'No castling' variant

Discuss anything you like about chess related matters in this forum.
User avatar
Christopher Kreuzer
Posts: 8781
Joined: Fri Aug 06, 2010 2:34 am
Location: London

'No castling' variant

Post by Christopher Kreuzer » Sat Jul 24, 2021 4:06 pm

Any views here on the recent Anand-Kramnik match with 'no castling allowed'?

A successful experiment: The Anand vs Kramnik "No Castling" match in Dortmund

Easier to play through than other variants, but am not sure what the point is. It can't be to discount computer training/use, as it must be trivial to get computers analysing games like this, or is it?

Roger de Coverly
Posts: 21291
Joined: Tue Apr 15, 2008 2:51 pm

Re: 'No castling' variant

Post by Roger de Coverly » Sat Jul 24, 2021 5:47 pm

Christopher Kreuzer wrote:
Sat Jul 24, 2021 4:06 pm
It can't be to discount computer training/use, as it must be trivial to get computers analysing games like this, or is it?
They seemed to mostly go for positions arising from 1. c4 c5 with the play being mostly normal with the occasional outbreak of weirdness. It's a variant where you might want to set an engine loose in search for opening variations where the absence of castling made a useful difference to the established evaluation.

There's an idea in the Hippo where (Black) plays all the g6, Bg7, b6, Bb7, d6,e6, Ne7, Nd7, a6, h6 stuff but then is faced with Be3 and Qd2. The solution, a King march e8,f8,g8 h7 thereby defending the h pawn and connecting the rooks.

User avatar
Christopher Kreuzer
Posts: 8781
Joined: Fri Aug 06, 2010 2:34 am
Location: London

Re: 'No castling' variant

Post by Christopher Kreuzer » Sat Jul 24, 2021 8:21 pm

Would you just insert dummy rook or king moves for both sides to remove the castling right if you wanted to do computer analysis? Trouble is, you can't do that from the starting position that easily.

Actually, it is easy. Both sides play their king knight out, move the kingside rooks to g1 and g8 respectively, then back to h1 and h8, and then put the knights back on g1 and g8 and after those 4 moves by each side you have the starting position for this variant.

Alternatively, you manually set up a starting position that was the actual starting position, and specify that the castling rights had been lost by both sides.

Wadih Khoury
Posts: 604
Joined: Sun Jul 12, 2020 8:14 pm

Re: 'No castling' variant

Post by Wadih Khoury » Sat Jul 24, 2021 8:38 pm

Christopher Kreuzer wrote:
Sat Jul 24, 2021 8:21 pm


Alternatively, you manually set up a starting position that was the actual starting position, and specify that the castling rights had been lost by both sides.
That should be the easiest way and take 2 seconds.


To Christopher's original question: it's because this variant is deemed to result in less draws as the kings are more exposed and entice players to go for nice attacks. As an additional bonus, it doesn't have centuries of theory and decades of computer analysis behind it.

NickFaulks
Posts: 8452
Joined: Sat Jan 02, 2010 1:28 pm

Re: 'No castling' variant

Post by NickFaulks » Sat Jul 24, 2021 8:41 pm

Wadih Khoury wrote:
Sat Jul 24, 2021 8:38 pm
To Christopher's original question: it's because this variant is deemed to result in less draws as the kings are more exposed and entice players to go for nice attacks. As an additional bonus, it doesn't have centuries of theory and decades of computer analysis behind it.
I would have put the motivations the other way round, but yes.
If you want a picture of the future, imagine a QR code stamped on a human face — forever.

Wadih Khoury
Posts: 604
Joined: Sun Jul 12, 2020 8:14 pm

Re: 'No castling' variant

Post by Wadih Khoury » Sat Jul 24, 2021 8:45 pm

NickFaulks wrote:
Sat Jul 24, 2021 8:41 pm

I would have put the motivations the other way round, but yes.
If theory was the primary concern, they would just play Chess960 (Fisher random).

NickFaulks
Posts: 8452
Joined: Sat Jan 02, 2010 1:28 pm

Re: 'No castling' variant

Post by NickFaulks » Sat Jul 24, 2021 9:34 pm

Wadih Khoury wrote:
Sat Jul 24, 2021 8:45 pm
If theory was the primary concern, they would just play Chess960 (Fisher random).
I don't think so. Kramnik is challenging all assumptions, and one of those is that 960 is the only approach which removes the modern game's dependence on theory.

It would be a fascinating experiment, which obviously won't happen, to remove castling from all chess and see how long it takes, with today's computer assistance, for opening theory to develop to the same level that it has reached in the traditional game. A year, perhaps?
If you want a picture of the future, imagine a QR code stamped on a human face — forever.

Roger de Coverly
Posts: 21291
Joined: Tue Apr 15, 2008 2:51 pm

Re: 'No castling' variant

Post by Roger de Coverly » Sat Jul 24, 2021 11:39 pm

NickFaulks wrote:
Sat Jul 24, 2021 9:34 pm
It would be a fascinating experiment, which obviously won't happen, to remove castling from all chess and see how long it takes, with today's computer assistance, for opening theory to develop to the same level that it has reached in the traditional game. A year, perhaps?
The evidence of the KvA match is that games involving 1. e4 would be out as too risky and (boring) positional stuff after 1. c4 the order of the day. Actually an impression I had of David Howell's play in his younger days was that he never castled if there was something else to do and this took place in all openings including 1. e4 ones .

E Michael White
Posts: 1420
Joined: Fri Jun 01, 2007 6:31 pm

Re: 'No castling' variant

Post by E Michael White » Sun Jul 25, 2021 8:18 am

Christopher Kreuzer wrote:
Sat Jul 24, 2021 8:21 pm
Actually, it is easy. Both sides play their king knight out, move the kingside rooks to g1 and g8 respectively, then back to h1 and h8, and then put the knights back on g1 and g8 and after those 4 moves by each side you have the starting position for this variant.
et QS

MartinCarpenter
Posts: 3041
Joined: Tue May 24, 2011 10:58 am

Re: 'No castling' variant

Post by MartinCarpenter » Sun Jul 25, 2021 9:30 am

NickFaulks wrote:
Sat Jul 24, 2021 9:34 pm
Wadih Khoury wrote:
Sat Jul 24, 2021 8:45 pm
If theory was the primary concern, they would just play Chess960 (Fisher random).
I don't think so. Kramnik is challenging all assumptions, and one of those is that 960 is the only approach which removes the modern game's dependence on theory.

It would be a fascinating experiment, which obviously won't happen, to remove castling from all chess and see how long it takes, with today's computer assistance, for opening theory to develop to the same level that it has reached in the traditional game. A year, perhaps?
Certainly not long now that the computers are actually very good at creating opening theory from scratch.

Nick Ivell
Posts: 1138
Joined: Wed Aug 31, 2011 6:33 pm

Re: 'No castling' variant

Post by Nick Ivell » Sun Jul 25, 2021 11:31 am

I sometimes wonder if castling is overrated (much though I like the right to do it). We have all 'castled into it'. I hardly ever leave my king in the centre - I suppose you could say I have been classically trained.

A word about connecting rooks. How much of this is simply for its own sake, or a way to avoid tactical weaknesses? Only occasionally does it improve the harmony of a position - a classic example is the sac on h7.

Wadih Khoury
Posts: 604
Joined: Sun Jul 12, 2020 8:14 pm

Re: 'No castling' variant

Post by Wadih Khoury » Sun Jul 25, 2021 11:43 am

Nick Ivell wrote:
Sun Jul 25, 2021 11:31 am
I sometimes wonder if castling is overrated (much though I like the right to do it).
Engines seems to have demonstrated the defensive value of castling: you can get between 50% and 90% of decisive games depending on sources.

User avatar
Christopher Kreuzer
Posts: 8781
Joined: Fri Aug 06, 2010 2:34 am
Location: London

Re: 'No castling' variant

Post by Christopher Kreuzer » Sun Jul 25, 2021 12:24 pm

E Michael White wrote:
Sun Jul 25, 2021 8:18 am
Christopher Kreuzer wrote:
Sat Jul 24, 2021 8:21 pm
Actually, it is easy. Both sides play their king knight out, move the kingside rooks to g1 and g8 respectively, then back to h1 and h8, and then put the knights back on g1 and g8 and after those 4 moves by each side you have the starting position for this variant.
et QS
Oops. Forgot that queenside castling still possible! :oops:

I wonder what the assessment would be in variants where only queenside castling is possible, or where only kingside castling is possible, or where White can only castle kingside and Black can only castle queenside, or where White can only castle queenside, and Black can only castle kingside?

Including the 'classical' game, I make that six variants on 'castling rights', is that correct? (All are sub-branches of the 'classical' game in the 'all possible moves' universe.)

Roger de Coverly
Posts: 21291
Joined: Tue Apr 15, 2008 2:51 pm

Re: 'No castling' variant

Post by Roger de Coverly » Sun Jul 25, 2021 12:32 pm

Christopher Kreuzer wrote:
Sun Jul 25, 2021 12:24 pm
I wonder what the assessment would be in variants where only queenside castling is possible, or where only kingside castling is possible, or where White can only castle kingside and Black can only castle queenside, or where White can only castle queenside, and Black can only castle kingside?
Kramnik popularised a variant without queens where White castled and Black couldn't. It was called the Berlin.

NickFaulks
Posts: 8452
Joined: Sat Jan 02, 2010 1:28 pm

Re: 'No castling' variant

Post by NickFaulks » Sun Jul 25, 2021 2:24 pm

Christopher Kreuzer wrote:
Sun Jul 25, 2021 12:24 pm
Including the 'classical' game, I make that six variants on 'castling rights', is that correct?
Sixteen I think. Each of the four rooks can be "on" or "off".
If you want a picture of the future, imagine a QR code stamped on a human face — forever.

Post Reply