What goes up must come down, right?

In the past, I have just guessed at the severity of a crash and then determined whether or not to replace the servos. In helis, where one servo failure will result a really bad day, I usually try to error on the side of caution.

Other than the usual checking for obvious signs of failure (cracked case, freespinning gears, etc.) has anyone come up with a test or tests to determine if a servo will fail under load? Does anyone know if there is a way to test them, other than just going for it and putting them in the air?

thanks