Today I played as White in an interesting bullet game in which a strange “fight” occurred.

[FEN ""]

1. d4 e6 2. Nf3 b6 3. e4 Bb7 4. Bd3 g6 5. Be3 Bg7 6. Nc3 Qe7 7. Qd2 Nc6 8. a3 a5 9. h3 h5 10. Qe2 Bh6 11. Bxh6 Nxh6 12. Qe3 Nb8 13. e5 Ng8 14. Qd2 d6 15. exd6 cxd6 16. Qe3 Nf6 17. Ne4 Ba6 18. Bxa6 Nxa6 19. Nxf6+ Qxf6 20. Qd3 Nc7 21. Qe3 Kd7 22. Nd2 Rae8 23. Nb3 Nd5 24. Qd3 e5 25. O-O-O Nf4 26. Qf3 Qg5 27. dxe5 Rxe5 28. g3 Ne2+ 29. Kb1

During the game, it seemed, to me at he least, that were playing out a “fight” of who would castle first. In my thoughts, it is advantageous to castle on the same side as the other side, so we tried to make each other castle first to know each others intentions. In other words, it was a drawn out tempo play revolving around who would castle first. This fight lasted in my game until Black played 21... Kd7.

My question is not about my game, but about this concept of such a castling fight. What should be done if you find yourself in one? Are their any strageties for these situations? Finally, have there been any top level games where a, preferably somewhat extended, castling fight occurred?

  • 3
    i may have time later to research this, but there is no doubt that top players have often held off on castling trying to see where the other guy went first. – PhishMaster Jan 16 '20 at 18:45
  • To what end? I can see weak players doing it with the intention of an attack on the king. But question if top players do it for that reason. More like they do it because they have other good moves to make first and have no pressing need to castle for safety reasons. – edwina oliver Jan 16 '20 at 19:04
  • 2
    Not a duplicate in my opinion, but some overlap with this old question. The answers there may be interesting. chess.stackexchange.com/questions/573/… – Michael West Jan 16 '20 at 19:25
  • @edwinaoliver Top players can still toy with the threat of a direct attack on the king, even if they know that their opponent will be able to defend - the threat alone may still divert some ressources. – Annatar Jan 17 '20 at 10:14
  • I know of no top player who are concerned about that. They will still look at the overall position not worry about some slash and burn attempt at a direct attack. If necessary they will defend as needed, but the threat is not meaningful wrt delayed castling. – edwina oliver Jan 17 '20 at 22:02

You may have been in a fight or not. Top players do not fight that way they make the moves best for their position. Sometimes they delay castling but not to fight the opponent.

This seems to be a tactic used by weaker players who might want to mount an attack on the other king. The position should decide whether that is the best strategy or not.

Not sure why you two did it that way as there had to be better moves available. it looks more like you all were struggling to just make moves so did what you did. So being a bullet game might explain some of the moves that I would not have considered.

That said I would have just moved the white king up earlier and not castled at all as there is no danger of an attack and it would ensure that my king has an endgame advantage should black castle. I note that is what black did to white later.

But then I would have made a number of different moves earlier on that would have changed the entire game making the question moot:)


I'd question whether its "advantageous" to castle on the same side as your opponent. Often it's safer, but that's not always the case and even so "safer" does not mean "advantageous." Remember, when you castle on the same side as your opponent, that means they've also castled on the same side as you; it can't be "advantageous" for both of you at the same time. Any advantage you have has to lay elsewhere; advantages come from imbalances, not similarities.

Most of the top players know which side their opponent is going to castle on within the first half-dozen or so moves, because the castling choice is indicated by the characteristics of the position on the board. It's very rarely that a player pushes their g-pawn one step forward and doesn't castle king-side, for example.

The position itself will tell you where it's advantageous to castle. Black will only rarely castle Queen-side in the Stonewall Dutch or the Sicilian Dragon, no matter where White castles. It just doesn't make positional sense to castle anywhere but king-side in those kind of positions.

My answer, as you might guess, is "Don't find yourself in one." If you intentionally delay castling, you delay completing your development because you can't connect your rooks until you've castled. An advantage often lies in being the first to complete your development, meaning being the last to castle may often work against you, not for you.


Let's ignore the whole "fight" aspect for now and consider why we castle in the first place. The key advantage of castling is that you 1) bring your rook to the center, and (more importantly) 2) get your king to safety. Exceptions apply of course, but usually, you're looking to castle on the side where the opponent isn't going to attack.

The other thing to remember is that the opponent's king is usually going to be behind a wall of pawns. If you're going to launch an attack, you'll want to advance your pawns to break apart that pawn shield. Of course, the opponent is also thinking the same thing.

The position of your king influences these plans. Consider your game position after 21. Qe3. If Black plays 21...O-O, then 22. O-O-O intending Rdg1, h4, g4 etc becomes a legitimate plan for you. This doesn't mean it's a good plan - Black can also play Rfc8, Rab8 and b5-b4 etc - but it's a legitimate option. Similarly if Black plays 21...O-O-O, then 22. O-O followed by a queenside pawn advance becomes a legitimate plan for you.

What if, after 21...O-O, you decide to play 22. O-O? Then neither player will be attacking on the kingside (advancing your king's pawn shield is not a good idea unless the opponent absolutely cannot mount an attack on your king). The game's battleground moves to the queenside and it'll take on a strategic theme where direct attacks don't happen. The same goes if you meet 21...O-O-O with 22. O-O-O.

Therefore choosing which side both players castle into changes the strategic character of the game, and accordingly the available plans. You'll want to consider these before deciding where to castle. For example, in your specific game, you could decide that 21. O-O is a fine move because if Black plays 21...O-O-O and attempts to attack your king, your queenside counterplay will be faster (I'd bet on it; Black's queenside has vulnerable squares plus you have a pawn majority on that wing).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.