Agree with the post proper. I think the headline is technically accurate but potentially misleading, because poll-dominated models aren’t the only kind of election models. Political scientists build models that rely more on fundamentals like economic statistics and military activity, and when Vox averaged 6 of those models together, they predicted that Trump would win the popular vote. The headline remains technically correct because predicting that Trump would win the popular vote isn’t the same as predicting Trump would win the election, but it’d be a shame if people walked away with the idea that election models in toto said Clinton would win.
I think models that rely on fundamentals are worthless. I don’t have time to explain why in details, though perhaps I will post something on that at some point, but if you want to know the gist of my argument, it’s that models of that kind are massively underdetermined by the evidence.
OK. That’s interesting. I disagree but I can see why you’d think that, and in a way I’m kind of sympathetic: I think overfitting definitely happens with some of the poli. sci. models. My go-to model is my go-to exactly because its author really seems to appreciate the overfitting issue, and is very insistent on aiming for proper explanation, not just prediction.
Agree with the post proper. I think the headline is technically accurate but potentially misleading, because poll-dominated models aren’t the only kind of election models. Political scientists build models that rely more on fundamentals like economic statistics and military activity, and when Vox averaged 6 of those models together, they predicted that Trump would win the popular vote. The headline remains technically correct because predicting that Trump would win the popular vote isn’t the same as predicting Trump would win the election, but it’d be a shame if people walked away with the idea that election models in toto said Clinton would win.
I think models that rely on fundamentals are worthless. I don’t have time to explain why in details, though perhaps I will post something on that at some point, but if you want to know the gist of my argument, it’s that models of that kind are massively underdetermined by the evidence.
OK. That’s interesting. I disagree but I can see why you’d think that, and in a way I’m kind of sympathetic: I think overfitting definitely happens with some of the poli. sci. models. My go-to model is my go-to exactly because its author really seems to appreciate the overfitting issue, and is very insistent on aiming for proper explanation, not just prediction.