That’s one statement made by Hillary Clinton. I think she might possibly have meant to say that there are lots of reasons why her campaign to be elected was unsuccessful, but she couldn’t bring herself to say this bluntly or clearly. She instead said that the thing that was not successful was “an election like this” — whatever that’s supposed to mean. No matter what she meant to say, the fact is she lost and indeed there are many reasons for that.
The primitive event study that I blogged is designed to gauge the impact that an event has on the existing level of voter preferences. It doesn’t explain and it’s not designed to explain that existing level. That’s where the “many reasons” come in. It’s like saying we do not know why IBM stock is selling at the $160 level; but if the company announces a computer breakthrough, we can measure the impact in changing that level.
We don’t know why Clinton and Trump were tied on October 27 (their levels), but we can measure the change in levels brought about by Comey’s announcement on October 28.
Trump and Clinton were very close at the outset of the USC poll on July 11. Their levels of voter support were not far apart. That meant that Trump had as much of a chance to win as did Clinton at that moment in time or if the election had been held on that date. But why was that? Why didn’t Clinton have a big lead? After all, she was the known quantity. She supposedly had the experience and qualifications. She was the inheritor of the Obama reign, and he was elected twice. Either their tie meant that Trump was exercising some appeal to voters that lifted him to her exalted level, or that Clinton had some negatives that were reducing her level to that of the novice Trump, or some combination of both of these possibilities. I think both possible paths contributed to making them poll at about the same levels: Trump’s appeals to certain voters (despite his own negatives) and Clinton’s negatives (despite her own appeals to certain voters).
It is not difficult to come up with many specifics on their possible appeals and negatives. Polling data can be used to pin this down to some extent because the polls ask voters many questions about many issues.
5:58 pm on November 12, 2016
