Maps and Analysis of the 2016 Democratic and Republican gubernatorial primaries in New Hampshire, and Implications for the General Election

On September 13, 2016, voters in New Hampshire went to the polls to select the Democratic and Republican Party’s gubernatorial nominees. While Democrats have held the Governorship of New Hampshire for all but two of the past twenty years, the 2016 race was widely considered to be a toss-up even before the nominees had been determined. The results of both parties’ primaries reveal interesting information about the strengths and weaknesses of both candidates, and what could be expected in the general election.

I shall start with the Democratic Party. Three major Democrats ran for the gubernatorial nomination: District 2 Executive Councillor Colin Van Ostern of Concord, former Portsmouth mayor Steve Marchand, and former state representative and Director of Securities Regulation Mark Connolly of New Castle. This primary received scant polling, and what polling it did receive showed the vast majority of voters undecided. Thus, although Van Ostern received the majority of endorsements from politicians and interest groups, it was unclear who the victor would be until the results started coming in.


In the end, Van Ostern received approximately 52 percent of the vote, while Marchand received 25 percent, Connolly received 20 percent, and minor candidates received the remainder. Van Ostern won the vast majority of towns, while Marchand won a scattering of towns across the state, and Connolly won a few as well. As the map shows, Van Ostern performed exceptionally well in his Executive Council district, which runs across the state from Keene to Dover and picks up Concord along the way. Van Ostern’s best town was Harrisville, a small, ultra-liberal town east of Keene in the Monadnock region. Van Ostern also performed very well in his hometown of Concord, as well as in some of its neighbors, including Hopkinton and Canterbury. In general, but with a few exceptions, Colin performed very well in towns that vote Democratic in general elections, although this may be partly because the Executive Council district that he represents was drawn by Republicans to be a Democratic vote sink, and thus has lots of Democratic towns in it. Van Ostern posted a weaker performance in the Manchester-Nashua area as well as in the Republican towns of Rockingham County and on the Seacoast, although he still won almost all of those towns except for the ones on the Seacoast.

One of the heavily Democratic towns where Van Ostern did not do well was Portsmouth, home to Steve Marchand. He swept Portsmouth by wide margins and also won the neighboring towns of Greenland and Newington. Most of the other towns where he won have very small populations and even smaller Democratic primary electorates, so their voting for Marchand really doesn’t tell us much about those tiny towns.

The only town where Mark Connolly received an absolute majority of the vote was his hometown of New Castle, just east of Portsmouth, and the only other towns of any size that he won were the three Seacoast towns of Rye, North Hampton, and Hampton. This demonstrates one reason why neither Marchand nor Connolly was able to gain any traction on Van Ostern – their geographic bases were very similar and had a lot of overlap. This allowed them to push Van Ostern into third place in a few of the towns in that area, but also allowed Van Ostern to sweep almost the entire remainder of the state.

So what does this mean for the general election? A cynic might say that it means almost nothing, since the turnout rate for this primary was quite low. They have a point, however I think there are a few general trends that can be identified. Some people might assume the Van Ostern will overperform in his Executive Council district in the general election, but I think that, based on the primary results, any such overperformance will be more localized, particularly in the Concord area and some of the towns near Keene. Might Van Ostern also underperform in the Seacoast area, where his primary performance was weakest, even though his two opponents both lived in that area? I believe the answer is yes, and for another reason why that will probably be the case, read on about the Republican primary.


The Republican Party’s 2016 gubernatorial primary featured four major candidates: District 3 Executive Councillor Chris Sununu of Newfields, whose father and brother also held statewide office in New Hampshire; State Representative Frank Edelblut of Wilton; Manchester mayor Ted Gatsas; and State Senator Jeanie Forrester of Meredith. The race had long been assumed to be a two-way race between Sununu, whose last name gave him substantial name recognition, and Gatsas, the mayor of New Hampshire’s largest city. However, in the actual primary, Edelblut outperformed expectations and came in a very close second to Sununu, losing by only about 1,000 votes. This map shows which candidate won each town:


First, I’ll discuss Chris Sununu’s results. I should note that the following maps for each candidate all use the same color scale, to facilitate easy comparisons between them.


Chris Sununu received 31 percent of the vote. As the map shows, he generally had a floor of about 20 percent in most towns that were not in one of his opponents’ home regions. Most of his best areas were within his Executive Council district, as the map shows. He received over 50 percent of the vote in every town on the Seacoast (and remember, this was Van Ostern’s weakest region in the Democratic primary – another reason why Sununu will likely overperform here in the general election). He also performed well in Carroll County, which is an old-school, ancestrally Republican area. He won most of the wards of Concord, but only by small margins. In western New Hampshire, Sununu’s support was a bit squeezed between that of Edelblut, further southeast, and that of Forrester, to the north. Thus, Sununu won Sullivan County, but not either Cheshire or Grafton. He and Edelblut ended up in a virtual tie in Strafford County.


Edelblut’s map, at first glance, looks like a winning map. And if Gatsas hadn’t taken so many conservative votes in the Manchester area, Edelblut may well have defeated Sununu. Edelblut performed strongly pretty much everywhere in the southern half of New Hampshire, except for the Manchester area and the Seacoast, enough for 30 percent of the vote. His best area was the western half of Hillsborough County, where, unsurprisingly, his state House district is located. He performed very strongly with the small number of Republicans in heavily Democratic Cheshire County, decisively won the Nashua area, and substantially outperformed expectations in the swath of Republican-leaning towns east of Concord. He even won several towns in Coös County, including Berlin. But without a high level of support in the populous Manchester area, it wasn’t quite enough to defeat Chris Sununu.


Ted Gatsas’s support was undoubtedly the most concentrated in one municipality of any of the gubernatorial candidates, of either party, this year. The map shows this very well – Gatsas received between 40 and 60 percent of the vote in every ward in Manchester, and under 40 percent everywhere else except for two tiny Coös towns, resulting in a total of 21 percent of the vote. Not only that, but he only received over 30 percent in seven additional towns, five of which are in the Manchester area. Although it didn’t appear that way throughout most of the campaign, Gatsas seems to have been a regional candidate more than anything else. Gatsas was also the only one of the four Republicans who did not win any counties – despite Manchester being in Hillsborough County, Gatsas lost the county to Edelblut, who performed strongly in pretty much the entire remainder of the county.


In some ways, Jeanie Forrester was also a regional candidate, however her region was larger in size but smaller in population than Gatsas’ region. Forrester performed extremely well in her state senate district (the map includes the state senate district borders to demonstrate this fact). This was enough to allow her to win both Belknap and Grafton counties with pluralities, and for her to receive 18 percent of the vote statewide. However, despite her being endorsed by the New Hampshire Union Leader, she was unable to gain much support in the southern part of the state. She received over 20 percent of the vote in only a single town south of Concord – Londonderry, just south of Manchester. In much of the southern part of the state, she received fewer than 10 percent of the vote. Forrester was undoubtedly crowded out from competing in the populous southern part of the state by the three other candidates, all of whom were from southern New Hampshire.

So there you go. That’s how Van Ostern and Sununu won their parties’ primaries, what coalitions they used, and what we can expect in the general election. It’ll be close. Let’s see who wins. Either way, after the election I’ll be doing a deep dive into the results to see if I can use their data to draw any interesting conclusions about either Colin Van Ostern, Chris Sununu, or the great state of New Hampshire.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s