I wish it was that clear cut. This is a tricky business in the sciences. Often, the quantity you are trying to validate is unobservable. In this case, I suspect (I'm not a squishy scientist) it is because evolution has such long time scales.

In other fields, it can be simply because it is prohibitively difficult or expensive to instrument the quantity of interest. This is extremely common, even in the 'hard' scientific fields and engineering. You very, very seldom have all the data you want.

This is all a long winded way of saying that confirmation of a hypothesis by mathematical models certainly is a mechanism by which science creates a stronger degree of belief in said hypothesis. Would it be a full fledged validation of the model? Probably not. But it is absolutely useful evidence.

Full disclosure, I am a computation scientist.