State Analysis: New Hampshire
New Hampshire has always been a bit more Republican than the rest of its New England neighbors, but its 2016 performance, in which Hillary Clinton won the state by a mere 0.37%, came as a big surprise to most of the nation. However, there is good reason to believe that in the overwhelmingly Democratic environment of 2020, there is little chance of this repeating.
With New Hampshire, it's important to consider the change the state has seen between 2012 and 2016; Zach Adams (@PVIGuy) has created a series of maps that allow us to examine the state's voting trend by PVI (Partisan Voting Index) on a county basis, and a quick glance shows that although the biggest counties in Rockingham and Merrimack actually stayed roughly static relative to the rest of the nation, areas like Sullivan, Coos, Belknap, and Cheshire saw a sharp swing towards Republicans that, when combined with a slightly more Republican environment than 2012, accounted for most of Trump's 2016 gains in the state.
|New Hampshire's 2012 (left) and 2016 (right) Presidential Election results by county PVI. Map made by Zach Adams (@PVIGuy)|
|New Hampshire elasticity by county|
Strafford (E 33.4) saw the margin narrow from a D +15 one in 2012 to a D +8 one in 2016. However, with its suburban structure and its proximity to the Boston area, a snapback towards Democrats is expected, and so one would be surprised to see Joe Biden win by anything less than the 15 points Barack Obama managed in a considerably less Democratic environment in 2012.
Hillsborough (E 30.4) and Cheshire (E 33.7) are two more counties that should see a significant snapback towards Democrats; although Obama won Hillsborough by 1 and the (significantly) less-populated Cheshire by 25, Clinton lost Hillsborough (albeit by less than a percent) and carried Cheshire by only 12. Hillsborough isn't too elastic of a county, but it is consistently only a couple points right of the nation, and when considering the ongoing suburban revolt against Republicans, Democrats can be hopeful of carrying it by possibly a bit over 5 points, a margin that would seal their victory in the state.
Perhaps the area of most interest, however, is Rockingham (E 27.0); although it is the second-most populous county in the state, it is also a bit to the right of Hillsborough and only slightly elastic. Clinton lost it by 5 points, but there are still enough elastic suburban voters here to think that if national polls and released Democratic internals of its district are to be trusted, Biden may become only the third Democrat to carry it since 1964, and the first since Barack Obama in 2008.
The elastic areas of Belknap (E 52.1), Sullivan (E 52.6), and Merrimack (E 41.5) are interesting to consider. Although Biden should have no problem carrying Merrimack and flipping Sullivan back, heavily Republican Belknap county may be a bridge too far for him; Trump, who won it by 17% in 2016, should easily carry this Republican bulwark once again.
The toughest area to analyze, however, is Coos (E 71.4). Although it voted for Obama in 2012 by a whopping 17 points, Trump swung it back to the Republicans and carried it by an astonishing 9 points, for a 26-point swing towards Republicans. However, if the national environment is, indeed, anything close to D +10, as polls are suggesting, Biden should be able to swing this extraordinarily elastic county back towards him and more closely replicate 2012.
Ultimately, the combination of a moderately elastic population and a heavy suburban concentration, combined with latent Democratic strength masked by Hillary Clinton's collapse in the state in 2016, means that Joe Biden should have little trouble carrying New Hampshire. A potential victory map for him could thus resemble a mix of Obama's 2012 and 2008 coalitions, and the sheer strength of a Democratic wave means that Biden could carry every county in the state except Belknap.
|A New Hampshire county prediction, made with Zach Adams (@PVIGuy)|
Prediction: Likely Democratic
A special thanks to Zach Adams (@PVIGuy) for his help with this post, from the PVI maps to the county-level predictions and analysis that we did together.