For a different view about numbers and their uses see P.Tillers & J. Gottfried, "United States v. Copeland: A Collateral Attack on the Legal Maxim that Proof Beyond a Reasonable Doubt Is Unquantifiable?," 5 Law, Probability and Risk 135 (Oxford University Press, 2006) and P. Tillers, Trial by Mathematics - Reconsidered, -- Law, Probability and Risk -- (forthcoming 2011).
But the 10th Circuit, having disavowed the appropriateness using numbers to discuss probable cause, proceeds to inform Mr. Ludwig why a 58% "accuracy rate" by a sniffer dog is more than enough to satisfy the probable cause requirement -- this is so, the court says, because "probable cause doesn't require an officer's suspicion about the presence of contraband to be 'more likely true than false.'" But isn't this statement by the court just a number by another name? (Yes, it is.)
The dynamic evidence page
It's here: the law of evidence on Spindle Law. See also this post and this post.