Then, of course, thousands of convicted murderers in this country should go free. After all, no one saw them do it, but there was evidence (derived by science) that placed them at the scene of the crime. Hell, the distance scale of the universe, the great lynchpin of determining distance is based not upon physics, but simple geometry. It is geometry alone that makes it readily apparent that the universe is old. The next step in the distance scale is just a simple relation between the period and luminosity of certain variable stars. It too doesn't require physics, but a careful examination of the behavior of several different types of stars to determine distance. Just a simple relation is enough to begin to drive the stake in the heart of a young universe. It is the application of simple physical laws (already well determined here on Earth) that puts the nail in the coffin of a young universe (through redshift, and other well-studied astronomical relations).
I still don't think you realize how based in local demonstratable physics much of astrophysics is based upon. There is no ad-hoc assumption of millions of billions of years in astrophysics, a lot of what we say can only be explained by an old universe. I was talking to a colleague about clusters of galaxies a couple of days ago, and he's been working on research that is finding clusters of galaxies that haven't dynamically relaxed yet at redshifts of 1.4-1.6, which will be a completely new finding as almost every cluster we've seen is in some sort of orbital equilibrium. Even more exciting is that he thinks he's finally seeing giant elliptical galaxies, the large central cores of galaxy clusters still undergoing star formation and mergers at that redshift. It's a huge discovery, and it's one of the final holes in the expanding universe paradigm that needed to be filled.
Now you may be saying "whoopy-doo", but that data would make no sense in any other model. We see developing galaxies of redshifts of 2 and 3 (and 4 and 6), and by the time we look at redshifts of 1.5, we see clusters of galaxies just finishing their development, and at redshifts of 1.0, we see the finished product that we see all the way into the nearby universe. It's all pretty straightforward when you look at it with an open mind.
Forensic science deals with current evidence and events. Evolution theory makes assertions stretching back into the unrecorded and unobservable. It doesn't take a scientist to figure out that any assertions based on such things is merely positing reasonable conjecture. It should not be treated as a matter of fact.
As far as criminal justice is concerned, there is always an element of uncertainty, as the prospect of false and unreliable witness is all but certain.