When you define natural and supernatural that way you are missing the point. A lot of supernatural phenomenon are usually easily explained, although the explanations invoke a power that isn't easily observed. "How did the oracle know that?" - She was given revelation from a spirit with more knowledge than her. "How was the Red Sea split?" - YHWH did it. Here the supernatural distinction is important - the Red Sea doesn't spontaneously split, but a supernatural power did it. Claiming that that would be natural is missing the point of the distinction. What's natural is what's regular, what's possible to recreate and predict by simple observation. The methodology for intervention by supernatural beings is different. "Okay, the Red Sea split. How do we explain this?" Regular empiric science fails here, because you can't recreate it, but when you take into account the religiohistoric context, "YHWH did it" makes the best contender for an explanation. Why? The being that appeared to Mose in the burning bush used that name, claiming to be the God of Abraham, Isaac and Jacob, the forefathers of the Israelites, to whom he had revealed himself 400 years earlier. Then he said that he would deliver the Israelites from the slavery of Egypt, then the ten plagues came over Egypt in the way Moses had described by the command of God, and then the Israelites were released. When something happens that is impossible to explain using the naturalistic sciences, using the religious methodology to give an explanation is the most reasonable way, and in that case, calling the explanation "natural" doesn't make sense.
The superficial counterargument to this would be that I don't believe most of that stuff ever happened. But let's set that aside because that's not the point.
If YHWH is an agent who is capable of interfering with the natural universe, then there isn't really a categorical difference between the things he does and the things e.g. I do. You can't recreate the things I do by waiting for the thing to happen by itself, someone has to interfere. You can try to predict the things I am going to do with some greater than chance and lower than perfect success, and if we understood anything at all about YHWH (e.g. if everything the Bible said about him was true and we used that info to predict his future behavior), we could say the same about him. He might be able to do things that I don't understand, but I am also able to do things that e.g. people in medieval Europe wouldn't have understood, so that's at best a difference in degree, not in kind. I could go say hi to some people, and 40 years later, go see them again and explain that I'm the same guy I was 40 years ago and that I had a plan for smuggling them out of jail or something and then I could do the plan and have it work exactly as I described. Would that be supernatural, and would that make "Awaclus did it" a likely and sufficient explanation for something like a sea getting split later?
Back to the topic of religious vs. non-religious explanations: I could (if we are not restricted to what seems realistic) also invent a method for splitting seas, invent a time machine and go back in time to split the Red Sea for Moses. That's not a bigger contradiction with what we currently think are the laws of physics than some stuff YHWH allegedly did according to the Bible. The idea of a future me time traveling around and doing all the things attributed to YHWH in the Bible but with the power of future science instead of god magic is obviously ridiculous and super random, but it's an equally good explanation as the Bible and really the only reason why it sticks out as ridiculous and random is that people hold non-religious explanations to a much, much higher standard than religious ones.
Another application of the religious methodology: I could give you a lot of evidence that parts of the Bible are of supernatural origin, that they are historically reliable, that it is psychologically helpful and so on, but the only reason I can give for why I believe it is true in its entirity is this: God has gained my trust.
Sure, and the followers of other religions would say similar things about their holy texts that obviously can't be true if the Bible is true and vice versa. So, necessarily, the religious methodology leads to a huge number of people having strong convictions in explanations that are not true, which makes it a terrible methodology.
Although ordinarily I'd say the Shelling point is the first item of the list.
