Were any of Silver’s previous predictions generated by making a list of possibilities, assuming each was a coin flip, multiplying 2^N, and rounding? I get the impression that he’s not exactly employing his full statistical toolkit here.
What Adams does is that he looks at Silver’s estimate, says that it is way too low and then takes 1 minus Silver’s estimate as his own estimate just to make a point. He does not attempt any statistical analysis and the 98% figure should not be taken seriously.
What Adams has said he’s doing is simulating the future along the mainline prediction—i.e. nothing too weird happens—and under his model, Trump is guaranteed to win. Then he says “well, maybe something weird will happen” and drops that confidence by 2%, instead of a more reasonable 30% (or 50%).
Were any of Silver’s previous predictions generated by making a list of possibilities, assuming each was a coin flip, multiplying 2^N, and rounding? I get the impression that he’s not exactly employing his full statistical toolkit here.
Isolated demands for rigor—what do you think Adams is doing? (I think he’s generating traffic.)
But sure, I agree, that’s more of a reasonable prior than an argument. There’s more info on the table now.
What Adams does is that he looks at Silver’s estimate, says that it is way too low and then takes 1 minus Silver’s estimate as his own estimate just to make a point. He does not attempt any statistical analysis and the 98% figure should not be taken seriously.
What Adams has said he’s doing is simulating the future along the mainline prediction—i.e. nothing too weird happens—and under his model, Trump is guaranteed to win. Then he says “well, maybe something weird will happen” and drops that confidence by 2%, instead of a more reasonable 30% (or 50%).