I understand the logic behind the threefold repetition draw and the fifty move draw. In those situations, neither player has an advantage and the game is not progressing.
But in a stalemate one player is clearly in an inferior position. If a player can’t make a move without going into check, I say it should count as a loss not as a draw.
So is there some justification for making a stalemate a draw instead of a loss?
I always assumed that, like “en passant,” it was conceived by some cheater bullshitting the rules as he went along. 
As for the serious answer, I haven’t the foggiest. Practically speaking, it gives the guy with nothing but a king something to play for, rather than just tip his piece over and give up. Catch a blunder, let your opponent snatch defeat from the jaws of victory!
Chess endgames are very thoroughly studied. There are certain combinations of pieces that can force a checkmate, king-and-queen against a king, for instance. There are combinations which can not produce a mate, and I believe there are others which can bring about a mate only if the disadvantaged player makes a mistake.
That doesn’t explain why the rules are the way they are, but to change them as you suggest would have some strategic consequences. A player would not need as much of a material advantage in order to force a victory.
No, I can see the sense of the en passant rule. Pawns can only cover one rank. If opposing pawns could skip over the second rank with impunity then there would be situations where it would be impossible for one pawn to defend against another.
If I’ve got nothing left except a king and you’ve got a king and two bishops, this situation didn’t just happen. We’ve reached this point because you outplayed me and you deserve the win.
I don’t know about an outright loss, because the player in the stronger position has made a dumb move. In tournament play, a player gets 1 point for a win, nothing for a loss, and half a point for a stalemate. Maybe a better compromise would be to award three-quarters of a point to the player in better position and one-quarter to the weaker.
Did he? Or did you force that person into an unwinnable position? Everyone knows the rules going in so I don’t see why having an extra useless piece at the end of the game means you’re better.
King and two bishops * wins easily against a bare King; King, bishop and knight is a hard one to win; King and two knights can’t beat a bare King barring a blunder, and - echoing Snarky Kong - if we come down to that ending, either you have made some bad choices on the way or I cleverly defended myself to reach a position you couldn’t win.
If stalemate were not a draw, one pawn up would win the game every time. As it is, you have to be careful to reach a position where you can actually queen that extra pawn, because there is a stalemate defence in some King and pawn endings; homework assignment for the student. Similarly, there are some amazingly clever and beautiful stalemate defences to be found in endgame studies, and it would be a shame to lose them. Finally, the rule puts a brake on the schoolboy showoffs who like to celebrate a win by queening every pawn in sight, and often end up inadvertently stalemating despite themselves.
Wikipedia has an interesting discussion of the stalemate rule, and mentions that as late as the 19th century you’d lose in England if you stalemated your opponent!
This is all good stuff!
Yes, it’s the logical result.
Consider this position
White is a bishop and pawn up, but can’t win against a simple defence. White can force stalemate however - why should this be a win?
You can create stalemate positions where the side being stalemated has a material advantage.
Why should White win this? He only has a lone King!
There are also endgame positions where one side has a forced win, but it takes over 50 moves. (The record is 223 moves with King, Rook + Bishop v King and 2 knights.)
There are many beautiful endgame studies where a precise line of play leads to a stalemate. There are no dumb moves here - study by V + M Platov 1907:
Apparently www.apronus.com doesn’t like two chess positions in one SDMB post.
So here is a correction:
You can create stalemate positions where the side being stalemated has a material advantage.
Why should White win this? He only has a lone King!
I know bugger all about chess but if it is an analogy of warfare then I can understand why such a rule is included. It makes sense that on the battlefield you can come up with unwinnable positions even in the face of superior numbers and the stalemate rule handily reflects that.
And did not WOPR state “the only winning move is not to play”? (Though he also, spookily, suggested " a nice game of chess " as an alternative…wheels within wheels my friends)
First off, I’m not a chess player, and I can’t actually see all those moves that would have to be made to actually make it to stalemate. But I fail to see why one side having more pieces than the other should matter in winning versus losing. In a system where a stalemate is a win, black’s having gotten himself in that situation is a tactical blunder.
It seems like the real answer is that that’s the way it’s been done, and changing it now would cause too much disruption in what is supposed to be a determinate game.
First off, I’m not a chess player, and I can’t actually see all those moves that would have to be made to actually make it to stalemate. But I fail to see why one side having more pieces than the other should matter in winning versus losing. In a system where a stalemate is a win, black’s having gotten himself in that situation is a tactical blunder.
It seems like the real answer is that that’s the way it’s been done, and changing it now would cause too much disruption in what is supposed to be a determinate game. To challenge this “logically”, I would need some sort of statistical evidence that the game is more balanced this way, with the same number of possible wins for white and black. This would be done by the best computer in the world playing both sides in a large number of trials. Unless, of course, there’s a way to control for how good someone is, in which case you could actually use real results.
This. This ads an element of challenge and uncertainty to games that would otherwise be essentially over, especially at the highest levels: it’s not uncommon for a grandmaster gaining a pawn’s advantage over another GM in the middle game, but the question of whether he can maintain that advantage into the endgame is less iffy than the question of whether he can convert that pawn into a win when he gets there, and there’ll be a lot of play aimed at having the best position to exploit (or prevent the exploitation of) that one-pawn advantage as the remaining pieces get traded off. Without the possibility of stalemate counting half a point, there’d be no need for any of that; like you say, up one pawn in the endgame would straightforwardly translate to a win, absent a blunder.
It’s the right constraints that make a game interesting, and in chess, stalemate=draw is one constraint that adds a lot to the game.
If a player deserved to win he would have won.
Perhaps because Black has played himself into a hideously bad position?
From the Micro standpoint of a particular circumstance, it makes sense that the player who can’t make a move without losing his King is the one who deserves to lose the game. This is different than having a player in a “weaker” position that can endlessly avoid checkmate. The second situation you agree to call the game because nobody wants to play a game forever with no resolution. The first situation you somehow decide that the player doesn’t have to move anymore, and end the game.
From the Macro standpoint, the playability and competitiveness of the game might be fouled up by changing the stalemate rules, so it makes sense to leave them be.
Looking at the thread, it’s interesting that the more experienced players are satisfied with the stalemate status quo.
OK, it’s stated above that stalemate is a ‘tactical blunder’.
I’ve already given one example where this isn’t true and here’s the simplest case:
This is a book ending (I use it to teach beginners.)
Although White (to move) is a pawn up, he can’t win.
He has the choice of either stalemating Black (1. h6 Kg8 2. h7+ Kh8 3. Kh6), or forcing Black to stalemate White (1. Kh7 Kf7 2. h6 Kf8 3. Kh8 Kf7 4. h7 Kf8.)
So under the proposed amendment, White either wins or loses - despite the fact that there’s no possibility of checkmate.
It doesn’t make sense to me that a player who can’t checkmate can win.
Also there’s no likely ‘disruption’ to what is indeed a determinate game.
As shown by this thread, experienced players know perfectly well what the full effects of changing the stalemate rule would be.
The major effect of counting stalemate as a win would be to make a large number of endgames which are now only a small advantage for one side into a large or winning advantage. This would count not only for endgames that have reached pawn vs king, etc, but the endgames leading up to them, and where one side has (what is now) a small strategic advantage. I presume this would mean strategic disadvantages in general would become riskier (one cannot cop out into a worse endgame, but into a lost endgame), and I presume this would restrict one’s options in the middle game.
What does this mean? Does it mean that if the rules were different, the player would be losing? But the rules aren’t different, so if that’s what you mean I don’t know why you think the “inferiority” of the position has any significance.
But is it not what you mean?