Only legal move forces a draw and lichess says it is not draw

If lichess fixes this bug, then I will have to rethink my bullet endgame strategy :P

#21 Search can be computationally expensive: see my partial implementation at github.com/ddugovic/Stockfish/tree/helpmate

#23 I have not read cpp code for decades. I am not sure if the objective you are looking for is to evaluate each position to determine if it is draw due to dead position. The alternative that I propose is less ambitious: simply, when the game ends -by time out-, a single evaluation of forced movements is launched to determine if it is a draw or not.

My invitation in #13 is open to anyone... pick any language and try to solve this "evaluate" problem and I guarantee either it will be computationally expensive or wrong.

White timed out. Black has winning material. End of story.
For black to have drawn he should have made his move in time.

Sure we could make the rules more complicated, but that shouldn't be a goal.

We can analyze [FEN 4k3/8/8/2p1p1p1/1pPpPpP1/pP1P1P2/P7/4K3 w - 0] as a no-way to win situation. There are only 2 sides to move, 15 squares for the white king and 25 for the black, so a computer could quickly fill 2*(15*25) = 2*(20²-5²)=800-50=) 750 entries in its hash table and see there is no way for either side to win, but there will be other positions where its difficult for humans and/or computers to see so quickly.

Add a rule for one move ahead and a sore loser will just complain about a situation 2 moves ahead, etc.
Live with it!

(The clock is part of the game. If you don't like it, play correspondence.)

#25 @Toadofsky I have published a small proof of concept. You can write a FEN string and it evaluates if there is a chain of forced movements until the end of the game and the result of it.

chess-forcedmov-eval.web.app

(*) It considers a forced move when there is only one legally valid move.

Very interesting idea, while I know nothing about computer programming, I can see while people would be angry at this. However, I believe it is simply better to get over it and move on to the next game. We all lose tons of games for stupid reasons, some our fault, some not. In the long run, a couple games like this won't make that big of a difference in your rating.

Given the position alone, Black's last move wasn't necessarily a blunder, maybe there was a white queen on g4 and black has exchanged queens. (In the game given that was not what happened. But I'm taking the perspective of what could happen in a real game).

White has to recapture but times out.

I do agree it has to be a draw under the rules.

#27 Here's an "interesting" joke problem which your proof of concept almost solves (EDIT: after 1. d4 if Black times out, Black should lose; if in the initial position White times out it's a draw although White has a forced mate): K1k5/P1Pp4/1p1P4/8/p7/P2P4/8/8 w - - 0 1
en.wikipedia.org/wiki/Joke_chess_problem#Self-solving_problems

@Toadofsky in the first of those positions shown white times out. There is no legal way for black to checkmate white (as the only legal sequence leads to white checkmating black). Therefore it must be a draw. Even if in the last move black promoted to a queen.