When do you Castle?

If I don't recall Karpov was top dog for a very long time and I would dare you to find a GM or IM that describes him as an attacking player...Capablanca was a well rounded player more and endgame specialist than an attacking player and he is in the list of hte top 5 in history. Botvinnik was no attack specialist. Anand is not attack an maniac. Kramnik is by far one of the most defensive players alive and he beat the worlds best attacker perhaps in history. So if it's history your looking for it tells a different story than the one you paint.

Actually it was someone from Sweden that first gave this advice ! :-)

"Castling is usually regarded as an important way of developing one’s game, but that has not always been the unanimous view. In C.N. 3119 Calle Erlandsson (Lund, Sweden) gave an English translation of the relevant passage of the first (1771) edition of C.W. v. Königstedt’s Kort Afhandling om Schack-Spel (‘A Short Treatise about the Game of Chess’). Chapter VII (About Castling) states, on page 23:

‘Great players never castle until the end of the game, and often never at all, as their king, although often in the middle of the board, nevertheless stands secure.’

Mr Erlandsson added that in the third, improved edition, printed in 1806, the start of the text was slightly different: ‘Good players seldom castle ...’ "

http://www.chessbase.com/newsdetail.asp?newsid=5966

Welllllll, I would castle if there were plenty of pieces surrounding the king.

If they have the same rating, there will be similar results. The great defensive players may have had a lower elo than the attackers, but how you can assume that it's not just because they're simply better players as compared to "it's the superior style". And of course by defensive, we don't mean they will always defend, but they are less likely to sacrifice for attack.

Mr. Nimzoindian, it does require a more advanced level of thinking to see the disadvantages of castling (or doing it soon), but even still castling is a very necessary move most of the time, in most queen pawn openings and many non sicilian king pawn openings (and even the sicilian kingside castling is done; the attack is harder to launch, but so is black's, and it has to be positional pressure if the king isn't there). But if it's not a priority to do so, you could postpone it if you feel you can really use that extra tempo, but not in a quiet position most likely.

I castle before certain lines open, or when it's convenient to do so.

Generally, I castle when it is safe to do so, and either I intend to initiate action in the center, or anticipate my opponent doing so.  If the position is very closed, I am more likely to leave my king in the center.  A king in the middle of the board even if on the back rank is usually much easier to attack than a castled king due to the greater number of lines open to the attacking pieces.

Casteling doesn't have to do with any kind of defensive/offensive strategy.  There are concrete things in each position to consider.

The general things you look for are:

1) Is the pawn shield still intact (a,b,c pawns on queenside and f,g,h pawns on kingside)

2) Sometimes there are more important moves to make, such as fighting for the center squares.  You can also put it off to watch how your opponent is developing when you have a choice of sides.

3) The king is sometimes safer in the middle when the center is locked as play will inevitably happen on the wings.

3a) If the center is open or about to be opened, you should quickly look to castle.

4) If casteling to a side where your opponent's attacking pieces outnumber your defensive pieces by 2 or more, calculate carefully as an attack is sure to follow.

The best time to castle is move four, . Actually a good rule of thumb: castle as part of your development not beacuse you can. Castling is a move to consider based on the needs of the position.

Castling is a surprise element that can out-fox your opponent. It should be an option that can be exercised when necessary.

Yes I know. I was being mischievous :)

Seriously though, an interesting scenario is when both sides have developed all the pieces and the "Where is the king going ?" question is asked. The two sides then have to think "Is there a USEFUL waiting move I can make in the meantime ?" I'm not sure this question has ever explicitly been asked in  chess middlegame treatises.

Simply and example position...

well, what i do is d4 then whatever happens play Nf3 then e3 or g3 and move away the bishop then i can castle.

oh and Queen side castle isn't because the "breathing room", it's because the rook comes out and it is very powerful, but be careful with the a file pawn you don't want to lose it

Another thing with Queen side castling, be careful when you move your queen out of the way, try not to put it infront of the King so it does not get pinned to it.  See the diagram bellow to see exactly what I mean.

A way to stop this is simply to move the Queen to e2 instead of d2...  But still most beginners or even more experienced players don't always see this.

Not true.  The main disadvantage of casteling queenside is the possible loss of a tempo when moving the king to b1.  However this is usually offset by the fact that the rook gets the d-file immediatly.

While the possibility of winning the queen with that kind of skewer is sometimes present in a position, it doesn't often happen in actual games.

Lol one possible tactic is not at all a major disadvantage.

Sometimes it is...  I've been caught off gaurd hundreds of times with that kind of scewer.  That's why I almost never castle Queen side if i can avoid it.

 You have forgotten the mighty Tigran Petrosian , also NOT an attacking player, his opponents rarely understood what was going on until it was too late. He is also known as one of the greatest defensive players of all time.

That's not a good habit, especially since tactics like those are so avoidable (without abandoning your castling plans!)

Smyslow wasn't really bad on the defence either, I think it´s safe to say that Cheater_1 is just full of, to use Rebs favorite word.

Hogwash.

Even in the position you gave, it's as I said, it contains the posibility of the skewer but it's very unlikely that it will happen in the game unless white is completely blind to not only that little tactic, but the genearl strategic themes going on in that position.  Black to move would love to open up the center with dxe4.  With his two bishops, an open center, white's poor kingside structure, and generally exposed king, black would be easily winning.

Qd2 takes non of that into accoutn and is a simply a blunder, there's no reason the queen should move there right now with f4 so obviously in black's hands.  If my opponent played Qd2 I would immediatly throw out Bf4 to stop casteling with a gain of tempo.

The natural move in that position is to play e5 to prepare to Qd2 (getting back f4) and 0-0-0.  Especially because of black's bishop pair you want to make black's center structure (d5, e6) static.  Black will probably pry it open with f6 pretty quick, but after d4 you have control of d4 and e5 anyway which will be important.  Also this sequence makes black's knight on h5 look off sides.  But most importantly you're getting your king to safety -- it won't be very safe on the kingside and certainly would be bad in the center.

It may have been in euwe kramers middle game books that I read it -- but there had been discussion on which side was generally better to castle.  The main consideration was that often the king has to spend a move to get to the b file -- the authors countered with the fact that the rook immediatly gets a center file and that in 0-0 setups often requires it's own 2nd tempo with moving the rook to the e file.