I've been thinking when should a Dark Chess game be declared a draw.
Obviously, games might occur where none of the opponents has a reasonable chance of winning, but one of them stubbornly refuses to agree to a draw and insists on playing on. The problem is well known in Standard Chess, and therefore there is a number of rules to determine under which conditions a player can demand that a game be declared a draw.
There are three occasions when a Standard Chess game can be declared a draw against the will of one or both of the players:
1. Stalemate;
2. Upon request of the player who is on the move, if the same position is repeated for the 3rd time, it’s been the same player’s move each time and the rights of castling and en passant capture have not been lost;
3. Upon request of the player who is on the move, if each player has made 50 moves without a capture or a pawn move. In certain cases (e.g. King + 2 Knights vs. King + Pawn), the number of moves required is 75.
As you can see, in the first case, the game is always a draw, regardless whether or not the players want it, whereas in the last two cases, the game can only be declared a draw if one of the players requires it. (In chess software, rules 2 and 3 are often enforced automatically, but never in chess tournaments.)
I should also mention that some national chess rules include a fourth draw rule, according to which a game is declared a draw when a checkmate is theoretically impossible, e.g. King vs. King, King + Bishop vs. King etc. However, FIDE rules at the time of writing state that if one of the players objects, the game must be continued until the 3rd repetition rule or the 50 move rule becomes applicable.
Now, let’s look if and how those draw rules can be used for Dark Chess games. After that, let’s discuss the possibility of other means for defining draws.
1. The stalemate rule is appliable and sensible in Dark Chess, as it’s impossible to continue a game when the player who's on the move can’t make a move. Of course, in Dark Chess, a stalemate can only occur in a position where the player on the move really has no moves. If he can make a move whereby he would move into check, the position is a stalemate in Standard Chess, but not in Dark Chess.
Example: W: Kh1, Rh2, Ng1, Pe2, f3, g2, h3; B: Kh5, Pe3, f4, g3, h4
White’s move. Stalemate.
Another example: W:Kh6, Ph7, B:Kh8
Black’s move. Stalemate in Standard Chess, but no stalemate in Dark Chess, as Black has three legal moves: 1. ... Kg8, 1. ... Kg7 and 1. ... Kxh7.
2. The Standard Chess rule of three repetions can’t be used in Dark Chess, as neither of the players can see the entire board (except in rarest cases) and thus won’t be able to notice a repetition and won’t know that he has the right to require a draw.
It is thus totally inappropriate that one of the Dark Chess environments, namely IYT, has the house rule that whenever a position occurs for the third time, the game is declared a draw not upon request, but automatically. It’s doesn’t matter who’s move it is, and neither do the castling and en passant capture rights. It’s a uniform rule for all types of games played on the site. For Dark Chess, the rule is totally out of place, and not only for the reason stated above. The other reason is the following. As the players usually don’t see the location of all the opponent’s pieces, positions might occur where a player on the move has no way of telling whether the move he intends to make would lead to an automatic draw. That games might end up draws because of such lottery is unlogical and unjust.
My firm opinion is that the third repetition rule must not be used in Dark Chess.
3. The 50 moves rule appears suitable for Dark Chess with adjustments.
Firstly, the right to require a draw can’t be made dependent upon pawn moves, as not all pawn moves are visible to both players. Therefore, the same argument applies here as with the 3rd repetition rule – a game may not be declared a draw when players won’t know that it can happen.
Captures are, however, always visible to both players. So we could modify the Standard Chess rule like this: a Dark Chess game can be declared a draw when both players have made a certain number of moves without a capture.
As to the number, I think 50 moves is too much. The number 30 or even 20 seems more appropriate. Sure enough, you could say that if certain positions require 50 moves in Standard Chess to enforce a checkmate, it should take even longer to enforce a victory in Dark Chess. That is correct, but Dark Chess is a game of probability anyway. If you have a large material advantage, it might take you thirty moves to capture the opponent’s king, or it could take three moves. It's unpredictable. I think that even when a player has a huge disadvantage, he deserves a draw when he succeeds to hold on for a large number of moves. How large that number should be, remains open for discussion. My Dark Chess player intuition says that we should start with the rule of 20 moves (that is, 40 half-moves) without a capture and see how it will work in practice.
If anybody has had a game where 40 or more half-moves were made without a capture, but one player won by the end anyway, please send the game to me by email (darkchess@gmail.com), on Brainking website (username: Lawless) or on Itsyourturn website (username: Olavi the Lawless).
II. Now, let’s look at the draw rules in the existing Dark Chess environments.
1. IYT
IYT's Dark Chess rules [http://www.itsyourturn.com/t_helptopic2020.html#helpitem1275] might make you think that games can be declared draws arbitrarily by the webmaster. Such a rule, whereby a neutral third party looks at the position and decides whether or not he thinks it’s a draw, would seem perfectly reasonable at least for non-tournament games, considering that the game is relatively new, so there hasn’t been sufficient time to develop just draw rules.
However, as I found out by asking, the formulation on that Dark Chess rules page is misleading. Actually, the webmaster declares a game a draw only when a victory is theoretically impossible – like King + Knight vs. King in Standard Chess. If a victory is theoretically possible, players have to play on, even if it makes obviously no sense.
The problem is that there are no theoretical draws in Dark Chess (apart from stalemates which, I believe, will occur once in a century or so). Even king versus king doesn’t have to end up a draw. As neither of the opponents can see the other’s king, it’s not too difficult to move your king next to the other one and thus lose it. Look at the following position:
W: Ka1, B: Ka8
Both players know that the opponent has only the king left. Suppose both players know their location as well. Let’s assume White wants to avoid losing at all costs, and therefore only moves his king back and forth between a1 and b1. Black, however, wants to play aggressively, and walks around the board, trying to locate the white king. The game could go on like this: 1. Kb1 Kg7 2. Ka1 Kf6 3. Kb1 Ke5 4. Ka1 Kd4 5. Ka1 Kd3 6. Kb1 Kd2 7. Ka1 Kc1
At this point, White can calculate that the black king could have reached c1, so 8. Kb1 is no longer a safe move, and neither is 8. Kb2. However, 8. Ka2 isn't safe either, as the black king might just as well be on a3. So after at most seven moves, white can no longer avoid the danger of colliding with the black king. Of course, the two kings could run around chasing each other forever, but they could also collide after just 3 moves (1. Kb2 Kg7 2. Kc3 Kf6 3. Kd4 Ke5).
I believe that it is wrong to allow a dead even game to be decided that way. In the position above, each player clearly deserves a half-point for the draw. However, this Dark Chess game cannot be declared a draw using a Standard Chess argument that a victory is impossible, because a victory is possible.
All right, you might say, let’s simply make a rule that King vs. King is automatically a draw, not because it’s theoretically impossible to win, but because it’s so even that it would be unfair to let one of the players win.
Well, too bad that we can’t have even such a rule in Dark Chess. Suppose we had the rule that King vs King is always a draw. Now look at the following position:
W: Kg6, Qe7, B: Kg8
What would happen if white played 1. Qg7 ... ? His joy would be premature, as black would play 1. ... Kxg7 and the game would be a draw according to the King vs King rule. That would obviously be a nonsense.
So we see that in reality, there are no draw rules on IYT except Standard Chess draw rules which are by their nature unappliable for Dark Chess.
2. Brainking
The Brainking game rules don’t state anything about Dark Chess draw rules. So I asked the webmaster quite some time ago what the draw rules are and how they are being enforced. He replied:
“[---] no such rules are "enforced" in any chess variant on BrainKing. When a draw situation occurs and one of players notifies me about it, I declare a draw.”
From that explanation, I’m none the wiser. When a draw sitauation occurs... Well, that's exactly the question – what is a draw situation. Apparently, the webmaster hasn’t given much thought to draws in Dark Chess. That is understandable, as he has lots of more important things to do. But I think it's time the we the players start a discussion about draw rules on the forum and when we have worked something out, propose it to the webmaster.
Comments ()
Comments ()
Comments ()