Name anything that is consistent with your same state past that does not FIRST assume there was one!??
A 'same past state' or a 'past state'?
Yes, we assume that there was a 'past state'. To do anything else would be Last Thursdayism.
But, with that, we *can* test to see if that past state was or was not the same as the current state.
How? Quite easily. If the past state was significantly different, then the results *now* would be significantly different. That they are not (i.e, consistency) is evidence that the state in the past was *not* different.
As a pair of examples, consider radioactive dating and lake lamellae. We do both types of tests and do them on several different lakes and with several different radioactive substances.
What we find is that the different methods (different radioactive substances, lamellae) give consistent results in all cases where they are compared.
Now, for this consistency to happen, assuming a different state of things in the past, requires several very different physical processes to respond to changing conditions (past state to current state) in a way that gives results that are consistent with each other *now*.
So, in lake A, layers are being deposited at a certain rate. We agree that this rate of deposition could have been quite different in the past. Also, say, C14 dates for the different layers give results consistent with the current rate of deposition and the current rate of C14 decay.
This alone is an incredible coincidence if true. Why should the rate of deposition of the layers be in any way related to the rate of C14 decay? They are entirely different processes. To stay consistent if the conditions *were* different in the past would require either that the decay rates in the past were correlated in a way that there is no physical reason OR that in the transition, things changed for both the depositional rates and the decay rates, but again in such a way that all the results stay consistent.
But this is only for one lake. If we instead look at a different lake, it will have a different rate of deposition. But the C14 decay rate (currently) is the same. Once again, for the layers we count to be consistent with a constant decay rate of C14 would require an impressive coincidence in how things changed in the past in order to give the results we see today.
BUT, the fact that it is the same rate for the two different lakes puts even more constraints on the coincidence. Not only do the changes (whatever they were) to the decay rate of C14 have to match the deposition rates (whatever they were) for the first lake, they *also* have to match the *different* deposition rates for the second lake.
And now, we can add in several different lakes, which makes the coincidence even more unlikely.
But, we can *also* go to a different radioactive nucleus, one that decays by a different physical process than C14 (say, U238). An this method *also* has to have its rates change in precisely the same ways as those for C14 (which would be unlikely given that they are different physical processes) and also in such a way that they are consistent with all the different deposition layers in all of the lakes.
The fact that different methods, based on physically different systems with different rates (and presumably different rates in the past) would change in *precisely* the way to stay consistent with each other and with all *other* processes and *still* have them all be wrong in the dates that they give would be an incredible coincidence.
And, at some point, you get to a different version of last Thursdayism. Instead of saying everything started Last Thursday with all records set up to give false information, we have a different version where prior to Last Thursday time was going 100 times as fast, but all physical processes were speeded up by the same amount, so there is no difference in what we would see.
And *that* gets to the very definition of what it means to measure time. A second is *defined* by a certain number of oscillations of a particular type of light. If that oscillation *speeded up* by a factor of 100, the very definition of a second would change along with it and instead of 1 second, the definition would give 100 seconds.
So, that is the evidence: the consistency across a large number of independent methods of dating, all based on different physical processes and, ultimately, the definition of time itself.