I don’t know about this... Forecasts in general can definitely be wrong. Assigning a 99% chance to Clinton winning, for example, was a modeling error. If you don’t have the outcome in your prediction interval, from a modeling standpoint you very likely did something wrong. https://twitter.com/zeynep/status/1323649467015376896

There’s also another troubling (IMO) assertion here that forecasts are useless unless they rule out victory for either party. But that’s not our goal! The goal is to tell you how close the polls are, to answer the Q of how often candidates should win given the numbers.

So YMMW when it comes to forecasting. I have certainly been open about the weaknesses of this whole enterprise (possible behavioral consequences, poor calibration on small N, etc) but I’m not sure these objections really hold up. Maybe I’ll have time to say more tomorrow.