It sounds like your thinking is getting more confused as we go on with this, so I’m not sure if this response will help or not. But this whole answer is a mess. Most of what you have me saying here is a complete mischaracterization. I don’t know if you’re doing it on purpose or if you haven’t comprehended the words I wrote. And you still haven’t given any evidence.

It sounds like you don’t even understand who has the burden of proof for which claims. You made a factual claim about the effects of oversampling, and you have the burden of proof to support it. In response, I summarized the common knowledge about how polls are analyzed statistically. That shouldn’t require any specific support, but I’m sure you can start at the beginning and read up on it if you don’t believe me.

Even if you don’t take away anything else, remember this: There’s a huge difference between the polls of voter preferences and the predictions of victory based on those polls. First you get the numbers for voter preferences, numbers like 48% for Hillary and 46% for Trump, and then you run further statistical tests to arrive at a probabilistic prediction for victory one way or the other. Virtually everyone did say that Hillary was overwhelmingly likely to win. Some people said her chances were better than 90%. But the fact that Trump won doesn’t make that prediction wrong. It just means he hit on an unlikely outcome. You definitely can’t point to the predictions of victory and say that they prove the polls of voter preferences are unreliable. One has nothing to do with the other.

Going back to the burden of proof, I originally made factual claims and cited long-term polling trends to back them up. The unstated (but safe) assumption is that polls themselves actually do give us knowledge about what real people are thinking about elections, especially when you take the averages of lots of different polls, and you look at the trends over time. That assumption is actually borne out empirically by comparing decades of opinion polling with the results of actual elections. Frankly, if opinion polling didn’t work, we would have long since stopped using it. People dedicate their entire working lives to this stuff. They’re not trying to waste their time on illusions.

For you to respond to my factual claims by saying that polls are by nature unreliable and useless is a really extreme position, that requires serious long-term analysis. It would take a book’s worth of facts and original research just to make people take a second look. You haven’t done that work, and you haven’t pointed us to someone else who has, so you haven’t supported your claim.

May the best argument win. And let us shake hands when it’s done.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store