All media outlets were predicting a landslide victory for Clinton. But then, as we all know I’m sure, the unbelievable happened: Trump won the electoral college. Why didn’t the data predict that?
Nate Silver was singing a different tune, though. I remember an interview he gave a month out from the election where he noted significant softness in support for Clinton. There were also a lot of undecideds who might swing elections in key states. That is, of course, exactly what happened. When the Comey letter was leaked by Congress, it likely cost Clinton the election. Her poll numbers dropped from +7% to +3%, well within the advantage that the Electoral College gives to Republicans.
On Election Day, the 538 model was about 3:1 in favor of Clinton. That sounds highly in favor of Clinton, and it is. But it still leaves plenty of room for a Trump win. And lo and behold, she lost.