explaining more does not necessarily mean better. I can explain all of the motions of all of the celestial bodies (better even than newton-einstein gravity can

explaining more does not necessarily mean better. I can explain all of the motions of all of the celestial bodies (better even than newton-einstein gravity can!) using epicycles, but that's trivial because really epicycles are just fourier transforms under another name and all continuous-valued functions have well-defined fourier transforms. 1 - give me a prediction, not just an explanation. Can you take epicycles and project the motions of any given body better than newton? No. 2 - Give me a multimodal observation. Imagine the world before we unified terrestrial gravity with heavenly motion - these were two separate phenomena. Could epicycles explain things outside of the category which it is designed to explain (heavenly motion)? No. Newton's theory also explains why if you throw a book across the room it will fall to earth, which epicycles do not.