Perhaps there's something in the water of the Chesapeake Bay that affects electoral politics. The Virginia Senate race, expected to be a romp by incumbent Democrat Mark Warner, wasn't; it still isn't settled. And the Maryland governor's race, expected to be a romp for Democrat Anthony Brown, was instead a romp for his Republican opponent, Larry Hogan.
We explored the Warner race here. In short: Turnout dropped where Warner needed it most, contributing to the fact that even counties he won voted far more for the Republican than they did in the Senate race two years prior.
In Maryland, Brown simply got beaten. This wasn't turnout; we don't need any complex divinations to determine what happened. Maryland voters preferred Republican Larry Hogan, and voted him into office.
In our look at Warner's race, we compared turnout and margin in counties Warner won versus his opponent, Republican Ed Gillespie. Here's the same comparison for Brown and Hogan. The first chart shows the average drop-off in each county backing each candidate between 2010 (the previous gubernatorial race won by Democrat Martin O'Malley) and 2014. The second chart shows how much each candidate's counties moved to the right.
Now you might be thinking: Well, Brown counties had a much bigger drop-off in turnout! Which was a set-up on our part. Because, you see, Brown only won four counties. Which isn't itself a big drop-off from 2010, when O'Malley won five. But it suggests that the turnout drop isn't as big a factor as you might expect -- particularly given that those four counties voted almost 13 percentage points more for the Republican candidate this year.
To put it another way: If you increased turnout three percent only in counties that Brown won, he'd still lose by almost 67,000 votes. If you increased turnout by three percent only for Brown in counties that he won, he loses by almost 62,000 votes. If you increase turnout by three percent only for Brown and in every county ... he still loses by over 52,000 votes.
Shift toward Republican candidate between 2010 and 2014
In fact, you would need to increase turnout for Brown by ten percent in every county before he actually has more votes than Hogan. Because Hogan won handily. (This is as good a place as any to note that I worked on two campaigns with Anthony Brown's campaign manager, but that we haven't been in contact in several years.)
So why did pollsters think Brown was going to win? Real Clear Politics, which tracks polling in races, gave Brown an average margin of 9.7 percent last Friday -- based mostly on fairly old polls. FiveThirtyEight's Harry Enten wrote about the polling in Maryland on Wednesday. "The only late polls," he writes, "came from the Maryland Republican Party or from Hogan’s campaign." And "it’s clear the polls sponsored by Republicans caught Hogan’s rise. The final poll taken a week before the election by Wilson Perkins Allen Opinion Research (WPA) for Hogan’s campaign gave the candidate a 5 percentage-point lead."
But people (for very good reasons) generally discount partisan polls in favor of ones from independent pollsters. (The Post polled in early October and had Brown up by a wide margin.) So no one expected Brown to lose because no non-partisan polls existed to suggest that it would happen.
Anthony Brown lost to the candidate that the voters preferred, simple as that. May not have been anything in the water after all.