chess is a draw but in how many moves?

[Uri Blass]

I think that we can define it by the following game:

Have a rule to decide about drawn games.

Draws with more than X plies are considered to be a win for white
Draws with less than X plies are considered to be a win for black
Draws with exactly X plies are considered to be a draw.

Assuming the theoretical result of chess is a draw It is obvious that there is exactly one X when the theretical result of the new game is also a draw and the Question is what is X

It is possible to ask the same question about what is Y that gives theretical result of a draw when
Draws with more than Y plies are considered to be a win for black
Draws with less than Y plies are considered to be a win for white
Draws with exactly Y plies are consdered to be a draw.

I guess that Y<X because white can force a draw faster than black.

[towforce]

If chess is a draw, which it very likely is, then it's a draw on move 1.

[Uri Blass]

If chess is a draw, which it very likely is, then it's a draw on move 1.

No.
To get a draw you need to play.
You cannot claim draw because the theroetical result is a draw.

For example stalemate is draw in 0 plies
The following position is draw in 2 plies (if white want to force a draw and black does not want to lose) in 2 ply(for example Kxh8 Kg6).
[d]6kB/4N3/7K/8/8/8/8/N7 b - - 0 40

It is a draw in 101 plies if black want to force a draw.

The game may start Kxh8 Nc6 Kg8 when black may claim a draw only after Kxh8 and 100 plies with no capture or alternatively if the last move of black is a capture by insufficient material because white does not allow stalemate or repetition or capturing the knight earlier).

The question is how many plies you need to force a draw with white and how many plies you need to force a draw with black from the initial position(assuming both sides never play a losing move).

[jefk]

with 'best' (*) play from both sides, you'll get something as
1.Nf3! d5 2.d4 Nf6 3.c4 e6 4.Nc3! (not g3 because of ... Bb4+ = 1/2-1/2)
Bb4! (or c6 maybe) 5.Bg5 and so on (...dxc4 6.e4 c5 7.Bc4) and it still will take
a lot of moves before you see the actual draw because of the basic
endgame rules (would an endgame tb lookup qualify? i think it would).

(very) rough estimate: 60 moves (that's 120 ply) and you may
hit some 6 or 7 men egtb. So let's try this out on the chinese database and
at move 29 i'm out of (this chinese) 'book';
rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1 moves g1f3 d7d5 d2d4 g8f6 c2c4 e7e6 b1c3 f8b4 c1g5 d5c4 e2e4 c7c5 f1c4 c5d4 f3d4 d8a5 g5d2 a5c5 c4b5 c8d7 d4b3 c5c7 a1c1 b8c6 b5d3 e8g8 a2a3 b4e7 e1g1 f8d8 g1h1 d7e8 d1e2 c7d7 d3c2 c6a5 d2e3 a5b3 c2b3 d7d3 e2d3 d8d3 f1d1 d3d1 b3d1 e7d6 f2f3 h7h6 d1e2 a7a6 h1g1 a8c8 c3a2 c8c1 a2c1 e8a4 g2g3

Still long way to go. Is it a drawn position? yep it is.

In correspondence chess they usually agree to make a draw on
average a *lot* earlier eg before move 40 (unless you happened to have plugged
in the wrong move, which is maybe still why some 2100+ people
play on ICCF

(*) with other moves the game would probably be shorter

[jefk]

more realistic 70 (instead of 60) moves, and then you're not yet in the egtb,
but the engine is certainly hitting the egtb at the endnodes in such a way you
can be certain of the result (guess what kind of result )

so that's 140 ply, which is a lot, which indeed means that a proper 'weak' solution
eg in checkers style is still far away (ten years maybe?). but chess is not checkers
(where you can have strange endgames with zugzwang); not going to dwell
onto that (what that means) further coz then i would be repeating myself.

[jefk]

here's another example
rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq - 0 1 moves g1f3 d7d5 d2d4 g8f6 c2c4 e7e6 b1c3 c7c6 e2e3 b8d7 d1c2 f8d6 f1d3 d5c4 d3c4 b7b5 c4d3 c8b7 e1g1 e8g8 f3g5 h7h6 g5e4 f6e4 c3e4 d6e7 b2b3 a7a6 c1b2 c6c5 d4c5 d7c5 e4c5 e7c5 d3h7 g8h8 h7e4 b7e4 c2e4 d8e7 a2a4 b5a4 a1a4 a6a5 b2c3 a8b8 a4a5 c5b4 c3b4 b8b4 e4c2 e7b7 a5a3 f8c8 c2d3 b4g4 g2g3 g4g5 f1a1 g5c5 d3d1 c5c7 d1e1 c8b8 a1b1 c7c6 b3b4 c6b6 a3b3 b6b5 h2h3 g7g6 g1h2

[Uri Blass]

more realistic 70 (instead of 60) moves, and then you're not yet in the egtb,
but the engine is certainly hitting the egtb at the endnodes in such a way you
can be certain of the result (guess what kind of result )

so that's 140 ply, which is a lot, which indeed means that a proper 'weak' solution
eg in checkers style is still far away (ten years maybe?). but chess is not checkers
(where you can have strange endgames with zugzwang); not going to dwell
onto that (what that means) further coz then i would be repeating myself.

Tablebases is a drawn position but not a draw by the chess rules.

Draw by the chess rule is or by the 50 move rule or by repetition or by insufficient material.

I believe that it is easy to get more than 200 plies at least with white because black has no way to force a draw by the chess rules in 200 plies.

If the rules are that white wins if it is a draw with more than 200 plies then I expect white to win every correspondence game at the high level.

[jefk]

ok when the egtb moves have to be played out, your estimate(s)
imo looks reasonable.

Such thoughts make sense when we would look at a rule modification proposal
Larry Kaufman made some time ago, to solve the draw problem (eg in
correspondence chess); a proposal which didn't receive much attention.

[supersharp77]

I think that we can define it by the following game:

Have a rule to decide about drawn games.

Draws with more than X plies are considered to be a win for white
Draws with less than X plies are considered to be a win for black
Draws with exactly X plies are considered to be a draw.

Assuming the theoretical result of chess is a draw It is obvious that there is exactly one X when the theretical result of the new game is also a draw and the Question is what is X

It is possible to ask the same question about what is Y that gives theretical result of a draw when
Draws with more than Y plies are considered to be a win for black
Draws with less than Y plies are considered to be a win for white
Draws with exactly Y plies are consdered to be a draw.

I guess that Y<X because white can force a draw faster than black.

Well Based On what I've been seeing with The Strongest Chess Engines (NN) with best play its a draw in 40-70 moves or so......."More Often Than Not".......100%.....