Constructed drawn games

[Christopher Kreuzer]

Is it possible to construct: (a) any game) and (b) a sensible game - where you reach the position of bare kings (or any other position where the game is instantly declared drawn) in the least possible number of moves? Or to put it another way, what is the quickest way you can reach that type of draw? I don't think it can be done sensibly in less than 40 moves, but I am sure it can be done in any number of silly ways. I will tentatively define "sensible" as moves that keeps the position evaluation (by a strong computer engine) within one pawn deficit for either side...

[IM Jack Rudd]

As far as I know, fastest known stalemate is move 10, fastest known bare kings is move 17, fastest known fivefold repetition is move 8. Don't know about sensible ones.

[Christopher Kreuzer]

Thank you Jack. Found a discussion about this here:

https://chess.stackexchange.com/questio ... ng-endgame

Interesting!

https://www.wismuth.com/chess/chess.html

In December 2003, I wrote a program to search for chess games that are completely determined by the last move. Among other chess problems, the search uncovered this brain-teaser: Black mates White on Black's 5th move by playing a rook, but the rook itself does not attack White's king. How did the game go? (the solution is unique).

And many other fascinating results, positions and 'proof' games.

[Christopher Kreuzer]

Oh dear.

It turns out that Americans Stuart [Rachels], an IM and Alex [Fishbein], a GM are part of a group of very strong chess players led by Stuart who set out to find a (not necessarily unique) shortest game for each of the 840 possible mating moves without disambiguation.

From Game-determining Moves by François Labelle.

I don't even want to think what the 840 possible mating moves are... (a rabbit hole I am happy not to explore). But it is strangely reassuring that the ability and motivation to explore such problems exists.

[Richard James]

If you're interested in synthetic games you should read this paper by George Jelliss.

https://www.mayhematics.com/p/syntheticgames.pdf

[Christopher Kreuzer]

Thank you, Richard. That is very interesting.

I am now rather engrossed in Labelle's page on the longest possible chess game:

https://www.wismuth.com/chess/longest-game.html

The longest possible chess game is 8848.5 moves long.

[Shh! Don't tell Peter Lalic!]

[IM Jack Rudd]

I think I've solved the proof game mentioned earlier: 1.f4 h5 2.Kf2 Rh6 3.Kg3 Rf6 4.Kh4 e6 5.Kg5 Rh6#. Quite a tricky one.

[LawrenceCooper]

I think I've solved the proof game mentioned earlier: 1.f4 h5 2.Kf2 Rh6 3.Kg3 Rf6 4.Kh4 e6 5.Kg5 Rh6#. Quite a tricky one.

If that is the answer, what on earth was the question?

[Christopher Kreuzer]

"Black mates White on Black's 5th move by playing a rook, but the rook itself does not attack White's king."

(It was always going to be a discovered check, which helps.)