Why nutty candidates are underperforming their polling so badly

Francesca Hong, the Democratic Socialists of America-backed candidate for Governor of Wisconsin, had led the field for months. The final Marquette Law School Poll (the gold standard in the state) had her at 38 percent and her nearest rival, Milwaukee County Executive David Crowley, at a measly 7 percent. The last survey, taken a few days before the election, had Hong leading by 22 points.

Then, on primary day, she lost in a squeaker.

There are only two explanations. One is that Hong simply got some really bad press in the final 72 hours of the campaign. The other is that the pollsters were wildly off.

Admittedly, it didn’t help Hong that her surprisingly recent online posts about abolishing Thanksgiving as a colonialist, fascist holiday suddenly resurfaced — the sort of garbage that no normal person can read without cringing. Also, there were other strange circumstances in that race, such as three major candidates dropping out, and one of them (Crowley) jumping back in at the last minute.

Even so, Hong was the clear leader, with more support in the polls than all of the other serious candidates in the race combined. Then she lost.

So I am still going to go with the explanation that the pollsters were very wrong — all the more because the difference between what the polls said and what actually happened in the Wisconsin race mirrors similar recent events in Michigan and in Minnesota. In three states, there were three primaries where candidates from their respective party’s extreme wing ran against what one might call the establishment or mainstream of the party. In all three, the polls badly overstated the performance of the extreme candidate.

Start with Michigan. In the Democratic Senate primary, Dr. Abdul El-Sayed (D) —........

© The Hill