The definition of faith is, "Complete trust or confidence in someone or something". It has nothing to do with whether or not there is evidence supporting the thing you have faith in.
That's just one of numerous definitions of the word, a historically legitimate definition to be sure. Sadly any innocent use of the word for that purpose is -- to my mind at least -- tainted by the other, socially dominant meaning.
And anyway, even using your cherry-picked definition, I still say I don't have faith in science, because I can't afford it "complete trust or confidence". While I do think the core principles of science are the best tools we have to describe the natural world, even these core principles can not be immune to questioning.
I am sure he doesn't have complete and total trust or confidence in evolution; evolution is after all, falsifiable. A rabbit in the Pre-Cambrian would surely rock his boat. The boat has already been rocked many times before, and course adjustments were made based on the new information conveyed by that rocking.
It just so happens however that none of the things that would kill the theory have ever actually been shown.
