Well, that's one way you can approach it: look at the graph of potential facts instead—but what I was considering was an interpretation of the Principle of Suff

Well, that's one way you can approach it: look at the graph of potential facts instead—but what I was considering was an interpretation of the Principle of Sufficient Reason where dead ends are fine at a single point in time, which adds another dimension to the graph and leaves freedom for not having dead ends, cycles, or infinite chains (as long as 'dead end' is interpreted as a node not only without outgoing edges, but which cannot develop any at any future time).