How about the only reliable weapon in your arsenal? Einstein used mainly intuition for the Special Theory of Relativity, but when it came to the (much more difficult) General Theory, he turned to math.
The biological (soft) sciences often stumble when they try to extrapolate solely based on empirical evidence, but lack an underlying (mathematical) model.
More to your point: Agreed, science doesn't really explain anything about nature---it just observes it and records connections. However, whenever you want a predictive tool...nothing beats a good model. The language that expresses those models is mathematics. Without models, science becomes a discombobulated series of statements. I feel that's what's so confusing about a lot of the health-related science that get reported in the news: "I did a study...", "Here's what I saw...(blah blah blah/statistical correlation)", "I have no model...", "Draw your own conclusions..."
Also my beef with AI/Neural-Networks---no model (or more precisely, a really shoddy one).
EDIT: I'll add the Zen Koan: "The map is not the territory.", i.e. all models are an imperfect simplification of the truth....but they can be darn useful.
OK. Now I've gone too far afield...even for my high tolerance for non-sequitur.
Back to D&D.
