chess online
« TAP TO LOG IN

Play online chess!

Castling - is there statistical evidence ?
« Back to club forum
Pages: 12
Go to the last post
FromMessage
servantmk1042
26-Dec-19, 07:09

Castling - is there statistical evidence ?
I know that I am kind of stuck on this one issue. I brought it up once before and did get some very good advice that helped to guide some of my playing regarding castling. I do try to make it in the 1st 10 moves if I am able. I know there are occasions when castling after that can make great sense. But I also know that some see castling as a detriment at times or even most times.

Questions :

1 . Is there statistical evidence that anyone is aware of overall that castling makes the outcome of
a game generally more successful than not castling ?
2. Is there just a preferred way of castling that people take in certain openings ?
3. Are the first two question to be regarded as bogus and lazy ? I seriously do not know this.
4 . Is castling to the Queens side a tactic to be used more after the 1st ten moves ?
5. Do some people feel that castling is really just a tactic that slows down the inevitable ?

I have played games where I have castled or not castled for various reasons. More often than not if I don't castle it is because I do not get the opportunity. If I castle on the Queen's side , it is rare and it is generally more a case of " Will this help right now ? " than I really want to do this. I do understand that castling may simply be a response to the other person's game.

What say you ?
shirlmygirl
26-Dec-19, 13:44

I learned chess from a book and chess set instructions, but was really very poor at chess. When I joined GameKnot, I started at 1200 and rapidly dropped down to 1100 or below. I had never castled when I joined GameKnot and had never heard of stalemate. First time I got into stalemate I was shocked as I had several major pieces still on the board.

With regard to castling, I now do it as early in the game as I can. I get myself into my opening position and castle late in that. But sometimes it is advantageous not to castle. Also, my opponent might get me in a situation where I can't castle to the correct opening side, but may be able to castle later to the opposite side. Whether it is advantageous to castle early or not, I don't know, but sometimes I can see the danger approaching and castle if I can avoid the danger, perhaps castling to the opposite side. Perhaps it is only a matter of choice as to whether one prefers to castle or not. Better chess players may have an different opinion.
mo-oneandmore
26-Dec-19, 14:55

why castle?
I personally see limited evidence for a need to castle.
apatzer
26-Dec-19, 15:44

vizual-statistix.tumblr.com


66.media.tumblr.com
shirlmygirl
26-Dec-19, 17:22

I don't understand those links at all, Robert. I am no expert.
shirlmygirl
26-Dec-19, 17:38

Now I see in the Percentage of Players Who Castled that approx. 75 percent did castle.
servantmk1042
26-Dec-19, 18:43

Thank you all
I will study these comments. I know that it all will be helpful to understanding this game better than I do now. I especially loved the comments of Shirley whose chess has surpassed mine. I just saw this and so the vizual-statisix.tunlr.com and the 66.media.tumblr.com I will check out tomorrow. Maybe it will all help me get past my constant basement play in the game. But if not, I will still always love this most marvelous game. Ever since I learned it as a kid, I have loved chess.
pawntificator
26-Dec-19, 20:18

In your previous thread I mentioned that for teaching purposes you should definitely teach castling as early as possible and possibly exercises practicing both castling and not castling.

In my own play, I will postpone castling for as long as possible. I instinctively want to avoid castling if possible. Here are the rules I use for my own decision to castle:

1. There is immediate danger. If my king is going to be in trouble or my pawn structure would be compromised I will castle to avoid it.

2. I have an opportunity to break open the center and attack. I will usually castle first just to get my king out of the way of counterattacks.

3. I can't think of anything else good to do. If it's generally early or middle game and all of our pieces are developed and there is an abiding tension in the center that isn't ready to break yet then I will generally castle and see if the other guy loses his nerve first and tries to force something to happen.

4. One of the rooks gets under attack. Often the b or g pawn will be captured by a queen. Sometimes it's a "poisoned pawn" and sometimes it's just a mistake I made. In the case of a mistake, I will usually castle in the other direction in order to connect the rooks.

I can't really think of any other "rules" I use. I try to avoid castling as long as possible because the king in the endgame is a powerful piece and wants to be in the action.

I doubt there are statistics like you asked for. Usually castling for both sides happens in most games. I'd be interested to see any analysis of such statistics.







lord_shiva
27-Dec-19, 11:56

Castling
You kind of want your rooks on center files, to support control of the center of the board. To that end, and clearing the back row between rooks so they protect each other, castling is useful.

Plus, you get two moves.
shirlmygirl
27-Dec-19, 12:12

I am picking up some good tips myself from these posts by the higher rated players  
servantmk1042
27-Dec-19, 15:46

thank you
the point about rooks on center files and clearing the back row will be useful I am sure. Thanks lord_shiva. Good tips for sure Shirley.
lord_shiva
27-Dec-19, 22:09

Grandmasters
The grandmasters tend to castle almost every game.

Steinitz felt the king was a strong piece and avoided castling. Stats on 53 players ranging from Morphy to Carlsen: 80% white O-O, 9.6% O-O-O

Black 81% short castle, 5.5% long. So about 90% of the grandmaster games castle.

Stats are from Reddit.

I saw a game once ending with O-O-O#. Awesome!
lord_shiva
27-Dec-19, 22:21

Castling: A Love Story
In my past ten games white castles every time, only once long. Black castles eight times--two times was forced to move the king--and never long.

Castling generally occurred by move six or seven for white, though sometimes not for 15 moves. Black tends to delay castling a few moves longer in my games.

I did not check to see if I was the one who did not castle those two games.
lord_shiva
27-Dec-19, 22:33

More
Analysis shows a couple of castings were inaccuracies or mistakes. Black castles in 5 in one game which analysis said should have been delayed to move six, and no castling was suggested in the other game.

Black castled long once, contrary to what I wrote above. Just my past ten games. I was curious.
kingseabass
27-Dec-19, 22:57

[I saw a game once ending with O-O-O#. Awesome!]


That is truly awesome! I would love to have watched that game.
5imon
28-Dec-19, 11:54

This game didn't end in 0-0-0# but should have done.
www.chessgames.com


Here is one where Morphy won by castling in a game against his father.
www.chessgames.com
kingseabass
28-Dec-19, 14:10

Excellent! TY!
lord_shiva
28-Dec-19, 15:17

5imon
Yeah, that queen sacrifice was inspired. And yes, white should have castled long to end the game, as that would have been an added touch of beauty.

If I have a choice of pieces with which to mate, I always go with the lower ranked piece. Just because. I win't take an extra move to do it, but yeah. Unless I could do it by castling.

Thanks so much for posting that!
pawntificator
28-Dec-19, 15:19

Not castling was definitely either a sign of respect or an insult.
servantmk1042
29-Dec-19, 03:46

Paul Morphy vs Alonzo Morphy
Looked like a great attacking game to me. Advantageous use of castling at the end. I can definitely here an " Awe nuts ! " by Alonzo early on as the king was exposed. It was like there was no time for effective development in the game. To see Alonzo's queen so helpless until almost the end was kind of amazing.
baddeeds
29-Dec-19, 15:51

Yes, I like to castle early in the game. And, I told one of my students that, unless there's an exception like winning material first or avoiding the loss of material, you should castle early. Although, you can castle right into the attack, but it doesn't happen so often. It became so interesting and heavily talked about in the KOH club that this is why I brought it up in TA's club, and Senior Master Joe agreed.
lord_shiva
02-Jan-20, 06:09

O-O-O#
While this is not that, it is pretty close and my team mate just finished it.

Mate in 16.

game
kingseabass
02-Jan-20, 06:47

Castling early is correct from a teaching perspective, and likely advantageous toward statistical win percentage. But I personally enjoy castling late, as it can bring about some interesting positions on the board that are fun to play.
servantmk1042
02-Jan-20, 10:03

o-o-o#
sweet. I looked this over. At first I saw the black knight move in # 10 as a questionable move to the rim. The more I looked , the more I realized he was just escaping. It was an interesting short game. Thank you
lord_shiva
02-Jan-20, 11:34

Rim
10. Nf3 Nh5 (the black Nh5 is the "rim" move) isn't really an escape so much as discovered "check" on the white queen. Black moved the knight to force the white queen to move (or hoped white wouldn't notice the queen was under attack).

The black knight wasn't under attack as the only piece that could take it was the white queen. It was covered by a pawn and the black's black bishop. [An important thing to keep in mind is that the white bishop is blind to the black squares, and vice versa].

Having said that I think 10. ... Nd7 would have been a little better, creating opportunity for discovered rook attacks once the rooks were moved to center files. Either square isn't ideal, and Nd7 blocks black's white bishop from the battlefield. In that sense Nh5 was the better choice.

Before pushing the queen, black should know where white will move it. Maybe black doesn't really want to force white into a better position, and should think of something else to play.

In this case white chose to delay response to the threat by countering with 11. Ng5+. Which was smart because moving the queen anywhere it wouldn't protect the knight opens white to:

11. Q somewhere BxNc3+
12. bxc3 Qxc3+ and white drops Ra1+.

This move at 10 seems like the cusp of the game, the hinge on which black pummeled white. But white should have interposed pieces to block black checks (which of course cannot be done with N+ but subsequent checks could have been blocked).

Hey, I annotated this game! My annotations are missing. Can someone delete annotated games?

shirlmygirl
02-Jan-20, 12:03

I have been trying a few games without castling early, as I usually do. Those games have not been going well for me so far.  
lord_shiva
02-Jan-20, 12:28

Oops
Apparently I had NOT annotated this game, but a different one instead.

Here is this one:

gameknot.com
bevo_xv
02-Jan-20, 12:32

Castling
Castling allows you to do two important things. The first is getting your King to safety. The second, which is extremely important, is to activate one of your Rooks. Rooks who sit on their starting squares are inactive pieces. The player with the most active pieces usually has an easier time controlling and subsequently winning the game. Moves that allow you to do two good things at the same time are the type of moves you want to make.

While castling is crucial, timing is everything. During the opening game, both players are fighting to control the center of the board. The only way to dominate or at least equalize control of the board’s center is to carefully but rapidly deploy your pawns and pieces to active squares, those that control the greatest amount of centralized board space. Therefore, before castling, beginners should ask themselves two questions.

The first question: Is my King in present or future danger? Present danger means that it’s your turn, your opponent’s pieces are in attack formation and ready to start checking your King immediately. If so, castling is a good idea. When I say future danger, I mean that an attack on your King is possible during the next one or two opposition moves. Advanced players have a bit more leeway regarding future danger and just when to castle. Future danger translates to “ within the next few moves can my opponent’s pieces attack my King, either forcing it to move, in which case my King loses the right to castle, or force me to weaken my position when I have to defend the King?” Of course, a potential immediate checkmate from the opposition within the next few moves should prompt you to castle if doing so saves the King! If the answer to this question is yes, then castle your King!

If you answered “no” to the first question, then its time to ask the second question, “are my pawns and pieces developed enough to control the board’s center more so than my opponent’s pawns and pieces? Most beginners consider castling before completing their development so the answer to this question is almost always “no.” Time to look at your development.

Many beginners learn the Italian Opening because it provides a relatively clear example of the game’s opening principles. For example, after 1.e4…e5, 2.Nf3…Nc6, 3.Bc4…Bc5, both players can castle on the King-side. This is where beginners get into trouble. They’ve been told by their chess instructors or by reading beginner’s books that you should castle early. Beginner’s take things literally, which often inspires them to castle as early as move four in the above opening move sequence. However, the opening is a fight for territorial control and the player that has it has a greater advantage. Advantages, both big and small, win games.

If your King is in no immediate danger, further development is in order. Keep developing pieces to active squares in order to shut down your opponent’s chance at staking a claim to those very same squares. In the
opening, it’s all about the center. Just because you’ve developed your minor pieces on one side of your King is certainly no reason to ignore the pieces on his majesty’s other side. Keep bringing those remaining minor pieces into the game. Pieces on their starting squares are not in the game. Those pieces are inactive and activity is the name of the opening game.

Then there’s the question of which side of the board to castle on. Beginners tend to castle King-side because its easier since you don’t have an additional piece to move (the Queen). However, Queen-side castling can be extremely effective. Why would you castle Queen-side? Here’s a good reason: If your opponent has aimed his or her forces at your King-side, castling there is going to put your King directly in the line of fire. Castling on the opposite side of the attack will force your opponent to redirect his or her pieces, which has a price. That price is tempo or time (wasting it). While your opponent is redirecting pieces, you can be strengthening your position or building up an attack against your opponent’s King. Don’t make your opponent’s job easier by castling into an attack or potential attack!

The next time you consider castling, ask yourself those two questions before doing so. If your do, you’ll know if you’re castling at the right time. Castling too early can make a position worse. Castling too late will send your King to an early grave.
lord_shiva
02-Jan-20, 12:35

O-O-O
There is also the disadvantage of that far flung, undefended pawn.
shirlmygirl
02-Jan-20, 20:11

CAPTAIN BEVO
Thank you for your very detailed and informative post, Ro. It took a lot of thought, effort and time to post it. Thank you to everyone who posted on this thread. I shall have to do some deep thinking about all of this. I wish you all a great 2020. Shirley.
Pages: 12
Go to the last post



GameKnot: play chess online, chess clubs, online chess puzzles, monthly chess tournaments, Internet chess league, chess teams, free online chess games database and more.