The site has moved to a new server, and there are now some issues to fix. Please report anything needing fixing with a comment to the homepage.

The Chess Variant Pages

[ Help | Earliest Comments | Latest Comments ]
[ List All Subjects of Discussion | Create New Subject of Discussion ]
[ List Latest Comments Only For Pages | Games | Rated Pages | Rated Games | Subjects of Discussion ]

Comments/Ratings for a Single Item

Later Reverse Order Earlier
The Fair First Move Rule in Chess. Every turn you flip a coin to see who goes first.[All Comments] [Add Comment or Rating]
(zzo38) A. Black wrote on 2011-02-28 UTC
Another idea is as follows: Win by capturing opponent's king, not checkmate (and castling in/through/out of check is permitted). At the first turn of the game you toss a coin to determine who goes first. And then on each turn, you roll a dice (decide before the game starts how many sides, you might use d10 or d12 or d20), and if you get 1 then you miss your turn. If your turn is not skipped and you have no legal omve, the game is drawn. Repitition of moves does not end the game. Perpetual check is not a drawn game; possibly the checking player will win or the other player will be able to stop it.

Rich Hutnik wrote on 2010-01-02 UTC
I am not really a fan of this variant, as I feel it adds too much randomness to chess. However, I believe if you get rid of check/checkmate, and replace it with capturing the enemy king, there isn't an issue with check.

H. G. Muller wrote on 2009-12-31 UTC
What you are asking for is not fairness, but stability. The two have nothing to do with each ther. And indeed, Chess is by nature an unstable game, where advantages tend to grow with the move number. The quality of earlier moves has a higher weight in determining the outcome than the quality of later moves. Take away someone's Queen on move 3, and 'restoring' the balance by taking away the opponent's Queen on move 33 will almost certainly leave him in a lost position, as the opponent hs been hammering on him with an extra Queen for 30 moves.

So in practice the game will be decided in favor of the first player who has a stretch of luck in the first-move lottery. But even if the first coin flip would fully determine the outcome, it would still be a far game. Just a less interesting one.

John Smith wrote on 2009-12-31 UTC
Sure, it is fair theoretically, but how about empirically? Does the game tend towards equality, or are things more skewed by this flipping? My contention is that the game tends toward inequality, because the first few moves, if they give advantage to one player, that advantage will not be reversed because they have the initiative. Even if an equal amount of extra moves are given to the other player, it is worth less because it comes later. To equate the first advantage, it would require more consecutive moves which are less likely to occur. Then again, positions can make moves more useful at certain times which mediates bias. To what extent I am not sure.

H. G. Muller wrote on 2009-12-28 UTC
Isn't that obvious from the symmetry? The rules make no distinction anymore between black and white, so they must have equal opportunity.

Of course you could simply acheive that by only flipping a coin to decide who starts.

Fairness is of curse not a mathematically defined concept; one could argue that a game that involves chance elements is intrinsically less fair than one that doesn't. If under the coin-flip rule I play two games against the same opponent, he might very well be lucky, and get the first-move advantage twice. Is that fair, just because I _could have_ gotten that advantage as well if I had been more lucky? Or is it just that it _could have been fair_ if I had been less unlucky (but is quite unfair in practice). Personally, I think it is much more fair to simply play two games with alternating colors, where white starts. Or in a tournament with only a single game per pairing, alternately give a player white and black.

John Smith wrote on 2009-12-28 UTC
I am no mathematician, but I am interested in the actual veracity of this claimed 'fair first move'. Despite whatever merit this may have as a good game, or to introduce randomness into Chess, is it really fair?

Anonymous wrote on 2003-04-04 UTC
The realtime chess game <a href=''>Kung Fu Chess</a> also provides a fair first move rule. In that game, both players move at the same time during the entire game. There are no turns. Pieces take time to move to their destination. You should check it out at <a href=''>Kung Fu Chess</a>

Anonymous wrote on 2002-06-06 UTCExcellent ★★★★★

gnohmon wrote on 2002-05-25 UTC
> This also deals with the discovered check problem in multiplayer

Not to mention variants where both (or many) sides move simultaneously and
then the moves must be resolved. For example, the multiplayer game
Diplomacy, which IM Bela Toth introduced me to, has everybody move
independently and then conflicts are resolved by complicated rules; and for
contrast, the multiplayer game Risk (and thanks to Bela I can win this
game) is completely turn-based.

These things are complex, and this is unfortunate in a way because we
susally seek simplicity; but it may be necessary.

Synchronous Chess is an ancient chess variant where both sides move at once
and the conflicts are somehow resolved. 

Realtime Chess is a first-draft inspiration of the moment by me, and the
idea is that you just move your pieces without waiting for the other guy to
move. Your clock is *always* running. ((I see no way to make rules that
keep it from bing a simple stupid clickfest.))

Peter Aronson wrote on 2002-05-21 UTCExcellent ★★★★★
<blockquote> 'The only other rule I can think of is that if it's your move and the other player is already in check, you cannot capture the King but you can play any other legal move you choose' </blockquote> This also deals with the discovered check problem in multiplayer variants: that is, when player A moves a piece that was blocking player B's piece, so now player B's piece attacks player C's King, and the turn sequence is A-B-C so player C never gets a chance to move out of check before being captured.

Mike Nelson wrote on 2002-04-18 UTCGood ★★★★
I've played it and I agree with Ralph--the best way to introduce randomness
into Chess.

A checkmate rule I find satisfactory:
If a player is mated by a single move, the game is over.
If a player is mated by two consecutive moves, if taking two consecutive
moves would relieve the mate, the mated player wins the next toss
automatically and can play two moves.

For stalemate the rule is the same.

11 comments displayed

Later Reverse Order Earlier

Permalink to the exact comments currently displayed.