Full Idea
Statistical relevance, not high probability, is the key desideratum in statistical explanation.
Gist of Idea
Statistical explanation needs relevance, not high probability
Source
Wesley Salmon (Four Decades of Scientific Explanation [1989], 2.5)
Book Reference
Salmon,Wesley C.: 'Four Decades of Scientific Explanation', ed/tr. Humphreys,Paul [Pittsburgh 2006], p.59
A Reaction
I suspect that this is because the explanation will not ultimately be probabilistic at all, but mechanical and causal. Hence the link is what counts, which is the relevance. He notes that relevance needs two values instead of one high value.