On Castling?

Originally posted by gorookyourself
No. I am wrong. Sorry.

No need to apologize, being wrong is part of being human, and owning up to it is a sign of honesty, and humility.

Ooh considering studying to be cheating that kinda thinking is like a sign of brilliance methinks.🙂

Originally posted by National Master Dale
Ooh considering studying to be cheating that kinda thinking is like a sign of brilliance methinks.🙂

What?

"only sissies castle" -rob sillars
"castle early and often" -rob sillars

I castle as soon as I can in the game-however, I will not castle if I can maintain tempo with another move that I believe improves my position.

And to be blunt, hundreds of years of chess theory have evolved to show the value of castling, every world champion has done it with extreme regularity, and all the strong players do almost as a routine.

For a new player to think he knows better than these is arrogant stupidity -and I'm sure your friend is neither, but he should be warned about appearances. I speak from personal experience, as I was in his shoes once!
.[/b]

I'm not sure I agree with this sentiment. I've often thought about experimenting through games without castling my King. I generally have the hardest time playing against opponents whose King is in the center of the board, as a castled King gives me a stationary object to focus my attack on. I realize that this difficulty is due to my own ineptitude, however I can't agree that since grandmasters castle in general, I should castle in general -- at least not in casual Chess. While playing by good Chess principles seems to give me better results, I don't like playing moves based off of principles without understanding the principles myself. Until I learn how to exploit a centralized King, I think I will remain ambivalent about castling. In my opinion, this is not an arrogant position to take. Ignorant, perhaps, but not arrogant.

Secondly, as computer Chess advances, I wonder if the principles we have played by for the last few centuries will pan out to be good Chess principles after all. Computers don't "intuitively" play by human principles, and in the beginning of computer chess this seemed to have made them significantly weaker. However, as Chess machines have advanced far beyond even the best humans now, I wonder if computers can give us some insight as to how effective some principles really are.

While we're on the subject of castling, I'd like to post the following game which I find particularly beautiful:

Aha, I knew castling was for chumps (and GMs).

In a few games recently I have been in positions where my castled king was under threat of perpetual check. Is this a common risk, and in a relatively even game is it better to play a slightly inferior move to avoid the possibility, or play a stronger move where my opponent could force a draw?

Originally posted by Willzzz
Aha, I knew castling was for chumps (and GMs).

In a few games recently I have been in positions where my castled king was under threat of perpetual check. Is this a common risk, and in a relatively even game is it better to play a slightly inferior move to avoid the possibility, or play a stronger move where my opponent could force a draw?

I would imagine that it would depend on which player is stronger. If you're stronger, you might decide to play the slightly inferior move to keep the game going. Otherwise, you'd probably be better off playing the stronger move and risk settling for a draw. This is similar to the contempt factor in computer chess.

This thread actually inspired the Caslting blog Blog 4
So felt I could add more.

These figures in the blog do show White 0-0 and an uncaslted White
as being very close.

White 0-0 and got mated = 3,444
White uncastled and mated = 3,642

But as this sample is taken from a database of weaker players games
it will be giving a false figure.

At that level first to attack usually wins as the standard or defending
is so low.

I've won a fair few games tossing a speculative pawn or piece to stop
a King castling and indeed there are many famous traps showing this idea.

However when you try these things v a good box then there is a good
chance you will come unstuck.

They have even busted some famous traps by discovering hidden resources
that just look bad but are infact OK.

These things have no fear about walking a King into an open board as long as
they cannot see the mate and I've seen 'White wins analysis' which has stood
for 100 years turned over by a wandering Black King handled by a box.

When you play a good box it will castle because it's still in it's book.

(I bet computers hate having to follow human books moves and I imagine
secretly inside they cannot wait to get out of the book and unleash their power).

Play it with 'book off' and yes it still castles but only if it thinks this is the best move.

It's not adverse to go on a pawn snatching mission and go through a nailbiting
attack which at first glance 'must lose' but infact wins.

Reti was ahead of his time with his.

“Castle because you have too, not because you can.” statement.

Do we have figures for when white castles queenside and then wins?

I generally got the impression it was a risky move.

My two cents.

1: It connects the rooks

2: It generally safeguards the king

3: The king tends to obstruct his rook from joining the action, which leads to...

4: Rooks like open files, which typically occur toward the center.

GP, a good summation?

Originally posted by Willzzz
Do we have figures for when white castles queenside and then wins?

I generally got the impression it was a risky move.

I sometimes castle Queen side. At my level it sometimes throws my opponent off because they're not expecting it, especially if they over commit to going after my king side.

It does often lose a tempo if I feel compelled to scoot my king over one more square for safety.

Originally posted by beetlebomb
Here's what he said:

"I read their posts. I think I might be looking at it incorrectly. Whereas I see fewer escape routes in case of attack.....they see fewer attack routes that would require attack".

And thanks again for the prompt responses.

The thing about this is that "attack" doesn't have to mean a mating attack, at least not immediately. When your king's position is opened, the opponent can target it for a variety of forks, pins, skewers, etc. (As well as deeper tactics, which of course are generally made up of these building blocks, as your friend may not yet be aware.)

Weak players are likely to simply overlook a dangerous threat, losing a bunch of material in short order. Strong players will have to defend against such threats, even if it means making moves that further weaken their position. An "escape route" doesn't help with this: in fact, if the escape route leads your King into the open, it'll only make the problem worse!

Originally posted by amolv06
I'm not sure I agree with this sentiment. I've often thought about experimenting through games without castling my King. I generally have the hardest time playing against opponents whose King is in the center of the board, as a castled King gives me a stationary object to focus my attack on. I realize that this difficulty is due to my own ineptitude, however Re6 Kg7 40. a7 h3 41. Rxc6 Rxc6
42. gxh3 Ra8 43. Be4 1-0[/pgn]

I consider this speculation in spite of the facts, and I am inclined to give Grandmasters the benefit of the doubt without solid evidence to the contrary. To each his own, though!

Castling is most certainly not an absolute (and I think no one has claimed otherwise), but Karpov's sentiment (as expressed by GP34) is an opinion I am not inclined to dismiss lightly. He thinks as he does, and acts on his thinking, for a reason.

Here's a game that is, for lack of a better expression, a synthesis of both ideas. Black cannot afford to castle per se, but wants to achieve the advantages of castling, so he does it "by hand". Enjoy!

Originally posted by greenpawn34
This thread actually inspired the Caslting blog Blog 4
So felt I could add more.

These figures in the blog do show White 0-0 and an uncaslted White
as being very close.

White 0-0 and got mated = 3,444
White uncastled and mated = 3,642

But as this sample is taken from a database of weaker players games
it will be giving a false his time with his.

“Castle because you have too, not because you can.” statement.

You've really only partly answered the question- to finish, how many times did white castle and NOT get mated, and how many times did he forgo castling and NOT get mated?

I suspect the total sample of games where white castled far exceeds the sample where white did not.

The numbers as given show white getting mated more when he doesn't castle, and I think the real story is worse, as white more often than not castles.