You are on the backup site for Any posts, moves, or other changes you make here will not be permanent, because the pages and database from the main site will be backed up here every midnight EST. Additionally, things may not be working right, because this site is also a testbed for newer system software. So, if you are not here to test, develop, or merely read this site, you may want to change .org to .com in the navigation bar and go to the main site.

The Chess Variant Pages

This page is written by the game's inventor, John Lewis.

Castling in Chess960 

An appeal for simplicity

By John Kipling Lewis


Bobby Fischer- "That's basically it. You can learn the rules in two minutes. It's a great game, and can become the standard for chess."

In 2003 David A. Wheeler contacted many active in Fischer Random Chess to determine the exact castling rules, including Eric van Reem, Hans-Walter Schmitt, and R. Scharnagl. All agreed that there must be vacant squares between the King and his destination except for the participating Rook, clarifying the castling rules for Chess960.

The intention of this article is to promote the idea that the castling rules as originally presented by Bobby Fischer for Chess960 are flawed from a game design standpoint and that changing these rules would benefit the advancement and acceptance of Chess960 by both the Standard Chess and non-chess playing communities. Obviously there will be resistance by some in the current Chess960 community, as any change might be seen as suspect. 

First let's review castling rules for Fischer Random and Standard Chess.

Fischer Random castling rules. (Quoted from the Chess Variant Pages.)

"Of necessity, in Fischer Random Chess the castling rule is somewhat modified and broadened to allow for the possibility of each player castling either on or into his or her left side or on or into his or her right side of the board from all of these 960 starting positions. However, after "a"-side castling, the King and Rook find themselves on the usual squares: King on c1 (c8) and Rook on d1 (d8), after "h"-side castling : King on g1 (g8) , Rook on f1 (f8). Sometimes castling looks odd in Fischer Random Chess: e.g. when your King is on e1 and a Rook is on f1, you only have to move your King to g1 ("King-move-only" castling)."

A much longer explanation can be found at, from the original Fischer Random text.

FIDE rules for castling. (quote from

"Castling This is a move of the King and either Rook of the same colour on the same rank, counting as a single move of the King and executed as follows: the King is transferred from its original square two squares towards the Rook, then that Rook is transferred to the square the King has just crossed."

Please note that for brevity I have excluded from both the conditions under which castling is not allowed. I will include these for the offered variation at the end of this article.

I think it's important to review a history of castling to understand part of the impetus for this change. Before the 1500s, in order to speed up the game, the 'King's Leap' allowed the King to move two spaces as his first move (jumping one square). Castling become a single move of a rather common two move opening where the Rook was moved next to the King and the King would use his King's Leap to move on the other side of the Rook on the following turn.

There are two points to be made here. The first is the King's ability to 'King's Leap' two spaces as his first move was changed to castling as a single move. This is a balancing of power from a game design point of view, because the King can no longer move forward or diagonally two squares on the 'King's Leap', but must instead interact with a Rook in order to gain the advantage of moving two squares as his first move. The second point is that the Rook always ends up in the square leaped over by the King. (further research available at

In examining Bobby Fischer's rules for Chess960's castling we see that it totally ignores the origins of castling in favor of appealing to chess players who are already familiar with the castling positions that result from Standard Chess. Unfortunately the system is confusing to newer players who aren't as familiar with Standard Chess. The final resting squares of the King and Rook feel arbitrary to new player who may not have played enough Standard Chess to intuitively remember them.

It seems that simplification of the castling rules for Chess960 could help promote the game for beginners, streamline the rules and reconnect the game with it's historical roots. To that end I suggest the following castling rules for Chess960:

Castling: This is a move of the King and either Rook of the same colour on the same rank, counting as a single move of the King and executed as follows: the King is transferred from its original square two squares towards (or over) the Rook, then that Rook is transferred to the square the King has just crossed (if it is not already there). If the King and Rook are adjacent in a corner and the King can not move two spaces over the Rook, then the King and Rook exchange squares.

(1) The right for castling has been lost: 1. if the King has already moved, or 2. with a Rook that has already moved

(2) Castling is prevented temporarily: 1. if the square on which the King stands, or the square which it must cross, or the square which it is to occupy, is attacked by one or more of the opponent's pieces. 2. if there is any piece between the King and the Rook with which castling is to be effected.

This rule is much cleaner than the current set of rules and is the most obvious to new players who are introduced to Chess960 as beginning Standard Chess players.

As an exercise for the reader, I suggest finding a new Standard Chess player, explaining the rules of Chess960 to them without revealing how castling works, and then ask them to guess how castling would work in the variant you have just shown them.