This is separate from the “What the? It’s groundhog day again” thread. This is hard data. One local reporter collated the results of all the past predictions he could find in the local paper archives. Mr Hog was correct only 39% of the time. Terrible. He can’t even get close to a 50/50 guess.
So what I’m thinking is that way back when the originator of this prediction method worked out the details he somehow got it backwards. I mean, a 61% correct prediction would be decent.
You mean the people that came up with this only got 39% correct.
NPR reported a 30% success rate.
I’m going to go with mixdenny. If he goes back into his tunnel after seeing his shadow, it means there will be good weather and he nap before doing his groundhog stuff, whatever that is.
Hogging ground, duh.
Why, of course!
If he sees his shadow, there will be six more weeks of Winter.
If he doesn’t see his shadow, Winter will only last another month and a half.
I assume that this is not adjusted for the baseline frequency. Because being consistently incorrect significantly more than half the time would be just as remarkable as the opposite. It would just mean that people were not reading the output correctly, and the groundhog is outperforming all supercomputer weather forecasting algorithms.
Yeah, I’ve always heard it more or less like this, too. The groundhog’s “prediction” is set up in such a way that it can’t be wrong.
But even if you do assign meaningful outcomes to it, how does one define what constitutes “the end of winter”? First thaw? Last freeze? Whether the temperature on March 16th (15th leap) is above or below the local median for that date?
My point exactly.
I thought I said that. At any rate, I agree.
Phil needs a NOT gate.
How much ground would a groundhog hog if a groundhog would hog ground?