know is that after he had access to the World Championship system again and did not have to get into the best tournaments by invitation only, he started proving his worth again.

    An important point here is that many of the top players are very good at beating Van Wely and other 26-a-lot players with high frequency. Gelfand a bit less so. So, when he play those tournaments where you have to make +4 or +5 to win rating, usually he doesn’t. But when he plays tournaments like this year, where all the players are top class and +3 wins, he is doing really really well. Does this make him a less great player? No. It just means he is lower rated when he does not get a chance to play the top guys. When he does, his rating climbs…

  24. Jacob Aagaard :
    No. It just means he is lower rated when he does not get a chance to play the top guys. When he does, his rating climbs…

    Rating is purely a reflection of the results done in the past. Ratings don’t lie so we can’t state that a rating is too low or too high. However as ratings are used to predict future results, they are very often inaccurate.
    Just think of the fictive example that a players rating is only based on his white games. Obviously when the player would switch to play only with black then I wouldn’t be surprised if the rating drops with 50 points.

  25. @brabo

    I look at ratings as an indicator of a player’s strength, not some absolute measure of what player X has done in the past, so in that regard ratings lie quite often. Playing against a young, up and coming kid or teenager isn’t the same as playing someone older who has been hovering around the same rating for a lot of years.

  26. Just in case you still need it for further investigation: contains all rating lists from 1969 until 2012

  27. Jacob Aagaard

    The question to what you say is, how come elo came in at no. 17 in a competition for making the best rating system? (unless my memory is flawed again). Clearly it is not flawless. Still it is the official list and I am checking out as frequently as most other fans…

  28. Jacob Aagaard

    Thanks. I got a few e-mails from “Christian” with offers to help, but my replies keep bouncing. Will get in touch with him somehow when I return.

    So far three wins and a draw with black. Today GM Paunovic awaits.

  29. @Jacob Aagaard
    I am sure some participants of the competition for making the best rating system, are better placed to react on your question but what I understood from following the discussions is that the today used elo-system gives not the best predictions as we are using a too simplified calculating method. This is understandable because when elo was introduced, there were no strong computers available as today. With todays computers we are able to use more complex and refined calculating methods. There are votes going up today to give up the old calculating method of elo and switch to one of the top accurate calculating methods. However there are also a lot of people, preferring to keep the less accurate elo calculating method as it has the benefit of being simple (no need for a computer) and avoids updating the whole chessworld about a new complex calculating method.

    I also want to tell that the calculating methods discussed in the competition, won’t solve Gelfands case. Anyway we are digressing from the original topic and that is age and not the elo calculating method. Still I am willing to discuss further the side-topic if interested.

  30. Also, the point was that Gelfand is a player that, compared to his peers, does relatively worse against Van Wely and players of that level and better against top players.

    Such differences between players exist and it will mean that Gelfand will gain rating points in tournaments with top players only and lose them in tournaments with a few top players and a few Van Welys (that would be my favourite type of tournament, he rules).

    Regardless of the rating system.

  31. Interesting post, Jacob, and as a 52 year old I am glad you wrote it. However, you seem to be ignoring recent discoveries in neuroscience:

    1. Myelination of neurons in the brain is the key to developing a skill. This was demonstrated by, a U. of Oxford brain-imaging study of adults who learned to juggle. See for details. So the ability of the brain to myelinate the neural circuits related to chess patterns is the dominant factor in developing chess skill. I.e., the more you practice, they more your brain will myelinate the “chess circuits.”

    2. The myelination process typically reaches its lifetime maximum at age 39, and starts to deteriorate at that point. In fact, myelin is constantly breaking down and being repaired, but at age 39 or so the repair process is no longer able to keep pace. And as we age, the repair process lags further and further.

    This has been demonstrated through a very clever experiment involving finger-tapping. A key quote from a popular write-up explains:

    “Studies have shown us that as we age, myelin breakdown and repair is continually occurring over the brain’s entire ‘neural network,'” said Bartzokis, who is also a member of UCLA’s Ahmanson–Lovelace Brain Mapping Center and the UCLA Laboratory of Neuro Imaging. “But in older age, we begin losing the repair battle. That means the average performance of the networks gradually declines with age at an accelerating rate.”

    “Significantly, the research suggests that the myelin breakdown process should also reduce all other brain functions for which performance speed is dependent on higher AP frequencies, including memory; it also supports the suggestion that myelin breakdown is a biological process of aging underlying the erosion of physical skills and cognitive decline, including the onset of such age-driven disorders as Alzheimer’s disease.”

    So I think Kasparov was on to something. At the same time, I would agree that:

    1. An adult who has not reached his/her peak in chess ability (i.e., 99% of the readers of this blog) can certainly increase their chess ability even as they move into the senior years. Myelination is slower, but it doesn’t stop altogether.

    2. A hard-working 45 year old and maybe even a 50 year old can stay very close to his/her peak ability, which for someone like Kasparov would make him dangerous today. And it is what makes Gelfand dangerous, too.

  32. Hasn’t the age thing been proven? Youth doesn’t equate to success. There was a match in the late 90s, I believe played in Parsippany, New Jersey in the United States, between someone who was then 67, I don’t recall who, and Irina Krush when she was I want to say 14. They drew the first and last games, and split the middle two, tying the match at 2-2 between two players 53 years apart.

    Nuff said!

  33. LOL, I spent about 20 minutes trying to find the name. I never could find the match, but I remembered he did analysis at tournaments for the CCA in the early 2000’s, and yes, it was Arthur Bisguier.

    The Bisguier/Krush match, 4 games, tied 2-2, shows that age doesn’t equate to failure!

  34. Patrick, I am not sure what the “age thing” is that you think has been proven. I would certainly expect an experienced GM age 67 to retain enough ability to draw a match against a 14 year old rated 2275. I would not expect the GM to have the same ability he had 30 years previously. In fact, Bisguier’s rating had already dropped from ~2500 to 2350 at the time of the match. His rating has since dropped to 2200. So the history of Art Bisguier very strongly supports what I have said.

    If you’re saying that an older player can still have fun and play strongly, I agree 100% with you. If you’re saying that he or she can play at the age of 70 with the same strength s/he had as a 30 year old GM, I’m afraid that the science and the ratings graphs do not support such a notion.

  35. Saying that Karpov or any other great from the past would have no chance because a computer finds non-perfect play in their games is close to nonsense; today’s top players don’t play with a computer in their pocket (hope so! :D). Remember how Caruana was well beaten with the white pieces by an already very veteran Korchnoi a couple of years ago in a Ruy Lopez for example.

    Same computers will find a lot of “bad” play in today’s top players as they find in Capablanca’s, Kasparov’s, Karpov’s games, etc.

  36. @Chris Falter
    The curve age/rating which I use in my blogarticle fully corresponds to what you are stating so limiting the peak at 50 for a player being always reasonable active.

    I also notice very often around me that when chessamateurs retire from their profession at age 60-65 that the 1-2 years after the retirement these people play much more chess and manage to give a new boost to their rating. However once they reach age 70 this retiring effect has been eliminated again by the aging effect.

  37. “Postscript: I would be very happy if someone wanted to do the work of comparing the average age of the top 10 or 20 historically. Please get in touch if this is you. I want to use this information for Thinking Inside the Box, where I will explain why we have the oldest World Champion for half a century. To me it makes perfect sense.”

    Please email me using the email I left as a part of this comment. I have gone ahead and compiled the 2013 results (which range 29-32) for Top 10, 15, 20. I will happily share this file via email with you if you get in touch with me.

  38. @Matt
    I didn’t intend this. I said that
    – Karpov’s play was considered flawless while computer assisted analysis find many improvements for his opponents that they didn’t find because they weren’t used to “think out of the box”
    – Computer showed that moves considered illogical can be played because they work and they bring the game in a direction where both players feel somehow “lost” and so both have difficult decision to make and hopefully the best player wins
    – Karpov (and his generations) weren’t used to this kind of chess, so even if they would try to avoid irrational position, most of the time they wouldn’t succeed
    – In those irrational position a today’s top-GM (Nakamura, Caruana and others) would be in great advantage against Karpov and his contemporaries. Of course Karpov would be the favorite in the endgame but he probably wouldn’t survive till that.

    Having said that I add
    – Caruana may have lost to Korchnoi but it was a single game, Caruana was very very young and Korchnoi is the exception as he was an irrational player 20 years in advance.
    – Anand’s crisis is in my opinion due to this fact. He hasn’t forgot how to play chess but he hasn’t accustomed to this new type of – let me call it in this way – irrational chess.

  39. @garryk
    To be honest I find these ‘comparisons’ pretty pointless. Who was more brilliant, Newton or Einstein? You could also argue that if someone like Morphy would have lived in this computer age he would have wiped away all his opponents including Nakamura. How do you proof these things? Obviously, if I could somehow time travel to the stone age I could easily kill the strongest Neanderthal person with my 21st century weapon, but I maintain this is not a fair comparison.

  40. @Ray
    In fact I don’t want to compare Karpov to Nakamura, of course it’s pointless. I’d like to compare Karpov’s style of play (that was typical for his period) with today’s style of play, that is highly influenced by computer assisted analysis. Of course I like Karpov’s play much more, if only because it’s logical and beautiful, but my point is that logic unfortunately doesn’t stand against the irrational but effective computer suggestions.

  41. I think there is a problem with this way of viewing it. Chess logic was never static. Neither was our understanding of the physical universe or any other complex subject. So, with the computers, obviously new discoveries were made; mainly regarding dynamics. It is a constant development. Put Nakamura in the 1980s, without those tools, and he would play like those guys as well.

    To me the interesting thing is that some players managed to absorb both ways of understanding chess, and they are still there at the top. Obviously new talents have come along, but look at the 1990s; not a lot of older guys were in the top 10 then.

  42. Regarding the statistic project. I will get in touch with those interested in participating and we will do it as a team. I am back to work next Thursday.

  43. Jacob Aagaard :
    Obviously new talents have come along, but look at the 1990s; not a lot of older guys were in the top 10 then.

    With statistics you can proof anything. As I mention in my blogarticle, if we look to the top 10 then the average age even raised in the last 10 years !! However if we look to the top 100 players then the average age dropped. You really need to look very critical to the numbers before making any conclusions about age. A few possible but not proofed explanations:
    1) Changes in the average age could happen for the top 10 in a delayed mode than for the top 100. Logical if you think that reaching the top 10 takes more time than the top 100 for a young talented player.
    2) Computers can maybe bring new talented players faster to a level of 2700 compared with before but beyond that bar it isn’t case.
    3) Today we maybe have much more players picking up chess than before. I am thinking mainly of the explosion of chess in Asia. Breaking in new markets leads to a temporarily drop of age. I say temporarily but it can easily be decades before the average age is back on the old level.
    So many elements are influencing the average age (i am even not talking about coincidence which certainly also plays a role when looking to e.g. only the top 10). In the end I believe, it is better to look to rating than ranking if we want to understand the effect of age. In my blogarticle I also showed some age/rating figures which despite the small samples give some interesting insights.

  44. Nice game! I played in Haarlem – Netherlands in the Nova tournament this weekend. Did very well with 4 out of 6, two draws against an IM and a GM, and no losses. Also played a short attacking game with a very beautiful last move:

    Indra Polak – Laurens van Twillert
    1. e4 e5 2. Nf3 Nf6 3. Nxe5 d6 4. Nf3 Nxe4 5. d4 d5 6. Bd3 Nc6 7. O-O Be7 8.
    Re1 Bg4 9. c4 Nf6 10. cxd5 Nxd5 11. Nc3 Nxc3? (not good) 12. bxc3 O-O 13. Rb1 Na5 14. Rb5 b6 15. h3 Bd7 16. Rh5 g6 17. Ng5! Be8(? but white is better in all lines) 18. Nxh7 gxh5 19. Nf6+!! 1-0 It is a forced mate now, the most beautiful line runs 19. … Kg7 20.Qxh5 Rh8 21. Qg5+ Kf8 22.Qh6+! Rxh6 23.Bxh6#. Note that after 19.Qxh5? black would turn the tables with 19. … f5!, being the point of blacks 17th.

  45. I personally think that at 47 I’m probably close to a breakthrough in my playing strength and certainly think I’m playing my best chess in 15 years if not ever.

    The decisive factor, especially for younger players, but not only younger players is access and exposure to better learning material.

    Not every player press conference at a top level tournament is good but there have been a surprising amount of astonishingly instructive ones. Not to mention great commentary. These teach you big picture lessons not available in the literature.

    A keen young chess player has access to all the material and online practice they could ever need to advance quickly if they want and even older players such as myself can suddenly have a breakthrough.

    I’ve also learned a lot from the exercises in Aagaard’s new series amongst other new excellent books.

    Below a certain level I think players can improve very significantly. Replacing unhelpful ingrained thought processes is the biggest barrier.

    Older players such as Gelfand and Anand have almost bullet proof opening repertoires which take years to develop. But top level chess is different to lower levels.

    Right now Gelfand is just playing well. Good openings matched by a new confidence in the rest of his chess. I’m not sure Anand has such confidence over the board. I do think however that if you’re an older elite player physical fitness is becoming increasingly important.

  46. @Mark Crowther
    I think it was Nunn in one of his books who said that amateurs usually haven’t realised their full potential, so that leaves much room for improvement even at an older age. For pro’s it’s more difficult I guess because they have already realised a lot of their potential.

  47. Jacob Aagaard

    Obviously the scope for improvement increases the less you have worked on chess! But the point is that it is still there, even for people in their 40’s and 50’s…

  48. Soren Pedersen

    I have made these quick calculation for the last 63 years. Since 2000 top 10 has become older !
    Regards Soren Pedersen, Danmark

    Top 10 1 jan 1950
    (Botvinnik, Smyslov, Bronstein, Kotov, Fine, Najdorf, Ståhlberg, Boleslavsky, Keres, Reshevsky)
    Average: 34,9 year

    Top 10 1 jan 1960
    (Tal, Keres, Petrosian, Spassky, Smyslov, Botvinnik, Gligoric, Geller, Korchnoi, Taimanov) (Fischer nr. 13)
    Average: 34,3 year

    Top 10 1 jan 1970
    (Fischer, Korchnoi, Spassky, Polugaevsky, Petrosian, Smyslov, Stein, Larsen, Portisch, Geller)
    Average: 37,3 year

    Top 10 1 jan 1980
    (Karpov, Korchnoi, Tal, Portisch, Polugaevsky, Spassky, Kasparov, Timman, Larsen, Petrosian)
    Average: 39,3 year

    Top 10 1 jan 1990
    (Kasparov, Karpov, Ivanchuk, Timman, Salov, Short, Beliavsky, Ehlvest, Gurevich, Korchnoi)
    Average: 32,8 year

    Top 10 1 jan 2000
    (Kasparov, Anand, Kramnik, Shirov, Morozevich, Ivanchuk, Leko, Bareev, Adams, Topalov)
    Average: 27,9 year

    Top 10 1 jan 2010
    (Carlsen, Topalov, Anand, Kramnik, Aronian, Gelfand, Gashimov, Ivanchuk, Wang, Svidler)
    Average: 28,6 year

    Top 10 1 july 2013
    (Carlsen, Aronian, Caruana, Kramnik, Grischuk, Karjakin, Nakamura, Anand, Gelfand, Topalov)
    Average: 31,9 year

    Source: and

  49. @Soren Pedersen uses a different formula than ELO, although it can certainly be useful. Still, those lists aren’t official and the FIDE didn’t adopt the ELO system until 1970. The lists prior to 1970 are estimates.

  50. Jacob Aagaard

    I still think the end result is reasonable. But of course, it would have to be done for every year to make a statistically valid conclusion.

  51. Jacob Aagaard

    At the same time I would like to apologise to those who wrote to me about this while I was away. I have been under a big log of work ever since returning and it is 26 degrees here in Glasgow, which makes it like an oven…

  52. Jacob Aagaard

    What I see from that list is that age is less of a factor, than the fact that good players tend to be good players for several decades. Larsen would be on the list in the early 1960’s and falls out in the early 1980’s. He was born in the mid 1930’s. Kasparov would be on the list today, had he not stopped. He was on the list in 1980.

    As with all things random; the great players tend to come in clusters rather than spread out evenly. Around 1990-1994 we had the maturity of a great generation, with Kramnik, Shirov, Anand, Ivanchuk, Gelfand and a few others. These guys are still around. We will no doubt see that Karjakin, Carlsen and Caruana will be in the top ten (maybe one of them in and out, like Ivanchuk and Gelfand) in twenty years time.

    Kramnik was young when he reached number one, Kasparov was, Karpov was, Tal was, and so on. The fact that they were great players is the consistent thing. Some players made the final leap later in life, but usually those who singlemindedly go for the top have this tendency early on.

    So, I do not think that chess is getting younger, we are just getting better at recording the early progress. To me, the biggest successes by 12-year olds is winning the u18 Soviet Championship: Kasparov and Kamsky did this. Kamsky was already world class at 15. Unfortunately the support system around him was not good, for life or for chess purposes…

  53. @Jacob Aagaard

    I agree. Most historically great players have had long careers at or near the top. Maybe the only two exceptions are Morphy who played in a completely different time and Fischer, but this also adds to their mystique. Someone like Kramnik has been going strong for 20 years now and I don’t believe people consider him “old”.

  54. Soren Pedersen

    Top 10 is very accidental. There could be a 60 yr just outside the 10, and a junior just inside, and this has a big influence on the average with 4-5 yrs.
    I have made calculations for top 50 from the latest FIDE list, and the list for 13 yrs ago. The first list, where FIDE keep top 100 list on there homepage. I must say for these calculations, the average is so close, that it is the same.

    Top 10 2013 32,6 in 2013 (29,3 in 2000)
    11-20 32,6 (28,4)
    21-30 27,8 (32,2)
    31-40 27,6 (34,8)
    41-50 32,8 (33,8)
    1-50 32,8 (32,2)

    The top 20 is about 3 yrs older today, but from 21-50 2,5 younger. This gives no clear conclusion, and depends on some good years with many World Class-players. As f.x 1990 which maybe the best ever. Players like Carlsen, Karjakin, Vachier-Lagrave, Andreikin etc.

  55. Soren Pedersen

    I have made the calculations again, and there are a few mistakes for the latest rating in 2013.
    The finally result is that in2013 the top 50 was 30,5 in average, and still 32,4 in 2000. So the top seems slightly younger. Sorry about miscalculations.

  56. @Soren Pedersen
    But when we do it in four years, we will find them to be older. Why? Because of Carlsen, Caruana, Karjakin and so on will get older. Remember Carlsen was one day at no. 1 at the age of 17. Top 10 was definitely younger a few years ago: it is all the same guys with the exception of Caruana!

    I think the evidence that chess is becoming younger is absent; while the same can probably be said about my theory of how you learn chess.

  57. @Jacob Aagaard
    Over 4 years new young stars are knocking on the door of which we haven’t heard today yet and this makes that today we can’t make any serious claim if the age of the top x will go up or down.

  58. If the relationship between age and skill follows its usual course, 4 years from now several of the current top 10 will be gone for good from the list. Almost certainly Anand and Gelfand will be gone, and perhaps Topy and Vlday as well. This isn’t to say that as members of the top 25 or top 50 they won’t be able to work hard, enjoy the game, and occasionally score the outstanding result. And then in a decade they will perhaps be in the top 100….

    Just look at Anand’s results over the past 2 – 3 years. Clearly he is not playing at the same level as previously–in a consistent fashion, anyway. I seriously doubt that it is due to a lack of hard work.

  59. If Michael Adams keeps playing like he does in Dortmund, it won’t be long before he is back in the Top 10 🙂

  60. @croflash
    A game or two. In the most recent New in Chess Gelfand says that he knows of several younger players who have reignited their dreams of becoming great players, after seeing him being so successful over the last few years.

    I do think that stability goes out the window at some point, but looking around the elite at the moment, it seems to be 30 minutes into the game :-).

  61. The idea that Kramnik is not in the top 10 in five years (if he still wants to be) is ludicris in my eyes; but I guess only time will tell…

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top