17

I've just lost an online game on time (lichess.com).

I had 3 pawns and my opponent had just a knight. I don't understand why it's not a draw.

  • 15
    It's not a draw because it's possible that you could lose if you played really badly and smothered yourself with the pawns. – Qudit Jul 28 at 20:15
  • 8
    Aside from the fact that you could theoretically lose such a position – it should be clear it can't be automatically drawn because you can obviously win with three pawns vs a knight, if you can manage to promote one of them! – leftaroundabout Jul 29 at 8:47
  • 10
    @leftaroundabout "6.9 Except where one of Articles 5.1.1, 5.1.2, 5.2.1, 5.2.2, 5.2.3 applies, 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." My reading of this is that if Player1 runs out of time, but Player2 cannot possibly checkmate Player1, then it's a draw, regardless of whether Player1 can checkmate. – Acccumulation Jul 29 at 16:42
  • 3
    Were you using pre-moves? If you really thought that your opponent could not possibly checkmate you, there's not much downside (unless you think you can win 3 pawns vs knight under time pressure). – Acccumulation Jul 29 at 16:45
  • 5
    @leftaroundabout It's irrelevant whether the guy who is out of time can checkmate the other player - because he's out of time... – Apollys supports Monica Jul 31 at 1:43
27

Draw with insufficient material is covered in article 9.6:

The game is drawn when a position is reached from which a checkmate cannot occur by any possible series of legal moves, even with the most unskilled play.

With a given material, it is possible to construct a checkmate (assuming your cooperation or horrible blunders), so it is not a draw.

  • 2
    Can you give an example of such a checkmate with a knight vs 3 pawns? – konsolas Jul 29 at 7:42
  • 4
    White: King on a6, knight on b6 Black: King on a8, (promoted) bishop on b8 – David Jul 29 at 10:12
  • 2
  • 1
    @konsolas White pawn on b7, black knight on a8, kings elsewhere. White: bxa (Q), then win the King-Queen-2pawns vs. King game – Cort Ammon Jul 30 at 14:54
  • 3
    @CortAmmon, I think we are looking for a checkmate from the side with a knight. – justhalf Jul 31 at 16:59
23

Worst-case scenario:

  • Upgraded all your pawns to knights

  • Your king is at Ka8

  • Your knights surround your king, so at Nb8, Na7 and Nb7

  • Opposition knight is at Nc7# - checkmate!

So it is indeed possible to lose, thus not a draw.

  • 4
    The advantage of this case s that ir does not depend on where the pawns originally are – Hagen von Eitzen Jul 31 at 6:22
  • 2
    The opposition king thought he was safe surrounded by his trusted guard of knights. – Lan Aug 1 at 14:20
9

From FIDE Laws of Chess (2018 version):

6.9 Except where one of Articles 5.1.1, 5.1.2, 5.2.1, 5.2.2, 5.2.3 applies, 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.

(The articles in chapter 5 all concern checkmate, stalemate, resignation or agreement to draw. I.e. any conventional consideration other than time that causes the game to be over.)

So your time ran out, and your opponent can checkmate you through some legal series of moves, which means you lost.

I do not know that Lichess folows the FIDE laws, but it seems like a reasonable assumption in this case.

  • 5
    I wonder what the effect would be if the rules were changed so that a player who noticed the opponent's time had expired could take over the opponent's pieces on his own time and win if he manages to achieve a checkmate before a draw occurs based on number of moves, repeated position, or mutual time expiration; good sportsmanship would imply that players should resign when their time expires unless they had reason to believe that the opponent would have some difficulty achieving checkmate under such conditions, but a player whose opponent does not resign in such circumstances... – supercat Jul 29 at 17:21
  • 2
    ...should simply checkmate the opponent without complaint (since any difficulty in doing so would demonstrate that the opponent's failure to resign was reasonable). – supercat Jul 29 at 17:21
6

As others mentioned, you cannot get the draw because you can still get checkmated.

In fact I have actually seen this kind of position get lost in practice, as follows:

You are on Ka8 and have just played a7 pawn (knight, rook or even bishop have very similar effect and could be obtained via promotion), opponents king is on Kc8 or Kc7 and opponent plays Nb6#

  • 3
    Already mentioned the other pieces, but updated to clarify they would need to come to promotion. As I emphasize the practical aspect I focus on pawn as you would rarely want to avoid promoting to a queen. – Dennis Jaheruddin Jul 30 at 10:54
  • 2
    Nice example. You can even set up the position so that the pawn move to a7 is the only move which avoids losing a pawn (N on c8 keeping an eye on a pawn at e6, K on c7 holding back a pawn on c5). In a blitz game with time running out, something other than a perverse desire to self-mate could lead to such a blunder. – John Coleman Jul 31 at 14:32
5

From the lichess.org FAQ:

In the event of one player running out of time, that player will usually lose the game. 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 (FIDE handbook §6.9).

Note that it can be possible to mate with a single knight or bishop if the opponent has pieces that could block the king.

I assume that the statement in the second paragraph is being used as the basis for ruling this as a timeout rather than a draw (and, as others have pointed out, this would be a correct call in this case.)

Perhaps it is worth noting that, in general, it is computationally infeasible to determine when a player cannot win, and furthermore, an automated system cannot yet be expected to identify all such cases that are obvious to an experienced player, so it must be expected that lichess will use tractable rules such as those quoted.

3

By USCF rules, you wouldn't have.King + Bishop and King + Knight are defined to be insufficient mating material unless that side can demonstrate a forced win (all forced wins are very short, so this isn't hard to do if it exists).

14E: Insufficient material to win on time:
The game is drawn even when a player exceeds the time limit if one of the following conditions exists:
14E1: Lone king
14E2: King and bishop or king and knight

The presumed rationale behind this is that avoiding checkmate from that material combination is so trivially easy that a win on time is the only realistic way to win.

  • 1
    I'm not doubting that that's what the rule says, but mating with K+N is possible if the opponent has pieces that can block king movement. – scatter Jul 31 at 19:15
  • 1
    @scatter There are checkmate positions in KNN vs K, but that's still declared a draw even in FIDE because they cannot be forced and are trivially easy to avoid. The same principle is (presumably) at play here. – eyeballfrog Jul 31 at 19:24
  • 1
    By what rule? The only insufficient material rule in FIDE I'm aware of is the one that says it's a draw if there's no legal series of moves that can lead to checkmate. – scatter Jul 31 at 19:26
  • 1
    @scatter Huh, I suppose FIDE doesn't prevent a player in KNN vs K from insisting on playing out the inevitable 50 move draw. Strange. – eyeballfrog Jul 31 at 19:35
  • 2
    So yes by USCF rules it would be a draw, by FIDE rules it's a loss. And Lichess follows FIDE rules. – user3445853 Jul 31 at 19:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.