Was walking through primary school and saw an 8*8 chessboard on the ground, with squares big enough for children to stand in, and immediately thought how useless that was.

Then paused.

Chess requires 32 pieces, getting 32 children to participate in a game of wizard chess would be impossible. Checkers 24 children, ditto. Go, Connect 4, etc. would all require too many children.

What would be the minimum number of children required for an interesting 2 player game on an 8*8 board? What about a chess endgame? Would 5 children suffice for an interesting chess endgame? Would 4?

what makes you think it was for children? I could have been for a large garden set.

I think you’re pessimistic about how many kids you could get to be involved in a big chess game. Schoolkids by and large do what they are told.

dv said:

I think you’re pessimistic about how many kids you could get to be involved in a big chess game. Schoolkids by and large do what they are told.

A lot of children can play chess if given the opportunity to do so.

I think 4 children would do.

King, knight and bishop against king.

50 moves allowed (single king keeps score).
50 moves or stalemate and the single king is deemed to win.

Without the 50 move rule, here’s a mate in 545 moves starting with just 7 pieces.
fun-with-chess/longest-mate-official—-mate-in-545

mollwollfumble said:

I think 4 children would do.

King, knight and bishop against king.

50 moves allowed (single king keeps score).
50 moves or stalemate and the single king is deemed to win.

Without the 50 move rule, here’s a mate in 545 moves starting with just 7 pieces.
fun-with-chess/longest-mate-official—-mate-in-545

On watching that 545 move win by white I can start to see some of the strategy.

White starts out in a very bad way, with the king boxed in by the rook in the upper left corner.

It escapes that trap moving to the bottom left corner where it is not quite so boxed in.

By move 200, the strategic positions look about equal with both kings in similar difficulty and both sides are controlling the same area of board.

The white king escapes to near the bottom right corner, where it is protected from the rook by its knight. The black king, by contrast, is sort of contained in the left half of the board, and black pieces are being pushed closer and closer in to their king.

The white king escapes the bottom right corner for the top right, amply protected this time by the queen. By this time white controls at least two thirds of the board.

The black king is forced into the top left corner on move 400 and never escapes, his pieces are pushed closer and closer until the only option is trading the rook for a knight, and the end comes quickly after that, in 32 or so moves.

Aha. This is what I was looking for.

A table of win probabilities for up to 5 pieces.

http://www.maths.usyd.edu.au/u/thiel/misc/endgames/

There are enough interesting options even with only 4 children. Especially if you count a stalemate as a win for black.

mollwollfumble said:

Aha. This is what I was looking for.

A table of win probabilities for up to 5 pieces.

http://www.maths.usyd.edu.au/u/thiel/misc/endgames/

There are enough interesting options even with only 4 children. Especially if you count a stalemate as a win for black.

Nearly 100% draw. One N or B on one or both sides. NN on one side vs nothing, N, B, R.
One exception is R vs NN, where R going first can force a 6.3% win in up to 41 moves.

Nearly 100% win. King loses against Q,R,BN,BP,NP,PP. Q wins against R,B,N,P. Q loses to QR, R loses to RR. B, N and P, individually, lose to RR,RB,RN,RP.
Exceptions are that first play of R vs Q can win 29% of the time, Q vs QR 24% and P vs all opposition on rare occasions.

Other games with 3 (lone pawn), 4 (2 bishop, Q vs Q, R vs R,B,N or P, or P vs P,N or B), and 5 pieces may be considered interesting.