4 ms·

Having a winning strategy for black seems impossible though; the discussion is whether white have it or is it a forced draw.

Someone 4y ago
Why would that seem impossible? One could argue that white has to open up his defenses first, allowing black to pick the best response to that. In general, there’s no guarantee of first mover’s advantage. For example, Hexapawn (https://en.wikipedia.org/wiki/Hexapawn https://en.wikipedia.org/wiki/Hexapawn) is a win for black on some boards (https://web.archive.org/web/20050330222720/http://www.chessvariants.org/small.dir/hexapawn.html https://web.archive.org/web/20050330222720/http://www.chessv...). Versions that are more complex than chess and are a win for black may exist.
- SamReidHughes 4y ago
  White could start with a knight move. Unlike Hexapawn, in chess, white moving first provides an increase in optionality for white, and the board positions are not forcing white on some losing track, because white is free to "burn" a move with many pieces by taking 2 moves to get to the same place they could with 1 move. White can use two bishop moves, or a pawn push to the 3rd rank followed by the 4th rank, to turn around the initiative early in the game.
  - SamBam 4y ago
    That would give black two moves in a row, which is a different scenario, and might still be optimal for black.
    - SamReidHughes 4y ago
      Black doesn't know it's getting two moves in a row when it makes the first move. And on the second move, it can only take advantage if white's first move reduces white's optionality in a specific manner, e.g. by blocking another piece essential for response to a certain tactic, which by construction, would mean it wasn't really an opportunity to transpose the game to hand over the initiative.
      - SamBam 4y ago
        Not sure I follow. The scenario we're positing is that perhaps, in fully-solved chess, white is actually in zugzwang, where every initial move is actually bad for them. If there were some way for white to hand the first move to black, then clearly this would be the best solution. But all the take-back solutions (e.g. move the knight forward and back again) actually give black two moves, which is a different situation. And the stutter-step solution of white moving a pawn only one space instead of two doesn't work because that's simply refuting the premise that white is in zugzwang.

Someone 4y ago

Why would that seem impossible? One could argue that white has to open up his defenses first, allowing black to pick the best response to that. In general, there’s no guarantee of first mover’s advantage. For example, Hexapawn (https://en.wikipedia.org/wiki/Hexapawn https://en.wikipedia.org/wiki/Hexapawn) is a win for black on some boards (https://web.archive.org/web/20050330222720/http://www.chessvariants.org/small.dir/hexapawn.html https://web.archive.org/web/20050330222720/http://www.chessv...). Versions that are more complex than chess and are a win for black may exist.

SamReidHughes 4y ago
White could start with a knight move. Unlike Hexapawn, in chess, white moving first provides an increase in optionality for white, and the board positions are not forcing white on some losing track, because white is free to "burn" a move with many pieces by taking 2 moves to get to the same place they could with 1 move. White can use two bishop moves, or a pawn push to the 3rd rank followed by the 4th rank, to turn around the initiative early in the game.
- SamBam 4y ago
  That would give black two moves in a row, which is a different scenario, and might still be optimal for black.
  - SamReidHughes 4y ago
    Black doesn't know it's getting two moves in a row when it makes the first move. And on the second move, it can only take advantage if white's first move reduces white's optionality in a specific manner, e.g. by blocking another piece essential for response to a certain tactic, which by construction, would mean it wasn't really an opportunity to transpose the game to hand over the initiative.
    - SamBam 4y ago
      Not sure I follow. The scenario we're positing is that perhaps, in fully-solved chess, white is actually in zugzwang, where every initial move is actually bad for them. If there were some way for white to hand the first move to black, then clearly this would be the best solution. But all the take-back solutions (e.g. move the knight forward and back again) actually give black two moves, which is a different situation. And the stutter-step solution of white moving a pawn only one space instead of two doesn't work because that's simply refuting the premise that white is in zugzwang.