It turns out that the coincidence involved in the null hypothesis is somewhat less improbable than I thought.

The situation: we were waiting for "the closest-ever predicted approach to Earth for an object this large" (asteroid 2012 DA 14), and with just hours remaining, "the largest recorded object encountered by Earth since 1908" suddenly showed up, above a populated area, and blew up spectacularly.

The consideration which reduces the improbability slightly, is that new records and notable events in the former category, now occur almost every year. So rather than a coincidence between a once-in-a-century event and a once-in-a-decade event, it's more like a once-in-a-century event and a once-a-year event, occurring on the same day.

Assuming the null hypothesis for the moment (pure coincidence), we should still be aware that this is a remarkable coincidence. The Chelyabinsk fireball would have been a notable event of the 21st century anyway, but now it will go into history accompanied by the spooky fact that the world was already watching the skies that day; so it will have a place, not just in the annals of the space age, but in those chronicles of weird coincidences that titillate agnostics and agitate fringe believers.

I see that Phil Plait says he thought it was a hoax at first. That reminds me of my own idea that it was probably an audacious covert operation. We were each applying a familiar template to an unlikely-sounding event, something which at first sight was too unlikely to be regarded as a coincidence, so either it had to be denied or given a causal connection to the other part of the coincidence, after all.

And I'm still wary of letting go of that feeling that here is a glimpse of hidden connections. If Earth is being trolled by the long tail of the FSM, part of the divine entertainment might be to see the efficiency with which certain humans can dismiss even the highly improbable as just a coincidence, if they can't see a satisfactory explanation.

More mundanely, although the hypothesis that the fireball was something artificial is looking weak - I cannot think of any scenario which makes much physical and political sense - I wonder what the relative likelihood of "natural but related" and "natural but unrelated" really is. If the probability of "natural but related" is as high as 1 in 500, it may after all compare favorably with the alternative.