I had enough materia to win, despite this, Lichess gave ½-½ when my opponent run out of time. He had an a-pawn and a King, I had a Knight and a King.
Imagine this position:
(I'm White)
Black: King on a1
pawn on a3
White: K on c1
knight on d4.
Black to play: he plays the only move: a2
White to play Nb3# winning.
http://en.lichess.org/a3gUVmce9mK5There's been very long discussions about this previously, and we've decided that lichess will decide as arbiter that having only a knight is insufficient mating material as it's only edge cases and self-mates that will result in a win.
Black king is not on A1, and the black pawn isn't on A3. Justifiation rejected and case closed.
It's true that's a checkmate, but I think you'd need more than 3+0 to get into that position from your final position. If you had more than 3+0, you'd be able to maybe get into that position (which requires some assistance from the opponent, of course - I think they would often willingly give up the pawn for the draw than garner a loss).
So yes, it is a technical win if your opponent is willing to be corralled into the corner, but a draw is more fair for the reasons Clarkey expressed. Benefit of the doubt is given to the defender in this case.
I did not say that Black king IS on a1, I said imagine I manage that position. I mean that I do theoretically have winning chances so Black should lose on time instead of receiving ½-½
"I do theoretically have winning chances"
Not in practice though, your position is drastically different from your hypothetical situation - which was an edge case, and so does a million other QN vs QP positions.
Surely the official rules would solve this?
I agree with you, as I am a huge proponent of using the correct FIDE rules, which would give you a win in this case. I also submitted an issue on Github about this a couple of months ago:
github.com/ornicar/lila/issues/1449. The "not in practice" argument doesn't really hold; the chess rules are theoretical, and there are already an infinite amount of positions where a player doesn't have winning chances in practice, yet does he win if his opponent runs out of time (random example:
http://en.lichess.org/x2PyYB1swky3). And people do manage to lose even the most unlosable positions.
It's obvious this case should be a win. There is a theoretical chance of checkmate, so the FIDE rules leave no doubt whatsoever - the player with the knight should win when the player with the pawn runs out of time.
The FIDE rules of chess clearly states:
§ 6.9
If a player does not complete the prescribed number of moves in the allotted time, the game is lost by that player. However, the game is drawn if the position is such that the opponent cannot checkmate the player’s king by any possible series of legal moves.
In other words, if there is any way to checkmate, no matter how unlikely or how bad your opponent has to play, you win. It also does not matter if you have almost no time left.
This would mean that these positions should be a win if your opponent loses on time:
(K=King,Q=Queen,R=Rook,B=Bishop,N=Knight,p=pawn. List might not be complete)
p vs anything. ( ex: Kp vs KQQQQQQQQQ)
B vs N, opposite colour bishop
N vs N,B,R,p (pawns can under promote)
R vs. everything
Q vs. everything
Unless of course the end position is a stalemate
Whether a player should win on time is not a matter of subjective interpretation. If a position can be won and your opponent loses on time, you win. It really is that simple.
I know that most chess sites have not yet implemented the instant win if your opponent loses on time in ex. knight vs pawn. But after my understanding this has been a coding issue, and in those cases all you needed to do was ask an admin to change the result.
Regards
Lasse
