Chess960/Fischer Random

My friend GM Csaba Horvath once had dinner with Fischer in Budapest, after meeting him in the street. At some point during the dinner, Fischer asked him, “What do you think of ‘my’ chess?” (meaning 960). Csaba said: “I like it, but…” Fischer caught him off with a big grin. “No! No, buts!”

I received the friendly email below after my appearance on the Perpetual Chess Podcast.

I know he asks for my opinion, but I am more interested in other’s opinions. And I have also made this commitment not to defend any opinions in debates since taking up a post in FIDE. For obvious mental health related reasons :-).

I just listened to you on the perpetual chess podcast.

I really enjoyed it and realise that you’re experience in the chess world is vastly superior to mine and with that your perspective is greatly appreciated.

My intention is honest and sincere. I’m not about to post your reply on a forum and anything like that. I’m kinda wishing for the future of chess to be chess960 and was interested to hear your recent comment in the perpetual chess podcast. Genuinely interested, im too stoic to get upset by differing opinions but also believe these views and discussions to be extremely important to the future of chess.

I am extremely interested in getting your perspective on something that you mentioned regarding chess960 as ‘the Fischer random circus’.

During the podcast you also mentioned your preference for classical chess due to the ‘deep thinking’ aspect.

I have interpretted these 2 snippets as on the one hand you’re pro deep thinking yet anti chess960 (compared to classical chess). I hope I haven’t completely misrepresented your beliefs here, I’m just making conclusions following 90 minutes of listening.

So I guess my question is really:

Why do you on the 1 hand love thinking deeply and on the other regard chess960 as a ‘circus’. What am I missing that you are seeing?

I personally love chess960 because to me it is an excellent opportunity each and every game from move 1 to think hard with no auto-piloting in the opening. I mean you can’t just be a d4 player in chess960 you’d at least need to make an assessment first.

I actually find that I think deeper (in chess960) from move 1. To me it’s the deep thinking of chess that I love. In standard I have won many a game thanks to london system solely due to my opponent not being as familiar with it as me. But in chess960 I don’t get these opening edges (and vice versa) – to me that’s more pure chess than memorised lines bring to the board.

To me being given more variety of opening positions leads to more varied positions to assess which leads to more deep thinking.

But… classical/standard chess is sooo much more popular so I’m in the minority. So there must be something I’m missing. I think chess960 is like jazz whereas standard chess is more like classical music. Chalk and cheese as far as prep’s concerned.

The thing that I see commonly mentioned are ‘unbalanced’ starting positions as a reason against chess960. But as long as each player plays same position as white and black what’s the big deal? Or if that’s impractical then to be honest, it’s rather obvious that even at top level e.g. Carlsen v caruana that even a computers 2+ edge is often not even realised. So until humans improve considerably I don’t think these computer assessments on unfamiliar positions really impact the human results much. I think the human aspect would likely be the difference.

To me this is a matter of taste entirely. I also don’t like jazz too much. It is too rich and there is an information-overload happening to me when I listen to it. Which I do on rare occasions.

I like the tradition of classical chess. I like that I begin the game in the same position as Kasparov did. I like the patterns of openings.

All of Jason’s arguments are valid. To me it is a matter of taste. What do you guys think?

55 thoughts on “Chess960/Fischer Random”

  1. I don’t see the music world struggling to decide which (jazz or classical) is the future of music. It’s not as if we have to eliminate one at some point. In fact, we probably *need* both, because there will be always be those who chafe at classicism and want more spontaneity (“improvisation”) and others who find greater depth in something composed over months or years rather than moments. (I don’t know whether I’m talking about music or chess any more, but perhaps it doesn’t matter.)

  2. Well, I happen to prefer classical music to jazz …..
    I also like the psychological aspect of theoretical battles: can I reach a position I like and my opponent doesn’t? Yesterday already at move 6 I had a position neither my opponent nor I was familiar with – quite amusing. The game was a quick draw, btw, probably because we both felt uncomfortable.

  3. In checkers (draughts in the UK), there is a variant called 11 man, in which one random piece is removed on each side, and then one random piece is moved on each side, before play begins. This is widely considered by aficionados to be *the* test of “over-the-board” (non-memorized) playing ability. I wouldn’t be surprised if the same were true with 960, if it were to be tested out by strong GMs.

    Other than that, I think it’s a matter of taste. But I don’t think classical chess will be replaced. Even in the checkers world, where the first 3 moves are often chosen by random ballot (call it checkers 156), 11 man (checkers 2500) has still not really caught on. Geeks will be geeks, and that means a taste for booking up.

  4. In the interview I mentioned two ideas. One is to have a spin on a computer between 100s of starting positions (Kramnik). The other is to have a year with 1.a3 a6 added. And so on.

    But I just don’t think we are anywhere this desperate point.

  5. I love the familiarity of the standard opening position, the way you can know your way around a bit and try to get the type of game you’re looking for. Memories of previous games in some given line. Every chess club that has the guy that always plays strange sidelines, the youth talent memorizing Najdorf lines, the maverick specializing in dodgy gambits, the stodgy types who play the London and wait for you to to make a mistake.

    You can see people’s personalities in their openings. Openings and their history are at the core of chess culture. We can’t play like the grandmasters we see in the big tournaments, but we can play the first moves of the same openings and feel a connection.

    960 may be the greatest game ever for all I know, but it can’t replace that.

  6. It is a matter of taste, I do much prefer chess960.
    But really if the idea is creativity from move 1 then I think 1 way to improve chess 960 would be to randomly select each white and blacks starting positions independently. So 960×960 chess. Bye bye opening theory and in the long run those with the better chess understanding, calculation ability, patience and thought process as at time of playing game I expect would reign supreme in the long run.
    I think the above variant (960×960) would work rather well as its still same pieces in the 8×8 world so doesn’t increase randomness at the cost of completely changing the game of chess.

  7. Jacob Aagaard :
    In the interview I mentioned two ideas. One is to have a spin on a computer between 100s of starting positions (Kramnik).

    To take this to an extreme, you could start the game with only pawns on the board. Then the players could take turns placing any piece or king onto their home rows; after the home rows are filled, play commences. If my calculations are correct, this would be chess25,401,600!

  8. I don’t think that classical chess is in a crisis, there should only be some more expertise in the organizers of top events. To play a top-tournament with the same faces over and over again would be boring in every kind of sports. Ali vs. Frazier was fine, but not if they had been boxing every year.
    It’s up to the organizers to find an interesting field for their tournaments.

    960 is an alternative game, why not? Play it, promote it but don’t think about chnaging classical chess. It’s simply something different and might be intersting to many.

    I like the Classic/Jazz comparison above a lot.
    I personally love to hear some Jazz – not all of it. And I love to hear some Classic – also not all of it. Keep an open mind.

  9. One aspect in that discussion rarely gets mentioned: I don’t buy that chess960 would eliminate opening preparation. The reason no one prepares for 960-openings is that there are few serious tournaments played in that format. Would 960 become the norm, grandmasters obviously would start memorizing tons of opening lines for all possible starting positions, since knowing just a bit more in any given position would give you a huge advantage. Wait for those “How to destroy your opponent with white in positions 400-499”-books to be published…
    Opening preparation would change for sure (preparing for individual players would make no sense), but for serious players it might in fact become a bigger rather than a lesser factor…

  10. One person suggested somewhere (no idea where) adopting one position of the 960 to be THE alternate starting position for two years. This way theory could develop. I think this is a very interesting idea.

  11. I think it is a cased of the “I can’t be ar$ed to do opening prep, so lets eliminate the need”.

    I think there is a trend in all sports of spheres towards the lazy wanting to “level the playing field” against those who are prepared to put in the work. e.g. T20 vs Test Cricket.

    Not in favour of 960 at all – see it as yet another chess variant not worth following.

  12. Ben :
    I don’t buy that chess960 would eliminate opening preparation.

    That’s where chess25,401,600 comes in! 😉

    Seriously, though, you may be right.

    Checkers has a precedent here as well…

    The game started to get very drawish in high-level events over 100 years ago, so they decided to shake things up by choosing each player’s first move by a random ballot (and play games in sets of 2, since the games could be unbalanced). They banned a few of the opening positions as being too unbalanced, but kept 43 of them. Not long after, people were booking up on all 43 openings, and books were being written about them.

    Then this variant (called the “2-move restriction” style) started to become too drawish, so they started choosing the first 3 moves of the game by ballot. You guessed it, people started booking up on all these opening positions (they’re currently up to 156). Some of these are so unbalanced that eking out a draw with the bad side of a critical ballot is considered a moral victory.

    Then a GM invented the 11 man variant with 2,500 possible opening positions. They don’t even know for sure which of these are unsound (a draw with best play), so they are allowed to pass on an opening that both players consider unsound. They haven’t started booking up on this variant so far, but that’s probably only because checkers as a tournament…

  13. > They don’t even know for sure which of these are unsound (a draw with best play)

    Should be “They don’t even know for sure which of these are sound (a draw with best play)”.

  14. They haven’t started booking up on this variant so far, but that’s probably only because checkers as a tournament…

    game has been on its deathbed for years, so there’s not enough money or glory in it.

  15. Well why bother with opening positions? Just randomize the way pieces may move as well?
    With 64 squares to go to and 16 pieces to start with, mathematically you can go a long way before you repeat yourself in that way…just pick any of the trazillion possible ways pieces might move, capture, jump etc. Or add mines, swamps, hurricanes etc. into the mix?

  16. Maybe it’s just something fun or interesting. What’s the value in trashing something that some people like? I don’t like cauliflower (not even a little), but I don’t argue to have it eradicated from the earth. And on the laziness thing: It actually takes a certain energy and adventurousness to try something new. So some of the people who *don’t* play 960 (including me) might actually be lazy in a way.

  17. An Ordinary Chessplayer

    There is always some inertial resistance to adopting new rules, unless you are Italian. So I take as a given that players brought up on classical chess will “prefer” it, and a new variant will have a difficult time gaining traction. My thought experiment is to imagine that Chess960 has been played for 100s (1000s?) of years and someone decided to advocate starting all games with one (best?) initial position. I think the suggested “improvement” would be mostly ignored…

    Of all the previous comments, Ben’s is the one that resonates with me. There has been a market for chess theory since the printing press was invented, both with the old rules and the new rules (la dama furioso). Chess960 wouldn’t change that.

  18. @Benjamin Fitch
    I agree, I was just reacting to people who present Chess960 as a solution to the alleged ‘draw death’ of classical chess. Imo one should like or dislike Chess960 in its own right, rather than presenting it as a solution to a non-existing problem.

  19. @Ray
    We just suffered through a World Championship match with 12 draws in classical chess. There were 3 Rossolimos, 2 QGDs, 2 Petrovs and 2 English Four Knights. The “excitement” came from 3 repetitions of the Sveshnikov.

    Maybe there’s something to this “draw death” thing?

  20. @The Lurker
    If every game had been drawn after remaining at or close to a 0.00 evaluation throughout, you might have a point. But in reality, there were many exciting games and missed opportunities for both players, which is the opposite of what you would expect if the “draw death” argument was true.

  21. @Andrew Greet
    A draw is a draw is a draw. Draws are a letdown. That’s why football (real football, not soccer ;-)) playoff games and Superbowls cannot end in draws; the people with all the money know that the spectators want a win-loss result.

  22. Comparing chess to American football, there may be more draws in chess but there are far fewer injuries. (I don’t have the actual statistics…) And there are many more women chess players than there are women players of American football. So advertisers should give chess more consideration.

  23. An Ordinary Chessplayer

    @kaimano – Perhaps I did not express my idea in the most exact way. But I think we can agree that Italians do not cling to the old rules! Historically this has been fortunate for chess. The Italians can rightly claim la Dama furioso. Isn’t it likely they also created castling, the Italian bishop and the namesake opening it made possible (before there was only Bf1-d3), promotion to other than a queen? These are the innovations universally adopted. How many other innovations remained in Italy? There is free castling, known in El Greco’s time and still played in the last century (perhaps played even today, for all I know). So the Italians are (or were) quite good at inventing new rules, are you suggesting these are meant only for others to follow?

  24. The starting position is the main basis for rejecting chess 960. Most positions that you get from there are ugly, and anti natural. Recently, there was a championship between Naka and Carlsen, considered (at least by the organisers) as the two best players of 960. The games were worst than their usual, because most of the job consists in getting a normal, playable, harmonic position; there is no opening, just reorganisation of badly coordinated pieces, the game is slower and basically dull.

  25. I’m not surprised that the opening play of Carlsen and Naka was weaker in chess960. They didn’t have reams of computer analysis and hundreds of years of practice to back them up. Personally I find top player relaying a long ‘interesting’ variation of home analysis ending in a draw far less interesting than a slower build-up where the players are on their own. In the later case you are more likely to get a game with a good fight.

  26. I agree with you, it is nice to have good fighting games; the thing is that the chess960 matchgave very little fights, and very few interesting games -compared to a “normal”match.

  27. @Benjamin Fitch
    I’ll bet more women watch football than chess.

    @Thomas
    Why 16 or 18 games? Because 12 games resulted in a draw?

    In the Carlson-Nakamura chess960 match, of the first 8 games (which had more classical time control), 5 out of 8 games had a decisive result (not drawn). And the match, if stopped there, would also have had a decisive result, as Carlson had 3 wins to Nakamura’s 2 at that point. Imagine if classical matches were that decisive.

  28. Have any of the players slating 960 actually played it?
    My club ran a tournament, 45 minutes each (iirc) at the end of last season. I did not go into it with any expectations but ended up having a great time.

    Everyone’s consensus was that it genuinely was “chess without theory” – the games really did feel exactly the same as normal chess, but with both sides playing an opening they don’t know.

    I only got to play against much lower rated players but with that caveat noted I will say that all of my games were short and exciting – not at all like how people described Carlsen-Naka but those guys are quite good…

  29. @Thomas
    I guess it must be an American thing. They hate draws and that’s why they in general don’t like soccer. But I agree there were many exciting games at the world championship match and the results could easily have been different.

    Besides, who is to say Chess960 won’t end in a ‘draw death’? The technique of the top players is just incredibly good, and the drawing margin of chess is rather wide. But on our level, who cares?

  30. @The Lurker
    Peter Heine Nielsen, Carlsen’s chief coach, noted that in matches where the score is tied (can only happen in play-off matches) the chance of a decisive result is 22%, otherwise it is 44%. Per game.

    The solutions can be many. Play-off for draw odds first. Champion keeps the title at a draw. Champion keeps the title but challenger has an extra white in an odd game match.

    There is no reason to blame the starting position for the fault of FIDE. Of course now FIDE have created a further 28 matches that can result in a rapid play-off in the cycle. I think they will regret this eventually. Kazan 2011 was not great.

  31. Maybe this comment will actually make it onto the blog…

    @Jacob Aagaard
    I’m not saying chess960 is the answer. I am saying there is arguably a problem; chess is drawish by nature at high levels, and this is one of the reasons many people have no interest in it. (Granted, many people have no interest in it because it isn’t sufficiently macho and tribal, nor does it make for good Sunday night parties, or simply because it requires thought.) Ignoring the problem, as many here want to do, is not going to make it go away. FIDE may not have the right solution, but they seem to recognize a problem.

    @Ray
    That’s it. It’s an American thing. What an intellectually lazy argument.

    Yes, the technique of top players is incredibly good, but as Carlson himself has demonstrated time and again by playing substandard sidelines and thus taking other players out of book early, many players aren’t so impressive when they can’t rely on their encyclopedic memories. 960 is a way of making it so players cannot be playing from book on move 30. At least not in all 960 opening positions; not unless they have phenomenal memories. Just imagine your book prep being spread 1/960th as thin as it is now…

  32. I would respond, but the past two times I tried, my post did not make it. Not sure if it’s a browser problem, or what. Maybe later.

  33. @Jacob Aagaard
    And yet it was 12 draws in a row. I leave it as an exercise for the reader to calculate how probable that was. Chesstempo leads me to believe that about half of all top-level games are draws.

    I’m not saying 960 is the answer. I’m saying there is arguably a problem with making top-level chess events interesting, at least to ordinary people. And it’s not a FIDE problem. It’s an intrinsic problem with top-level chess. Ignoring it won’t make it go away. At least FIDE seem to recognize that there is a problem.

    Of course, for patzers like me and my friends, this is not an issue. We get a decisive result most of the time.

    Maybe the problem is the same as the problem with most other sports: money. When players can make money at this, they will memorize reams of theory to do it. 960 is a possible way to make memorization less of an advantage, by making players spread their memory banks much thinner. That way, over-the-board playing ability may be favored more. I admit this may not help the draw issue at high level, because over-the-board playing ability may be more even than I suspect. And even if it results in less drawishness, it may not make chess more popular. Most people may just be too lazy to want to think that hard.

  34. Also, I was only half-joking about what I facetiously called chess25,401,600. (Yes, that’s really the number of possible opening positions in this variant, if my math is right.) The benefit of this variant is that the opening position is not randomly selected, but chosen by the players, interactively, over the course of 16 ply. It might not have the problem with “bad” opening positions that some here think 960 has. But I could be wrong.

  35. It really depends on which events we are talking about. The Grand Chess Tour is a disaster in my opinion. I hardly follow it. But the Candidates, Gibraltar, Isle of Man, Poikovsky and quite a few other top events (Wijk, US Champs) have excellent chess and lots of fighting games. In GCT everyone wins, almost, as the only prize that matters for more than half the participants is to be invited back the next year. This makes for peaceful games. Also, they play the same players all the time.

    The Olympiad was exciting.

    Yes, Caruana did not take chances, although he must have known that the only challenger to ever win a tie-break at a World Championship match is Kramnik, who defeated a player that had never won a World Championship match in it. But still I blame this on the system.

    I am not against half the games of chess being a draw. I am against half of them being a dull Berlin…

  36. Chess 960 is horrible. Too many starting positions are forcing for one side.

    All Chess variants are lame. Play fucking Scrabble or Risk or Go if you need to do something different than chess.

    And the idea that draws are anathema to chess is the height of idiocy. It’s a game of perfect information and should be a draw.

    Professional game players deserve no sympathy for the manufactured problems they suffer.

  37. @Hard Truther

    A game of perfect information is not necessarily a draw. King’s Row is a checkers variant in which the first player to promote a man wins. It’s proven to be a win for the 2nd player.

    Anyway, I was never saying that draws are “anathema”, only that they are boring to the great majority of people, and that’s one reason why chess as a spectator sport is not so popular with the great majority of people. So, if you want to make chess more popular, something might have to give.

    @Tim S

    We don’t know for certain that chess is a draw with best play, no, but given Zermelo’s Theorem, chances are that if it were not so, we would at least have a notion of a winning strategy.

  38. @The Lurker
    It seems quite obvious to me that chess is objectively a drawn game – all evidence points that way. Still, I didn’t make any comment on that before. My previous comment was that I’ve heard people complain about some chess 960 positions being too unbalanced on several occasions. I would like someone claiming that to provide examples.

  39. @Tim S
    You’re absolutely right, about chess almost certainly being a drawn game, and about you not making a comment about it before. I accidentally hit Reply to your post, instead of Ray’s. Sorry about that.

  40. @The Lurker: “Anyway, I was never saying that draws are “anathema”, only that they are boring to the great majority of people”
    Chess being boring to the majority of people has nothing to do with its drawish nature but more with the fact that you have to know quite a bit to actually understand what’s going on. Otherwise it is like watching a political debate on TV in a foreign language you don’t know: You can grasp that something interesting is going on, but you have no idea what those people are talking about nor can you determine who has the better arguments. If things turn into a fistfight after a few minutes, things are getting more interesting though and my impression is that this is where some would like chess to go as well (and I’m not even talking about chessboxing here).
    If you want huge crowds following chess, you need a chess-educated audience. It worked in the Soviet Union, tournament halls were crowded with visitors. These days, Youtube, Twitch etc. might help to build up a larger audience for chess in general, but it remains to be seen, how much of these kids will stay tuned in in the long run.

  41. Ben :
    @The Lurker: Otherwise it is like watching a political debate on TV in a foreign language you don’t know

    That struck me when I was abroad watching such political debate: politicians sound the same in every language. To quote singer Richard Thompson:
    Oh there’s some who believe
    Oh there’s some who believe there are reasons to lie
    And there’s some who deceive
    And the truth is right there in their eye

    Well, happy Brexit everyone 🙂

  42. @Morten Nielsen

    Ahh… I searched for the sequence you provided, and found this:

    https://translate.google.com/translate?hl=en&sl=da&u=https://theworldnews.net/dk-news/sadan-vinder-du-kalaha-hver-gang&prev=search

    I also found this link that says that there is a sequence of length 30, that wins all the stones. Haven’t tried it yet. It’s only for the game with 6 holes on each side, and 4 stones per hole to start. Kalah(6/4)? Not sure I’m using the game theory notation correctly.

    https://blogs.mathworks.com/loren/2017/05/22/how-to-win-all-marbles-in-mancala-on-your-first-move-with-matlab/

Leave a Comment

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

Scroll to Top