A couple weeks ago I started writing about an analysis I put together of the 2020 Senate races, based upon an examination of PVI and generic congressional ballot polling. I got a thousand or so words in before I realized that what I really had was a standalone article explaining the consequences of over-investing in sexy but unwinnable races, while ignoring genuinely winnable races. That article also explains what the hell PVI and generic ballot polling are, so if you haven’t already, please read that first.