Bug: Live Chess reports draw by insufficient material in winnable positions

I just played this game: http://live.chess.com/simple?v=2011072201 . I 'won' on time. However, to my great surprise the game reported a draw, insufficient material according to the Live Chess client. This is incorrect, for example white Kf1, Bf3 and black Kh1, pawn h2 is a mating position. There are many possible mates with the pawn promoted to a bishop or knight too. According to normal FIDE rules that qualifies as a win. I'm not sure if this bug extends to other positions with possible mates, but I would appreciate it if this could be resolved.

You are correct that should be a win if your opponent flags! as long as mate possible by an series of legal moves.

Thanks for those links! According to the second link USCF rules are differing from FIDE ones in this aspect: They'd rule this a draw. Assuming an American coding this that explains the origin of the 'bug'.The chess.com chess rules state:

Draws

Occasionally chess games do not end with a winner, but with a draw. There are 5 reasons why a chess game may end in a draw:

1. The position reaches a stalemate where it is one player’s turn to move, but his king is NOT in check and yet he does not have another legal move
2. The players may simply agree to a draw and stop playing
3. There are not enough pieces on the board to force a checkmate (example: a king and a bishop vs. a king)
4. A player declares a draw if the same exact position is repeated three times (though not necessarily three times in a row)
5. Fifty consecutive moves have been played where neither player has moved a pawn or captured a piece.

These are definitely the FIDE rules. Since we can therefore expect the Live Chess to abide by FIDE rules my bug still stands. Thanks for explaining the origins, though!

It isn't a bug.  It is an attempt to more closely follow FIDE rules, most specifically Article 10.2, the "opponent is not trying to win by normal means" provision where a player may make a claim to an arbiter that their opponent is merely trying to run their clock down and "not trying to win by normal means."  Since there is no arbiter, the site is programmed to claim certain positions like they one you mentioned when time runs out.

1...Kf4 2.Kg2 h3+ 3.Kf1 Kg3 4.Bh5 Kh2 5.Bg6 Kh1 6.Bf7 h2 7.Bd5# *
Pawn doesn't even have to promote!

Sorry forgot to ask were there increaments in this game if so then 10.2 doesn't apply. otherwise 10.2 applies

In theory it is possible for black to win, providing that white makes a series of extremely bad blunders, but it is not possible for black to force a win, and it is possible for white to end the game in a draw at any time simply by taking the pawn.

This a very simple position where even players new to the game understand that white must never allow black to promote the pawn.  Since white can only win on time, not by "normal" means, the comment by Omnipaul seems reasonable.

It is possible for both side to win if the other side flags the only question is if there are increaments in the game then like i said 10.2 doesn't apply. understanding has nothing to do with it in this case, unless there wasn't any increaments in which cases I would agree with you.

The correct ruling, according to Chess.com, is draw. In this scenario, a win is extremely impractical to expect, for the reasons that omnipaul and BigHickory gave. Since the final position is drawn, Chess.com's system made the correct ruling.

What where the time controls here?                                                                     

If there were increaments then it made the wrong ruling period.

if there were no increaments then it made the correct ruling as long as the draw is claimed by the player before he flags!

White Wins by

1...Kf4 2.Kg2 h3+ 3.Kf1 Kg3 4.Bh5 Kh2 5.Bg6 Kh1 6.Bf7 h2 7.Bd5# *