Arguments in favor of miracles are always more improbable than rational explanations.
Summary
Quotes
“David Hume's Pythian test for proving a miracle comes irresistibly to mind: "No testimony is sufficient to establish a miracle, unless the testimony is of such a kind that its falsity would be even more miraculous than the fact it attempts to establish." It may seem unlikely that seventy thousand people could have allowed themselves to be deceived at the same time, or agreed on a mass lie. Or that this historical data - that seventy thousand people say they saw the sun dance - is false. Or that they all saw a mirage at the same time (they'd been persuaded to look at the sun, which can't have been great for their eyesight). But the least of these apparent improbabilities is far more likely than the alternative, namely that the Earth was suddenly torn from its orbit and the solar system destroyed, without anyone outside Fátima noticing. I mean, Portugal isn't that isolated.”