If this is true, then why insist that the definition of science entails predicting results that we can fit into a theory? How are you going to separate "chance" from "predicatability" and still be a scientist?
What I said was that chance was not a principle of science. Taken to mean "statistics", chance is a tool of science, not unlike linear algebra or microscopes.
What you appear to be encompassing in the word "chance" is the principle of verification by successful prediction.
Perhaps you are intent on pointing out that some issues in evolutionary theory are verified by detecting what statisticians call central tendencies in data. As in, most small dino bones seem to be buried in Jurassic dirt, but not all. Creationists make much of not understanding this, because an understanding of central tendancies in statistical evaluation would terminate a massive clot of their arguments.
Simple example: Let's say there's a new Ice Age. Biologists predict, using the theory of evolution, that animals will either become extinct or adapt to the cold. Of those that survive, we can predict which that they will have some means of insulation, some means of food/energy strorage, etc.
However, since the mutations that produce these are random, we have no way to predict which species will have thicker fur, more blubber, hibernation, etc. Nor can we predict which species will not receive any useful mutation, and go extinct.