If ‘miracles’ occur (i.e., God affecting the universe, manifested through non-chance coincidings – etymologically, miracle just means an event that causes wonder), the pattern that is observed is that they (at least tend to) occur somehow and they occur through things.
(It is probably useful to note before proceeding that these non-chance coincidings attributable to God, according to Christianity, seem to happen quite often. For example, although a dramatic non-chance coinciding might not happen often in a given person’s life, if you multiply that by the number of people on the globe, the number of supposed dramatic miracles occurring is very high indeed.)
So, to explain a miracle by showing how it happened isn’t to really explain the miracle. What requires explanation is the probability – was it likely or not given some standard background repertoire of explanation?
For example, take a story from the Old Testament (because it can act as a common point of reference), of Moses striking a rock and water flowing, which supposedly replenished the people and livestock. Let us assume here that this is based on something that actually happened. How would this have happened? Well, presumably there was a water source somewhere in the desert, and Moses somehow found it.
This seems like it would be unlikely given conventional explanations. So, if you say “Aha! There was just this water behind or under a rock, and then Moses just happened to strike the rock, and so release the water! No miracle involved!” this is to misunderstand the situation, because miracles occur through things – there will probably be some back story like this. Rather, what has to be explained is the probability – is it reasonable to posit that this occurred through conventional mechanisms?
(The secular response to this is to explain away most seeming non-chance coincidings where the events are recent (unlike various events in the Gospels, say, where the evidential situation is different due to the intervening time). This can be done because the probability frameworks are difficult to establish, and so one can maintain that it was chance, or some more conventional causal explanation that doesn’t require great improbabilities, even though it might seem otherwise.)