Your knowledge is insufficient
When evolution is combined with game theory, it can make concrete predictions about what behaviours we expect to see in which populations, and why.
One of the first non-trivial predictions of this form was detailed by Edmond O. Wilson, and it involved the treatment of ant larvae depending on gender, and depending on whether the workers or queen were in control of the colony. (The workers are normally in control, except in slaver ants, and there only if there are few workers and many slaves.) Because of a quirk of ant genetics, 2 daughters share 3/4 of each other's genes, while a daughter only shares half with a parent, child or full brother. Hence one expects favourtism towards sisters, and IIRC game theory predicts that they will get fed more by a ratio of 3/1. This prediction is born out in observation. The queen, of course, has no reason to favour one over the other, and so when she is in control you expect an even ratio. And in slaver ants, the treatment of larvae varied as expected.
Incidentally this quirk is why ants and bees tend to form colonies. The daughters do a better job of passing on their own genes by helping mom have more daughters than by having their own daughters. The result is a colony with one queen, hordes of sterile daughters, and a smaller number of fertile sons and daughters.
If you talk to people in fields like ecology and animal behaviour, the ability to make predictions like these directs a lot of their research.
(There are other ways that evolution matters for prediction. For instance research on the role of genetic diversity in rapid evolutionary responses was utterly critical to figuring out how to breed better crops in the "Green Revolution" that started in the 60s. But those results are about theories of how evolution might progress, and are not consequences of its having progressed for a while.)
Cheers,
Ben
