In October, I pointed out that Ann Selzer, Iowa’s Queen of Polling, was so unnaturally good at polling elections that she actually had a better track record of predicting elections than Nate Silver. I therefore introduced the Selzer+Silver model, which aims to improve the FiveThirtyEight model by (basically) treating Selzer’s polls as the absolute truth and using them to anchor the FiveThirtyEight model results. At the time, my analysis was entirely retrospective and therefore could have been caused by cherry-picking results....

What if I told you that there’s a way to improve the election forecasts from FiveThirtyEight? That it only uses publicly available information and doesn’t require any advanced math? The catch is that it only works in Iowa (but there’s a workaround). Here’s the trick: instead of using FiveThirtyEight’s weighted average of all the polls and demographic-based predictions, use the results from Ann Selzer’s final Iowa poll....

Finding scientific papers with a plausible premise but terrible statistics is, if not an everyday occurrence, unfortunately common. This was the basically the entire shtick of psychology papers for a couple decades - find some plausible sounding premise to write a hypothesis about, run an experiment on the first 16 undergrads you can find, and keep running regressions until you find one where p < 0.05. And then you end up with experiments whose results can’t be replicated, because their conclusions were always based on random chance....

Note: This was written in Fall 2021, when Delta was predominant strain of COVID. It is not medical advice, and I have not revised or updated it in any way to account for new papers, new COVID variants, or new boosters While the initial clinical trials on the effectiveness of the COVID vaccines were astoundingly impressive, more recent studies have suggested that the vaccines may be less effective against Delta and that their effectiveness may decline over time....