So, today's the day that Americans prove once again that when given the choice between a scientific model, reached by the consensus of hundreds of climatologists and amply supported by evidence (e.g. climate change/global warming) and the prognostications of a rodent, they'll go with the rodent every time.
Today is Groundhog Day, which is the day that winter-weary northerners wait eagerly for Punxsutawney Phil, a groundhog who lives in Punxsutawney, Pennsylvania, to emerge from his burrow. The idea is that if Phil sees his shadow, it scares him back into the burrow and we'll have six more weeks of winter.
Living in upstate New York, I've always found this grimly amusing, because up here, six more weeks of winter would be good news. This would mean that spring would arrive in the third week of March, right around the equinox. (Up here we don't call the equinox "the First Day of Spring" because all that does is call attention to how miserably cold it actually is.) In upstate New York, we still have hard freezes at night at the end of April, and I remember twice having snow on Mother's Day.
In any case, let's assume that we give up on the "six more weeks of winter" thing, and just call it "sees the shadow, long winter; doesn't, short winter." How well does it work?
Tim Roche, a meteorologist with Weather Underground, has analyzed the 99 years' worth of records of Phil's predictions, and compared them to the actual weather that occurred that year. He found that when he predicted a short winter, he was right 47% of the time; when he predicted a long winter, he was right 36% of the time.
Me, I find this significant, especially in his predictions of a long winter. Note that his predictions of a short winter are right around where you'd be if it were a completely random flip of the coin -- just what I'd expect. (Yes, I know that "long winter" and "short winter" are relative terms, and there are "medium-length winters" and so on, but just play along, okay?) But look at his predictions of long winters -- he does considerably worse than you'd do if you just flipped a coin. I haven't done the statistical analysis, and honestly probably won't bother to, but I'm guessing that that deviation from a random 50/50 split is actually statistically significant. Does it count as a paranormal phenomenon if a psychic predicts an outcome wrong far more often than you'd expect?
Be that as it may, the whole Phil phenomenon is fantastically popular, and in fact has spawned a number of spinoffs. I know of two in my own home state of Louisiana. There's Pierre C. Shadeaux of New Iberia, who is a nutria, not a groundhog (if you don't know what a nutria is, picture a huge brown rat with orange teeth, and you've got the idea; they're sort of like the Rodents of Unusual Size from The Princess Bride, only less cute). Another nutria, T-Boy, is in the Audubon Zoo in New Orleans, presumably in the "Dear God What The Hell Is That Thing?" exhibit, and he is also coaxed out of his home early on February 2 to see if he sees his enormous, hairy, fanged shadow, which in this case will give us six more weeks of nightmares.
Of course, I know the Phil foolishness is all in fun, and I'm perfectly willing to take it in that spirit. And if you're curious, this year Phil came out in the middle of one of the worst storms to hit the East Coast in the last five years, and because of the cloud cover, he didn't see his shadow. Thus we have the results: Huge Snowstorm = Short Winter.
Makes perfect sense to me.