Chess324 – A Chess Variant to Reduce Draws
chessprogramming.org
chessprogramming.org
Though stronger players can no longer give 'presents', where you force a draw on a weaker opponent by ensuring both players end up with the same amount of points.
Is there amendment to chess that could work similarly? Nothing is coming to mind, but Chess is not my domain.
Hence the original 4.5 points komi about 100 years ago. The half point is the part that turns a draw into a win for white, that starts playing after black. But it's not strictly necessary: we could state in the rules that white wins draws.
Black started playing less conservatively and komi was soon adjusted to 5.5 points. It became 6.5 points at the turn of the century. It's 7.5 with Chinese and area rules.
This variant of chess seems to me potentially unfair to the player that gets a bad starting position, one that loses most games or that could even be winning but only with a narrow and difficult path to find. It could work in a tournament if all players have to play all the initial configurations. In thnk
But Chess960's castling mechanism is totally unintuitive. So, Chess 744 was born. The exact mechanic (rook moves to king and king swings behind it) from standard chess is used and all of chess 960's setups that had a rook and the king in a corner were removed. This website goes into detail. There is no better method, IMHO, of creating an intuitive chess variation that addresses this issue. It is very hard and unintuitive in Chess 960 to remember where to move your pieces to castle and whether or not it is even legal given the squares the king has to move through.
Basically, the rook, assuming the spaces are open and he and the rook havent moved, either moves adjacent to the king (if he isnt already adjacent) and the king moves two spaces around him.
Even on 9x9, like look at GoQuest for example, they must have thousands of game a day at 7 komi and ~nobody complains if there's a jigo, it just means the game was really close.
In chess the problem is just high level play has _so_ very many draws, so it can be worth trying to reduce that.
In go you get a draw once in a while. In chess you get a draw by default and have to really work to avoid it (assuming you're at a level vastly vastly higher than I am).
And then have either an imbalance in allowed time for each player or have them bid on it.
The mechanic for standard chess is:
(If Rook and King have not moved (and no funny business with pawns promoted to Rook), and none of the three spaces the King is involved with are attacked, and all the spaces between the King and the Rook are unoccupied) King moves two spaces, Rook moves to the space the king passed over. The order is important, at least in tournament play.
I guess something about being on the squares since the beginning of the game.
The king could still move the same two squares in the direction of the rook, but it might mean that the space is sometimes occupied by the rook or you may be simply jumping over the rook.
The way I'm interpreting this, 324 is different in that the rooks & kings are fixed, but the position is asymmetric.
Is it called 324 because there are 324 possible starting positions? Also, I'm inferring that this is popular in computer chess specifically, is there a reason for that?
[0]: https://en.wikipedia.org/wiki/Chess960
edit: there's a quote from the inventor in the wiki that answers my first question and sort of answers the second:
> This version has huge advantages over chess960. First, no special castling rules, any engine or GUI or human can play with no instruction after seeing the initial position. Second, since all but 18 of the 324 positions are asymmetrical, opening play should be much more interesting and complex. Third, the normal positioning of the rooks and kings and normal castling makes the game feel closer to normal chess. Fourth, matches of up to 648 games can be played with no repeat positions, generally enough for most purposes. Most important, no matter how many cores or how much time the engines get, there should be plenty of decisive games for the foreseeable future since many positions are at least not too far from the win/draw line.
Yes, each side has 18 possible setups. The bishops have 3 x 2 possible opposite-color placements among the 5 squares, and then the queen and 2 knights have 3 ways to occupy the remaining 3 squares. 18 x 18 = 324.
I also like that in this variant, castling works like normal -- that is one of the most unintuitive aspects of Chess960.
I have never played this variant of chess, but on the surface it seems that having both bishops on the same color would be a sizeable disadvantage.
The other randomized pieces (queens and knights) can get to any square, so having two knights start on dark squares, for instance, doesn't seem to really matter.
Or, more simply, player A is shown N random positions, picks one of them and lets player B pick the color.
The other idea works but is essentially just discarding all of the lopsided starting positions, in which case they might as well not be in the game.
Edit: A comment further down points out that only 18 of the 324 positions are symmetrical, so the vast majority of Chess324 positions could not happen in 960.
Secondly, at other levels one of the most important things is always whether a player can show enough sustained technique to convert some advantage and there’s always a chance for the person who is behind to swindle some draw if the guy who’s ahead can’t do this. This sort of a change would make that conversion technique somewhat irrelevant. The person ahead could just turtle up and guarantee themselves a win.
I know for a lot of people having draws in a game is very unsatisfying to some spectators[1], but the existence of draws in a game changes the strategy a lot meaning a player who is ahead has to press their advantage in order to secure the win and not allow a late heroic rearguard defence to secure a draw. That is (in my view) generally good for the game.
[1] I am a big fan of cricket and this is a common criticism of the game from folks who don’t like it.
It would be easy enough to test if one side having an advantage is common: have a chess engine play against itself many times using the same setup. The try a new setup. Find out what percentage of the time one side seems to have been given an advantage. (i.e. an advantage that's greater than the built-in one that white has.)
If there's a different reason this reduces draws, what is that reason? If it's simply reducing the reliance on book openings, that's fine, although there are other solutions for that. If there's something in a particular piece structure that reduces draws, what is it?
This is essentially just a subset of the Chess960 positions that don't require any modification to the castling rule.
How big is hard to know. Maybe it doesn't matter much.
Chess engines can figure that out empirically.
PS I checked, and this has several mentions before April 1!
Having unbalanced positions could even be better, as you'd be more likely to have a decisive game pair.
“All Kings, Rooks, Pawns are in their original locations as chess but other pieces are placed randomly in their first and last ranks, with no symmetry requirement between two sides, with the only restriction for bishops of each side must be on different colored squares.”
My point not being that that is a link straight to the exact thing you've described, but that if you ask that site for all the variants it has with a blank search [1], the first 500 variants on file (give or take some administrative garbage) gets you the variants starting with a number and up to "Avalanche Chess" alphabetically... that is, you can't even get all the way through the As before the search bails out.