I.D can explain more of the observed facts evolution can: how evidence works.

[struggle4progress]

39. A key requirement for a good logical theory is that it not prove too much.

Many useful logical theories not only begin by accepting certain claims but also reject certain claims, and among the rejected claims are always claims that (in some sense) have too many consequences

If we suspect Grumpy as the person who gunned Dopey down sometime between noon and 1:00PM on Saturday at The Palace Nightclub in Queens, but Grumpy has alibis for the whole time, we might scratch him off the suspect list. But should Grumpy's alibis prove too much, we will still doubt them: if (for example) Bashful testifies he was with Grumpy in Manhattan from 10:00AM until 12:30PM, and then Doc testifies he was with Grumpy in Las Vegas from noon until 2:00PM, we conclude Grumpy's account of his location midday Saturday has too many consequences to merit serious consideration

Different people in different contexts might disagree on when exactly a theory has too many consequences. Certainly a theory from which one can deduce anything whatsoever is a theory with too many consequences

To posit beings, with powers vastly beyond our own, does not lead to useful logical theories: the assumption permits arguments from which one can deduce anything whatsoever. This, of course, does not necessarily constitute a disproof of any and all religious ideas: it is merely the observation that one cannot reason reliably about beings with powers vastly beyond our own

Intelligent Design cannot be considered a scientific theory, in part because the underlying assumptions appear to prove too much to allow us to construct a useful logical theory there