Given the following definition:

Let $M$ be a move set and $A,B \subset M^\omega$ are disjoint, we define the winning conditions for the game $G(A,B)$ as follows: if the play of the game is $x \in M^\omega$, then player I wins if $x \in A$, player II wins if $x \in B$, and otherwise the game is a draw. A strategy $\sigma$ is winning or drawing in $G(A,B)$ for player I if for all strategies $\tau$ , we have that $\sigma * \tau \in A$ or $\sigma * \tau \notin B$, respectively. A strategy $\tau$ is winning or drawing in $G(A,B)$ for player II if for all strategies $\sigma$, we have that $\sigma * \tau \in B$ or $\sigma * \tau \notin A$, respectively.

what would a "drawing strategy in $G(A, B)$" mean? A strategy such that a player can at least play a draw (so either draw or win), or a strategy such that a player can force a draw (so the game always ends in a draw). I find the formulation of the definition a bit unclear in this regard (but maybe that's just me?), and I think these two options are not equivalent.

  • $\begingroup$ Note that it may be possible that a player has no winning strategy but has a win-or-draw strategy and yet has no strategy that guarantees exactly a draw! There is almost no reason one would be interested in an exactly-drawing strategy. $\endgroup$
    – user21820
    Sep 6, 2020 at 17:03

1 Answer 1


A drawing strategy is one which either draws or wins against every strategy - or perhaps more snappily, one which never loses. The point (looking at drawing strategies for player $I$ for simplicity) is that $\sigma*\tau\not\in B$ means that player $II$ does not win the play of $\sigma$ against $\tau$ - that is, that $\sigma$ either draws against $\tau$ or wins against $\tau$.

(In particular, $\sigma*\tau\in A$ does imply, but is not implied by, $\sigma*\tau\not\in B$.)

  • $\begingroup$ OK, this is what I guessed. But a strategy that forces draws exclusively is strictly stronger than a drawing strategy? Is there any game theoretic interest in this kind of strategy? $\endgroup$
    – Jori
    Sep 5, 2020 at 22:48
  • $\begingroup$ @Jori I wouldn't say that. The condition of forcing a draw is strictly stronger than forcing a non-loss, but a strategy which forces a draw is less strong, in the sense of quality of strategy, than a strategy which forces a draw-or-win but sometimes allows a win. If we have the option we want to win more. So I don't see immediately what the interest in that stronger condition would be, but of course I could be missing something. $\endgroup$ Sep 5, 2020 at 23:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.