Quote:
Originally Posted by MarchHare
Worth noting: the way 538's model works, Biden's chances of victory will increase the closer we are to election day even if the polls remain unchanged. There's a purposeful amount of uncertainty built into the model to account for October surprises (aka Comey memos) that can derail an otherwise leading campaign late in the game. The closer we are to election day, the lower amount of uncertainty there is.
|
Very true. Something I've mentioned before is that their model gives this election more uncertainty than 2016 because of the number of huge news events going on (as measured by full-column NYT headlines). But I think that's a mistake, given that the polling so far has proved impervious to the news cycles, especially relative to 2016. It's hard to imagine a story or event (other than the death of a candidate) that would significantly swing the polls at this point.