Evidence for Believing

[Inqwizitor]

If it was a testable, verifiable phenomenon that spontaneous remission occurred significantly more frequently during or as a result of prayer, we should look at some powerful psychosomatic (natural) cause, like perhaps another order of magnitude in the placebo effect. If something is observably repeatable within the order of empirical phenomena such that you can form an inductive conclusion from predictable data, that makes it less likely to be a miracle, not more likely.

Yes, it's not an "end-all, be-all" procedure but such observations could still put us on a good starting point for seriously considering the plausibility of miracles. We would have to go beyond just this basic experiment to establish the reality of miracles of course (say different controls, e.g. "praying to rocks").

But more importantly, this seems a concession on your part of the difficulty of attaining conclusive evidence for miracles.

ETA: Nevertheless, I disagree that repeatability of observations means they can't [likely] be supernatural/divine.

We seem to be using different semantics again. Miracles are extraordinary events; if it is an event that you can set controls for in an experiment, then you are establishing an order to it. If it's an observable, predictable order of things, that fits neatly within naturalism — unless you're locking down naturalism to (reductive) physical materialism.