Are miracles less likely than not?

I'm posting something I said in recent correspondence with some friends:

I just find the whole business of probabilifying miracles nonsensical. It's said that miracles are inherently or antecedently unlikely. 

Take the miracle at Cana. By that logic, it was less likely than not (indeed, far less likely) that God would perform the miracle at Cana. But how is anyone in a position to say in advance (and after the fact it's moot) whether or not God intended to perform the miracle at Cana? How do you lay odds for that hypothetical? 

If, moreover, God did in fact perform the miracle at Cana, how is it less likely than not (indeed, far less likely) that he wouldn't do what he was going to do? If he did it, then isn't it at least more likely than not that he was going to do what he did? 

Perhaps an atheist will say the evidence for atheism renders a miracle improbable. But in that event, it's not the probability of a miracle, but the probability of a miracle-working God, that's at issue.

Since, moreover, any evidence for miracles would subtract from any (alleged) evidence for atheism, is it not viciously circular to make atheism the gauge for assigning a probability value to miracles–even if you're an atheist? 

Not to mention that it would only take one bona fide miracle to falsify atheism. The threshold for falsifying atheism is exceedingly low. 

To take a comparison, what's the probability of a royal flush? Assuming the deck is randomly shuffled, that's a straightforward mathematical calculation.

But what's the probability of a royal flush if the deck is stacked? Well, assuming the card sharp is good at his job, it's inevitable.

So that becomes a question of how probable it is that the deck is stacked, which in turn, becomes a question of how probable it is that the dealer is a card sharp.

I don't see how treating probability statistically enables us to lay odds on whether or not the deck is stacked. That's a question of what would motivate a dealer to stack the deck.

In my illustration, the uniformity of nature is analogous to randomly shuffled decks, while a miracle is analogous to a stacked deck. 

I don't mind defining a miracle as an action that inhibits the world from continuing in the way it would if left to itself.

But since a miracle involves personal agency or personal intention, overriding how the world would continue if left to itself, the question is how to assign a probability value to God's will to perform (or not perform) a miracle. I don't see how statistics or background knowledge regarding the general uniformity of nature is germane to how we anticipate or estimate God's intention to perform a miracle.