Your argument here raises two questions:
- Is paleontology a "hard science"?
- Were dinosaurs reptiles, and if not, why not?
Is paleontology "hard science"?
Definition of "soft science":
- "any of the specialized fields or disciplines, as psychology, sociology, anthropology, or political science, that interpret human behavior, institutions, society, etc., on the basis of scientific investigations for which it may be difficult to establish strictly measurable criteria."
Definition of "hard science":
- "any of the natural or physical sciences, as chemistry, biology, physics, or astronomy, in which aspects of the universe are investigated by means of hypotheses and experiments."
Chemistry, biology, physics and astronomy are here listed as "hard science" and all provide evidence confirming evolution theory.
Evolutionary biology is sometimes called "soft", but that was in the days before detailed DNA analysis began upsetting the old order of biological classifications & relationships.
Paleontology itself is not found listed in either "hard" or "soft" category.
Bottom line: it's not clear if the pejorative term "soft science" is anything more than a polite way of saying: I disagree with your conclusions, so I'll call you "soft", pal.
Were dinosaurs reptiles?
Until Bob Bakker in the late 60s, dinosaurs were thought to be cold-blooded like reptiles.
Today dinosaurs are classified along with birds and reptiles among the archeosaurs a clade-order in the class Reptilia.
But where, exactly, they belong is still unclear.
When & how, did ancestors of dinosaurs split from those of birds or mammals, is not fully known.
But the real question here is whether that term "soft science" is simply euphemism for "inaccurate guess-work", and if so, then the biological classification system has gone from "soft" to harder & harder over many years, as more and more hard data (i.e., DNA comparisons) came into the picture.
Archeosaurs include two main sub-clades, "Pseudosuchia, which includes crocodilians and their extinct relatives, and Ornithosuchia, which includes birds and their extinct relatives (such as non-avian dinosaurs and pterosaurs)":
Examples of Archeosaur Ornithosuchians (Avemetatarsalia) and Pseudosuchians:
