If it's in the News, it's in our Polls. Public opinion polling since 2003.


For Sanders, New Hampshire was a Glass Both Half-Full and Half-Empty

A Commentary By Kyle Kondik and J. Miles Coleman

The battle is breaking in his favor, even as his own performance has not been that strong.


— There are mixed signals from Bernie Sanders’ narrow victory in New Hampshire, but for now he’s supplanted Joe Biden as a weak frontrunner for the nomination.

— Overall, though, the race remains very uncertain as the scene shifts to the more diverse states of Nevada and South Carolina.

— Center-left candidates got substantially more support than progressive ones in New Hampshire, but the center-left vote split in such a way that Sanders was able to win. Pete Buttigieg has slightly more delegates so far, though.

Sanders now a weak frontrunner

In the wake of his victory in the New Hampshire primary, Bernie Sanders seemed to some like an unstoppable frontrunner. To others, he seemed like a spent force. That there is evidence backing up both arguments is a testament to the uncertainty of the Democratic race following the first-in-the-nation primary.

The argument for Sanders is not only that he won New Hampshire, but also that the person who seemed throughout 2019 like the biggest obstacle to him winning the nomination — Joe Biden — sputtered to a pitiful fifth-place finish, failing to crack double digits. Sanders’ other seeming top rival, Elizabeth Warren, is arguably in even worse shape than Biden: At least Biden can try to make last stands in more diverse Nevada and particularly South Carolina, where he hopes that the state’s majority African-American electorate can resuscitate his flagging campaign. Warren has no such redoubt: Iowa and New Hampshire could or should have been good states for her; she did OK in the former, and terribly in the latter.

Meanwhile, Pete Buttigieg probably would have won both Iowa — which he narrowly carried in terms of delegates over Sanders despite Sanders receiving more votes — and New Hampshire had Amy Klobuchar thrown in the towel after her mediocre fifth-place finish in the Hawkeye State. Instead, Klobuchar used a strong debate performance Friday night as a springboard into a solid third place in New Hampshire, finishing closer to Sanders and Buttigieg in first and second than to Warren and Biden in fourth and fifth.

Here again, Sanders benefits from this alignment of rivals: While Sanders has not demonstrated widespread appeal to African Americans — a backbone demographic of the Democratic Party that has not really been heard from yet — Buttigieg and Klobuchar are significantly weaker among black voters than Sanders is, at least right now. Sanders was threatened by a candidate who could dominate the black vote against him. Biden might’ve been that candidate, and hypothetically still could be, but it seems likelier now that the black vote might splinter, which probably helps Sanders.

As Michigan State University political scientist Matt Grossmann argued Tuesday night, “Bernie won & both his main national competitor (Biden) & his ideological faction competitor (Warren) lost badly. But field remains too muddled for him to see full assault. All good for Bernie.”

Note, though, that the pro-Sanders argument we just made above largely focuses on the challenges of his rivals, as opposed to his own performance.

While it’s unfair to Sanders to measure him by his 60% vote share in a head-to-head race in New Hampshire against Hillary Clinton, his 26% plurality in a much larger field put him a few points behind his pre-election polling average in New Hampshire, and this was less than half his share from four years ago. He turned in this weak performance even as Warren sank, meaning that Warren’s former support probably split among several other candidates as opposed to flowing mainly to her colleague, Sanders, on the leftward edge of the Democratic Senate caucus. So if Warren were to drop out, we can’t assume Sanders would disproportionately benefit. Sanders did great among younger voters and liberals, but not necessarily with other kinds of voters.

While turnout was up from 2016, the demographics of the turnout were not really favorable to Sanders, calling into question his claims that he can expand the electorate in his favor in both a primary and general election setting. And, as Grossmann added at the end of the tweet quoted above, “media coverage does not seem likely to help him bounce or expand his coalition.”

It makes some sense to compare this race to the 2016 GOP contest, at least in the sense that both featured large fields and a New Hampshire primary winner who party elites didn’t really like (Donald Trump and now Sanders).

Trump got 35% in New Hampshire four years ago, finishing about 20 points ahead of second-place finisher John Kasich. Sanders’ win was much more of a nail-biter, and he only got 26% of the vote, a historically weak showing for a Granite State primary winner. Trump benefited from GOP nominating rules in 2016 that made it easier for plurality winners to accumulate delegates earlier in the process. For instance, Trump followed up his New Hampshire win by taking 32% in South Carolina, but he won 100% of the state’s delegates while doing so.

The uniform Democratic rules, which require at least 15% support to be included in a proportional allocation, can distribute the delegates much more evenly. Sanders tied with Buttigieg in delegates in New Hampshire, and the upstart former South Bend, Indiana mayor actually leads the veteran senator by two in the overall delegate count (granted, there is a long time to go — the two lead-off states, combined, awarded only about 1.5% of the total pledged delegates available). There is more below as we update our UVA Center for Politics/Decision Desk HQ delegate tracker.

Biden clearly took a hit in Iowa, and New Hampshire only exacerbated his problems. National polling, where Biden has almost always led, has moved against him, and Sanders has taken the top spot. Michael Bloomberg, powered by his eye-popping television advertising, has seemed to benefit more from Biden’s fall.

Friend of the Crystal Ball Chaz Nuttycombe noted Wednesday morning that, over the last month, Bloomberg is up 7.5 points in the FiveThirtyEight national polling average, while Biden is down that same exact total, 7.5 points. The field is so fractured and fluid that it would be wrong to suggest Bloomberg is siphoning all his newfound support from former Biden supporters, but it’s also not a total coincidence, particularly as one notes the growth of Bloomberg among black voters in polling crosstabs while Biden slips among that same demographic. Honestly, it may be that the current numbers are catching Biden only in the midst of an ongoing freefall.

Politico’s Natasha Korecki, Marc Caputo, and Maya King reported a telling quote from Quentin James, who runs a PAC that backs black candidates: “Black voters are starting to leave him now. … A big reason lots of black voters were with Biden is they thought he was the best person to beat Trump. And they thought one reason for that is that he had the support of white voters. Now they see he has done so poorly with white voters and he no longer looks like the electability candidate.” Or, as Theodore Johnson, an expert on black voting patterns, put it, “For a bloc whose top priority is defeating the president, electoral pragmatism necessarily rules the day.”

Black voters may have been attracted to Biden out of pragmatism; could they move in sizable numbers to Bloomberg, Buttigieg, or Klobuchar out of that same kind of pragmatism, rooted in a belief held by at least some Democrats that Sanders is too left-wing to win a national general election? Potentially. The possibility of a single candidate winning a significant share of the black vote against Sanders has always seemed like a major threat to his nomination, and it remains a threat — just a diminished one thanks to the diminished standing of Biden.

How does one sort this out? Sanders has replaced Biden as the race’s weak frontrunner, and certainly Biden is nowhere near being a frontrunner. Sanders can bolster his case with a strong showing in the next contest, Nevada, nine days from now. We are flying blind into Nevada, a caucus state where there is no fresh polling as of this writing. Our best guess is that Sanders would win it if the caucus were held today but, as we’ve seen, this is a fluid race.

There will be another debate, next Wednesday, and Bloomberg may be on the stage. He probably would prefer not to be given his strategy of carpet-bombing the airwaves while fending off increasingly negative coverage, including widespread reporting of comments he made defending the New York City police’s “stop and frisk” policy that struck many as racist. Given how Klobuchar surged following last Friday’s New Hampshire debate, the other candidates perhaps can hold out hope that they can benefit from the next encounter. New Hampshire winnowed the field a little — Andrew Yang, Deval Patrick, and Michael Bennet ended their campaigns — but not substantially (only Yang, who acquitted himself well as a newcomer candidate, had any measurable level of support). We’ll also have to see whether Tom Steyer enters the mix now; he hardly registered in either Iowa or New Hampshire, but he has polled decently in Nevada and South Carolina.

For Sanders, the more viable candidates remain, the better his position. New Hampshire did not demonstrate his ability to win a wide breadth of support, but no one else can demonstrate such support either. In a race where everyone seems to have a ceiling, it helps to have a high floor, which Sanders does appear to possess.

Before the voting started, Biden appeared to benefit from split opposition. That description now fits Sanders. He seems better-positioned than the others, but not overwhelmingly so.

New Hampshire: A closer look

Throughout much of the night, Sanders retained a slim lead, which Buttigieg came tantalizingly close to surmounting. Perhaps not surprisingly, Sanders tended to run best in the rural counties that touch his home state of Vermont; he carried all but a handful of towns bordering Vermont. His best county in 2016, Cheshire — the southwestern corner of the state — again gave him the highest share of any county.

Despite New Hampshire’s proximity to her own home state, Warren didn’t perform especially well with the voters that would presumably be most familiar with her. In fact, the opposite was true — across the 18 Granite State towns that border Massachusetts, she took just 8%, a tick under her statewide share. Instead, much of the Greater Boston towns acted as a base for Buttigieg (Map 1).

Map 1: 2020 New Hampshire Democratic primary by town

Mirroring a trend from the Iowa caucus result, Buttigieg’s strength was distributed relatively evenly across the state’s geography. He had more second-place finishes than any other candidate, at 101. Though her third-place finish was impressive, Klobuchar didn’t finish second in any county, though she did so in 72 towns. In a break from some recently contested primaries, the state’s two congressional districts voted in unison: Sanders won by less than two percentage points over Buttigieg in each.

Map 2: 2020 New Hampshire Democratic primary, second place

While New Hampshire, like Iowa, told us little about how nonwhite voters might break, the state did tell us something about the differences among the candidates between those who do and do not have a four-year college degree.

Political distinctions among voters who do and do not have a degree have become highly salient. White voters without a degree powered Donald Trump in both the 2016 Republican primary and general election, although he pushed away some white degree-holders in the process. On the Democratic side, the non-college vs. college distinction has also been salient, particularly because the electorates in both Iowa and New Hampshire have both been roughly evenly divided between voters who do have degrees and those who do not. That’s according to the Associated Press/Fox News VoteCast, which is essentially an alternative to the exit polls that other news agencies conduct in conjunction with each other. Table 1 shows the Iowa and New Hampshire VoteCast results based on whether respondents did or did not have a degree. As polling throughout 2019 suggested, Bernie Sanders was stronger with voters who don’t have a degree. Elizabeth Warren and Amy Klobuchar are stronger with those who do. That seemed to be the case with Pete Buttigieg too, but his level of support among these two groups was basically the same in New Hampshire.

Table 1: IA/NH college vs. non-college preferences among top five candidates

Source: AP/Fox News VoteCast

The results in the 20 most highly-educated towns bears this out. Note that Buttigieg ran just a tiny bit ahead of his statewide share in these locales, while Klobuchar ran a little ahead and Sanders a little behind. This is based on the most recent U.S. American Community Survey estimates of four-year college attainment. Statewide, 36% had a bachelor’s degree or higher, compared to 63% in this group.

Table 2: New Hampshire primary results in 20 best-educated towns

Though educational attainment is an increasingly pertinent factor in general elections, the pattern that took hold in Tuesday’s primary was, in large part, an extension of the 2016 primary. In 2016, Sanders would have carried those 20 highly-educated towns by 14.5 percentage points over Hillary Clinton — this was significantly worse than the 22.5% margin that he carried the state by. The gap was smaller this year, but Sanders’ share in this sampling of towns was roughly 3.5 percentage points worse than his statewide showing.

Hanover, which houses Dartmouth College, is the most highly-educated municipality in the state — 81% of its residents over 25 years old hold a college degree. Buttigieg carried this Ivy League area but Sanders tended to dominate in towns that house public universities. As the youth vote is a valuable demographic to Democratic candidates, the contrast — or, perhaps, accord — between public and private colleges could be a persistent hallmark in other state primaries this season.

Strong turnout on both sides

After a disappointing turnout in Iowa, Democrats beat the high-turnout 2008 Democratic primary in terms of raw votes cast (about 300,000), although that might not be quite as impressive as it seems once one takes population growth into account. Still, those looking at turnout as a sign of overall Democratic engagement — we’re honestly not sure whether it has any predictive value for November — had to be more heartened by voter participation in this contest than in Iowa.

By the way, Donald Trump was also running in the New Hampshire Republican primary. He got 84% of the vote, which is a pretty good showing even for a shoo-in nominee (his performance was on the higher end of the range enjoyed by the last four reelected presidents: Ronald Reagan, Bill Clinton, George W. Bush, and Barack Obama). Polling has consistently shown Trump with very solid support from his own party, and the results from Iowa and New Hampshire reinforce the polling. Additionally, Trump got significantly more than double the vote of Bush in 2004 or Obama in 2012 (about 130,000 votes). So Republicans could point to strong engagement on Trump’s behalf as well.

Split opposition facilitates Sanders plurality

Going into the primary, Buttigieg, Klobuchar, Biden were competing for supremacy of the “moderate” lane while Sanders and Warren tailored their campaigns to more progressive-minded voters. Though Sanders ultimately prevailed, with 26%, in this very fractured field, 53% of Tuesday’s vote went to more center-left candidates, compared to 35% for the main progressive candidates (Map 3). If establishment elements within the Democratic Party are hoping to preclude Sanders’ potential pathway to the nomination, this is a sharp reminder that they must better unify their efforts behind a single candidate.

Map 3: Aggregate center-left vs. progressive vote share in New Hampshire

The delegate math

For nearly a week after the conclusion of the Iowa caucuses, we held off on counting IA national delegates due to the slow tabulation and identified errors. The Iowa Democratic Party put out final numbers — pending a partial recanvass — for the caucuses, including official national delegate counts. As a result, we updated our 2020 delegate count for the first time, confirming those numbers barring changes to the underlying SDEs in a recanvass. We also have a delegate count for New Hampshire, and full New Hampshire results can be found at results.decisiondeskhq.com.

While Bernie Sanders won the most first and final alignment votes in Iowa, national delegates are awarded based off of state delegate-equivalents, which does not necessarily align with the former two vote measures. Pending recanvassing, Pete Buttigieg has narrowly beaten Bernie Sanders in SDEs, 364.3 to 361.5. As a result, he gains a small edge in statewide delegates. Buttigieg did especially well in CD3, while CD2 was Sanders’ best district. Klobuchar gained one delegate as a result of her performance in CD4, the only location where she cracked the 15% threshold.

Table 3: Current Iowa delegate allocation

The popular vote to delegate formula also does not work in Bernie Sanders’ favor in New Hampshire. While Sanders won by about 1.5 points statewide, the math works out to a tie at the top with Pete Buttigieg on delegates. No difference is found based on congressional district.

Table 4: New Hampshire delegate allocation

With Iowa and New Hampshire now updated, Buttigieg has a slight lead in delegates pending the Iowa recanvass.

Kyle Kondik is a Political Analyst at the Center for Politics at the University of Virginia and the Managing Editor of Sabato's Crystal Ball.

J. Miles Coleman is an elections analyst for Decision Desk HQ and a political cartographer. Follow him on Twitter @jmilescoleman.

See Other Political Commentary by Kyle Kondik.

See Other Political Commentary by J. Miles Coleman.

See Other Political Commentary.

This article is reprinted from Sabato's Crystal Ball.

Views expressed in this column are those of the author, not those of Rasmussen Reports. Comments about this content should be directed to the author or syndicate.

Rasmussen Reports is a media company specializing in the collection, publication and distribution of public opinion information.

We conduct public opinion polls on a variety of topics to inform our audience on events in the news and other topics of interest. To ensure editorial control and independence, we pay for the polls ourselves and generate revenue through the sale of subscriptions, sponsorships, and advertising. Nightly polling on politics, business and lifestyle topics provides the content to update the Rasmussen Reports web site many times each day. If it's in the news, it's in our polls. Additionally, the data drives a daily update newsletter and various media outlets across the country.

Some information, including the Rasmussen Reports daily Presidential Tracking Poll and commentaries are available for free to the general public. Subscriptions are available for $4.95 a month or 34.95 a year that provide subscribers with exclusive access to more than 20 stories per week on upcoming elections, consumer confidence, and issues that affect us all. For those who are really into the numbers, Platinum Members can review demographic crosstabs and a full history of our data.

To learn more about our methodology, click here.