A positional game is a kind of a combinatorial game described by:
- $X$ a finite set of elements. (Often $X$ is called the board and its elements are called positions.)
- $F$ a family of subsets of $X$. (These subsets are usually called the winning-sets.)
- A criterion for winning the game.
For $n$-pile Nim, wouldn't $X$ be the union of disjoint posets, e.g. $\{(1,2,...,k_1), (1,2,...,k_2),..., (1,2,...,k_n)\}$? Then a move consists of choosing a point in one of the posets and removing that point and everything above it in said poset.
Then $F$ would be $\{\{{(1,2,...,j),(),()}\}, \{{(),(1,2,...,j),()}\}, \{{(),(),(1,2,...,j)}\}\}$
And the winning criterion is being the first one to "receive" a position in $F$?
This seems right to me, but it appears that player two can have a winning strategy, which contradicts the notion of Nim being strong positional.