So the fossil record, the molecular clock, Potassium-Argon dating and Mitochondrial DNA don't count.
The fossil record disproves evolution. The evidence is the Cambrian explosion and the lack of both intermediate fossils and intermediate species.
The molecular clock does not exist, looking at different genes one comes up with a different 'tree of life'.
Potassium-Argon dating is not exact enough to tell us what came first because of the large +- variable. In addition, because the fossil record is spotty, one can never be sure which species arose first or later.
The problem with mtDNA is quite simple - there is no standard (or for any other genetic dating technique). We do not know what the mtDNA was of any creature a billion years ago for the simple reason that DNA decays quite fast (hardly any usable samples over 100,000 years of age). So how can you measure a distance when you only know the end point but not the starting point? The answer is that you cannot.
They count as observations that support the theory, but they are not proof. According to the scientific method I learned, proof follows controlled, reproducable experiments that demonstrate the predictions implied by the theory.