I asked someone with a PhD in statistics to, "Explain regression to me like I know some math but have never used statistics." For those of you who don't know, regression can be taught to people either with a single course in calculus or people with a single course in linear algebra. It is also a, maybe the, fundamental statistical tool. I'm generally satisfied with people drawing some points on a 2-variable graph, explaining we're trying to calculate the least error fit, then telling me what that the error specifically is squared error. If the candidate can discuss penalties and outliers and glms and exponential family distributions and IRLS and sampling distributions and gauss-markov then great, but I'm satisfied with a really basic understanding. This guy couldn't get any of it. I simply don't understand how you can have a phd in statistics and not be able to explain in 60 seconds roughly what regression is, how one interprets the betas, and at least one basic inference method.