Fixing Chess

Chess is broken. A game can be broken and still be popular, entertaining and beautiful. Chess is still a beautiful game, and is even growing in popularity. But that doesn’t mean that it’s perfect or that it doesn’t need fixing. What we are doing to this game by refusing to make any changes is terrible.

Every major video game in the world does regular “balance changes”. These are just standard changes that are made to the game to make it more interesting or fix it because something about it is broken. There are so many things that are not ideal about chess, that the chess community refuses to address due to adherence for tradition, lineage and an obsession with sitting in the same spot for 6 hours.

Magnus Carlsen himself, the World Champion of Chess and number 1 player in the world, even admits it in his latest interview. He says “In general it’s good to incorporate more rapid and blitz in the world championship because to some extent it is a purer form of chess because preparation plays less of a role”.


Photo by Maria Emelianova/

I have lost all patience with the diehard “never change anything about chess” crowd. They are dinosaurs and are making a mockery of what a game should be

What are the main problems with chess:

  1. The draw rate is absurdly high at the top level. Do I have to see another article at some major super tournament where the author writes a glee filled headline because a few games actually ended in wins? A perfectly designed game should not be a tie as frequently as chess is and this is super obvious. The drawing margin is too high, and you can get away with too many mistakes and still draw, especially with the White pieces. Yes some of the draws were interesting, but how about create a format where it’s not a draw every single time? Maybe that would be even more interesting!
  2. Computers have taken a lot of the life out of the game. Chess players are no longer willing to play any risky looking moves in the opening, because they fear their opponent’s computer preparation. They know that it will be absolutely perfect and that they will stand no chance if they walk into home prep in a complicated position. This could be seen in Game 2 of the most recent World Championship match, in which Magnus intentionally avoided what looked to be the critical line. He knew it was the critical line, but he also knew that Fabiano would know everything about it. Magnus didn’t want to play against a computer, so he played a tamer line and was satisfied to just play for equality.
  3. The games are too long. The top players play exceptionally well in faster time control games. Give them a time control of 90+30 and they will still play near 2800 level. But for some reason we have to add an extra two hours so that maybe they will play at 2820-2830. It’s so ridiculous and a complete misuse of time and resources. You have to be so stuck in the past to think that a three or four hour chess game just isn’t enough, and that it simply must be six hours. The extra benefit is that when the time controls are faster, even just slightly faster than they are now, people win more frequently.
  4. There are almost no more memorable games in World Championship Matches. We have grown up studying the great games in the matches with Capablanca vs Alekhine, Botvinink vs Tal or Kasparov vs Karpov. Tell me one memorable game from the last 6 World Championship Matches? They will basically all be forgotten. You can pretend that Carlsen’s fortress defense from Game 6 will be remembered, but it won’t. Actually I take it back…..there was one extremely iconic moment from a recent World Championship Match that I don’t think will be forgotten easily. It’s when Carlsen played the beautiful 50. Qh6!! to win the title against Karjakin. It just so happens that came from a rapid game! Once rapid chess starts getting taken more seriously, we will see that the games and ideas are so much more beautiful and much more digestible to 99%+ of the chess audience. Players will once again be able to go for speculative sacrifices against the best opponents in the world.
  5. The current format is not likely to determine who the best chessplayer is! Because classical games end in draws so frequently, even if you increase the match length to 24 games, you are often going to see someone win the match with 2 wins to 1 and 21 draws. This does very little to differentiate who is actually the better player. Sticking to the low number of games, slow time control format, increases the chance that the weaker chess player will win.

I have already written my solution in a previous blog but I’ll reiterate it one more time, with a small change, because I do think it’s the perfect balance of retaining that classical chess tradition, while not allowing all of the life to get sucked out of the game:

This is how a chess game would work:

You play one game at 90 minutes plus a 20 second increment. The winner gets 10 points the loser gets 0

If that game is drawn you reverse colors and play one game at 20 minutes plus a 10 second increment. The winner gets 7 points and the loser gets 3

If that game is drawn you keep the same colors as the rapid game and play one game at 5 minutes plus a 3 second increment. The winner gets 6 points and the loser gets 4. If this game is drawn, both players get 5 points.

A perfect mix of Classical, Rapid and Blitz.  The Classical game is still worth 5 times as much as the blitz game. The Classical game is worth 2.5 times as much as the rapid game. The rapid game is worth 2 times as much as the blitz game. But all the games count, and all forms of chess count. Every single player will need to be equally versed at Classical, Rapid and Blitz chess to consider themselves the best in the world.

The best overall chess player will almost always win handily with such a format. If it’s truly very close, then we will see some really tight and exciting matches. Right now everyone is so good at the top and the draw rate is so high, that most matches with absurdly long time controls are going to end with the majority draws and maybe one or two wins sprinkled in.

I think that Fischer Random is also a great idea, but there is something beautiful about the starting position in chess, so I decided to retain that chess tradition in favor of simply speeding up the time controls. But Fischer Random has the added benefit that Computer Preparation is rendered almost meaningless.

The chess community needs to wake up. The World Championship was just all draws. The one before it was 10 out of 12 draws.  The idea of making the match even longer so that eventually someone wins is ridiculous. Where exactly were players avoiding risks? You could say it was Carlsen in Game 12, but I believe he only got that position because Caruana was so hell bent on trying to win and avoid the rapid tiebreak. This is all so stupid and it’s so frustrating to be part of a community who can’t see it.

Everyone loves to resist change, and they eventually get left in the dust because of it.





12 thoughts on “Fixing Chess

  1. There is always room to argue about detail, but your proposal is motivated by the right ideas – especially the over-rewarding of preparation. It is particularly loathsome, of course, for a classical match to be determined by a rapid tiebreaker. Making the rapid and blitz part of the fabric of the match is fairer, given suitable weights. The matches might still have the same result (classical draws and decisive short games) but not in the dreadful irrelevant way they do now.


  2. Every other professional sport is aiming for a game to be played in 2.5hrs. The draw in game 12 came after 3 hrs. It’s impossible to rewatch the great commentary when it’s 6hrs of video. I’m not sure what the solution is (maybe 45/45 with a forced draw after 4hrs), but an aim of 3hr games with a max of 4hrs games would be much better than these 6-8 hr marathons. I don’t understand how any sizeable fan base could watch from start to finish right now.


  3. I do not think this idea is very good at all but I do agree the present system sucks. Here is a simpler idea; change time control to championship level, game in 90 minutes each. Then play randomly selected openings 2 games a day, best of 10 games, no rest days. If still tied, then whoever played the most exciting game, as voted on by astute judges, wins the championship. In this way we try and encourage creative play and eliminate having to have rapid or blitz play offs, which really should have no bearing on classical chess time controls. Like asking 2 people who tied in a marathon race to now run a 100 yd dash to decide the winner. Another solution, also better than proposed, (imho) is armageddon classical chess with a randomly chosen opening, including those considered dubious, like Barnes Opening (f3). No way to prepare for every opening known to man and we force a winner to be determined more fairly than playing dubious faster time controls where one is clearly better than the other. Game to be played on day 6. By the way, when the current best in the world in rapid and blitz chess (by rating) thinks it is a great idea for faster time controls to be a part of the classical chess world championship, you should already know there is a problem. Everyone knows that MC is best at faster time controls and they would only add to his advantage.


  4. I think the match was badly played by both to a degree. All of the play took place in the 6 White games of Caruana. One side was winning in 5 of them. Every time the player with advantage failed to convert – clearly outside of computer preparation and therefore completely shutting down the idea that computers are killing chess. Human players are still making human mistakes in matches, as they always have.
    There was also a vote on one of the largest websites which showed that most prefer classical chess as the only option. I agree with them as it defeats the logic of the match. Expand the match to 24 games and maybe FIDE/Agon can stop playing chess with their income statement.


  5. Why Greg is Wrong:

    There’s already a World Rapid Championship. It lasts 2 days and no one cares about it. It has no suspense and no buildup. The quality of the play is low, below 2600 standard on even the top boards.

    The World Championship this year was extremely tense, and we saw 6 Classical games in which one side had +1 or a clean win. That means in half of the games one side was winning and fans were rooting, cheering, and screaming for the victory that would have pushed the match to the brink and likely won the match on the spot. In half of the games, the match could have been decided outright. This made it an amazing match, with so much suspense and tense moments in almost every game.

    The number of draws had nothing to do with the amount of suspense and how many people watched. Three times as many people watched this match as the second Anand match. Anand was beaten easily and decisive games weren’t what made it interesting. What made this interesting was the close level of skill, and the desire for the games to be decided as fairly as possible. Playing some blitz games just adds a huge factor of randomness that the spectators didn’t want. It’s not what we came to watch. See who watches the world blitz championship from the press. Answer: no one you stupid idiot.

    As another commenter mentioned, the biggest chess Facebook groups voted and there was a 9 to 1 ratio of people who wanted Classical only to decide the World Championships.

    Nakamura said that about rapid and blitz chess because he’s good at blitz chess and has no chance in classical. It’s called a biased comment, Greg. Have you never understood what bias is?

    Your solution is very stupid because: 1. This was the most watched match of all-time already. 2. A 24 game match has never been a drawfest in chess history, and only in the case of the 1987 match was it fairly drawn on the board, with Kasparov dramatically winning the final game. 24 game matches solve all of the problems everyone is asking about. 3. Your solution makes the strongest rapid chess player a major favorite to win the match. It would mean the most important thing in match preparation is preparing to play rapid games. That’s extremely stupid and completely changes the match. 4. Fans watched this match more than any other. You can check the statistics on how much rapid chess is watched by chess fans and you’ll see it’s about 10 times less. This is just your wet dream, and you fall into the 10% of people who want it. You aren’t objective on this topic at all. We get it. You love decisive results and fast games. 90% of the chess world has voted and disagrees with you.


    • While I agree with some of what you’ve written, Greg, disagreements are much more interesting; so here goes…
      I’ll start with your point #4. For me, Carlsen’s 50) Qh6 in the final game of his match with Karjakin is hardly an iconic moment and I don’t think the move deserves two exclamation marks. My feelings about this move are similar to your sentiments, which I share as well, about authors writing glee-filled headlines about tournaments with a few wins. It’s a forced mate-in-2 after Karjakin’s 49th move that most of the Class-A players I know could spot. So why is everyone so ga-ga over it!? Sergey surely saw it and wanted to allow it and Magnus would’ve been laughed at if he didn’t play it. The game itself was not that great, mostly because Karjakin was in the unenviable situation of having to win as black in a rapid game and so compromised his position to avoid making a draw more likely. To me, 50) Qh6 is more novel–the only time a World Championship match has ended with a queen offer–than beautiful. The novelty may make the move more memorable, but the game is not. Bent Larsen beat Tigran Petrosian twice in the 2nd Piatigorsky Cup tournament in 1966. In their first game he sacrificed his queen on g6, which he described as not that difficult to calculate. While he acknowledged that spectators may have been more pleased with his queen sacrifice game, it was his later win over Petrosian of which he personally was more proud.
      As for memorable games in the last 6 WChs—that only takes us back as far as Anand-Topalov–I’ll site Anand’s game 3 win over Carlsen in their 2nd match. Game 6 of that same match saw GMs and lesser players light up social media after the back-to-back blunders by both players. And there is also game 11, which is memorable to me because it contained what I thought was the best move of the match—Anand’s 23) …b5!!, which even took Carlsen by surprise—followed by the worst follow-up, in which Anand unnecessarily sacked the exchange in a poor way for no compensation and Carlsen won the game, clinching the match. Game 6 of their 1st match was riveting to me and probably showed Carlsen at his signature best, pressing in a pawn-up rook ending where Anand came up with the brilliant defensive resource 44) h5!!—another move Carlsen missed–going down two pawns to fracture Carlsen’s kingside pawns. What was looking to be one of the great defensive displays in WCh history went for naught when Anand, in a position the computers rated equal, played 60) Ra4?? and resigned seven moves later.
      How about Anand-Topalov? Are you sure you don’t remember game 4, which Nigel Short described as “Wonderful stuff from Anand” after the Champion sacrificed his knight on h6? How about the nice attack Anand played in game 12 starting with 30) …f5! to win the match? Even if these games are not remembered much in the West, I imagine there are a lot of young players throughout India who have these games etched in their minds along with Anand’s game 5 win versus Kramnik in 2008.
      And while I know this goes further back than you asked, I have to mention Peter Leko’s fantastic win against Kramnik in game 8 of their Brissago encounter.
      I don’t like rapid tiebreaks, but what I don’t like more is going from 40/100; 20/50; sd/15 for twelve rounds to suddenly G/25. That’s just stupid to me. Why not play G/90 or G/75 with a 3 second delay initially? The World Championship IS about classical time controls, not rapids or blitz. If those faster time controls were meant to be a part of it they would be incorporated into the main body of the match, e.g. play 6 classical/6 rapid/6 blitz and whoever has the most points wins. There’s a reason rapid and blitz are relegated to tiebreaks: we don’t want the title decided this way. Saying “It says World Championship, not CLASSICAL World Championship blah, blah, blah” is really just a horseshit excuse. There was a time when top level tournaments and the World Championship had time controls of 40/150; 20/60; sd/60. We are now looking at time controls that give each player less than 3 hours. I don’t think we need to make the game faster and I wouldn’t advocate anything less than 2.5 hours per player. When was the last long, brilliantly subtle—dare I say ‘memorable’–endgame you saw played in a rapid game? The videos for the Go match between the AlphaGo computer and Lee Sedol run 4-7 hours. Go fans don’t seem to care and why should they?
      It also annoys me that Carlsen has twice now played to win in the rapid. Against Karjakin he didn’t even try—with the white pieces—to win game 12 while versus Caruana he had a very nice position, which many thought would be difficult for Fabiano to defend, and offered a draw (much to Caruana’s shock and delight). In lieu of this I have two ideas that the players may not like but organizers and sponsors could demand (I don’t think the players would turn down the payday). First, no draw offers until at least 40 moves (or 50?) have been played. Not original, I know, but it does increase the chance of a decisive result and it is not clear how Caruana would’ve survived 8 more moves in G12 of his match.
      Second, create a financial incentive to win in the classical time control. If you have a $1million prize split 60% (winner) 40% (loser) make it a 50-50 split if the match is decided in the tiebreaks. Do you think Carlsen would’ve kept playing in G12 if an additional $100,000 was on the line? Another way to approach this would be to offer, say, a $1million prize fund with an additional bonus of $100,000 if the match is won in the classical time control. Another, more radical, idea that I would propose is that if the match is decided in the blitz tiebreak, the loser actually gets 52.5% of the prize fund. The winner hasn’t really proven they are better at long time controls, but by virtue of being declared the Champion will enjoy a greater income and perks the title brings.
      I agree with your point #2 and I think it contributes somewhat to point #1, though I don’t personally feel there are too many draws—just to many uninteresting ones. Many of the draws in Carlsen-Caruana were fascinating and complicated games, which is fine with me.
      Finally, why do we keep trying to make chess popular to people who have only a cursory interest in the game—who don’t even know what an en passant capture is or the correct rules of castling. Chess is growing in popularity. We’ve had World Champions from India and Norway with the most recent challenger from America. China and the USA won the last two open Olympiads while China also won the women’s. It will continue to grow… until computers have sucked all the human spirit out of it.


  6. the idea is to expand the market share of chess to people who are not devoted to chess already, craig. In order to do this, we need to cut out the boring 4-8 hour games that 90% end up in boring draws. People who are already chess fanatics are fine with classical chess, but the average joe who just knows the rules and doesn’t already follow top level chess, is not gonna be ok with that. Right now its almost impossible for many IM’s, GM’s to make a living playing chess unless they are in the world top 10. This is because the market for chess is too small. Think about how many minor league baseball players make a livable wage… as an example…maybe you are fine with the status quo but many people want more…


  7. Pingback: A Tiebreak Win and the Problem of Draws | Gödel's Lost Letter and P=NP

  8. I don’t know about the need to remove draws from chess. At the highest level, chess is just a drawish game: the world championship has shown that, as well as the recent AlphaZero-Stockfish matches. Gimping play so that there are fewer draws feels like deciding not to play chess at its highest level any more, and I don’t think a sport can survive if no one is trying to play it at its best.

    I think the movement towards rapid or mixed chess has been great. It’s a lot of fun to watch, there is more opportunity for must-win games or players coming from behind which incentives aggressive play and interesting chess. Perhaps those formats will end up replacing classical chess, in which case there’s nothing to do: we’ve already found the non-broken way to play chess, and just need to wait for the broken part to die. But I think that there will always be a desire to see what humans can do if they aren’t constrained by the clock.

    There was a lot of tension in the Carlsen-Caruana match with wondering who would be able to strike first. It was too bad that the extent of computer preparation did seem to make the players, especially Carlsen, more risk-averse.

    Maybe there are ways to make a draw more of an interesting result. One crazy idea might be to only award the world championship to someone who is able to win it. If it can’t be won in 12 or 18 games or whatever, then the title would be empty, there would be no incumbent to seed into the Candidates Tournament, etc. That might give real incentive to play for a win towards the end of a match, as drawing a match at that point is almost the same as losing. This would probably mean that a slow World Championship match would have to be done in conjunction with some faster match, as finding a sponsor for a possibly inconclusive match would be hard.


  9. Solution is so easy and still nobody can see it. Carlsen’s right, you ought to include rapid & blitz games, but not as appendices, and not at the beginning (like someone else already suggested); you’ve got to include them IN BETWEEN.
    Let’s say we’ve got a bucket of 16 classical games (label C), 4 rapid games (label R), 4 blitz games (label B) – 24 games (!) like nostalgia is suggesting – the solution is simple: you’ve got to play them ALL, regardless the outcome of the classical games. It’s up to FIDE decide what’s the best order. I’d say to leave blitz as a “tail” still, but to put rapids in between classical games. E.g.
    C C C C R R C C C C R R C C C C B B C C C C B B
    If you don’t like blitz at all you can leave them out as last resort if after classical and rapid games players are still tied. E.g.
    C C C C C C R R C C C C C C R R C C C C | B B B B (blitz games not compulsory)
    Not need for funky scoring systems, no need to play rapid & blitz at the beginning just to get an artificially unbalanced, awkward situation. Easy!


  10. The high percentage of draws, while it does indicate an unwillingness to take chances, does show the high level of skill. Only the primitive would enjoy a lopsided contests of Morphy.

    If you want to make the game more exciting, let’s introduce alcohol. If the computer thinks you’re ahead, you have to take a shot.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s