Skip to main content
ABC News
How Clinton And Trump Are Using Their Running Mates On The Campaign Trail

Remember how Tim Kaine was supposed to help the Democratic ticket appeal to Latino voters because he speaks fluent Spanish? And how, if selected as Hillary Clinton’s running mate, he would help her win votes in his home state of Virginia? Or how Mike Pence would be able to serve as an emissary to Midwestern and conservative voters if selected as Donald Trump’s running mate?

Did Clinton and Trump really believe the veepstakes punditry that Kaine and Pence could deliver these electoral advantages? And do these considerations help explain how the running mates are being used on the campaign trail right now?

To help answer these questions, we’ve been tracking where the presidential and vice presidential candidates have been traveling throughout the campaign. Our database of campaign visits differs from existing candidate trackers in that we record not only where the major-party candidates have been traveling, but also the exact locations of their events and the demographic and political characteristics of each locale visited. The logic of our research is simple: If a presidential campaign really believes that its running mate can help win votes from a particular demographic or political group, then it will send the vice presidential nominee to campaign in places where those voters are disproportionately represented compared to the places the presidential candidate visits.

So, for example, does the Clinton campaign believe that Kaine has a unique appeal to Latino voters? If it does, then we would expect Kaine to campaign in cities and towns with significantly higher Latino populations than those visited by Clinton. Likewise, does the Trump campaign believe that Pence can win over conservatives in rural areas and the Midwest? If so, then Pence should be visiting areas that are, on average, less densely populated, more concentrated in the Midwest and more conservative than those visited by Trump.

Not surprisingly, the presidential and vice presidential candidates on each ticket are similar in one respect: Typically, they campaign in the same battleground states. From the time he was introduced as Clinton’s running mate on July 22 and going up through Oct. 26, Kaine made the most campaign stops in Florida (18), North Carolina (11), Pennsylvania (10), and New Hampshire (7) — these swings states account for 46 out of his 71 total appearances.1 The same three states that top the list of Kaine’s solo visits also appear in the same order at the top of Clinton’s most frequently visited list (although Pennsylvania is tied with Ohio and Nevada), and they even account for roughly equal percentages of each candidate’s total solo visits. However, Clinton has made appearances in New Hampshire only twice in her 50 solo stops.

On the Republican side, since Pence was announced as the vice presidential candidate on July 16, 87 of his 98 campaign appearances without Trump have been in 12 battleground states also visited by Trump — most frequently in Ohio (18), North Carolina (13) and Virginia (10). Interestingly, Pence and Trump have visited Virginia far more often (17 of 200 total visits) than has Kaine (3), the state’s senator and former governor. Clinton has not visited Virginia at all, either solo or with Kaine, since selecting her running mate. The Clinton campaign appears to believe Virginia is safe; the Trump campaign, not so much. Meanwhile, Pence is the only candidate to visit Indiana, his home state (he appeared there twice). It is also worth noting that the Republican ticket has been much more active on the campaign trail than the Democratic ticket, with 200 versus 121 solo visits, and about the same number of joint appearances.

Now, let’s delve deeper. Where in these (mostly battleground) states are the running mates more likely to campaign? And what types of voters are they trying to reach? To find out, we constructed an empirical model for each presidential ticket that analyzed the campaign’s visits according to one of two outcomes: a city or town was visited by the presidential candidate or by the vice presidential candidate. The model estimates the statistical relationship between these outcomes and a series of demographic and political characteristics for each visited area, such as the percentage of Latinos in the local population or the median income level.2

For all the hype about Kaine’s appeal to Latino voters — and despite his high-profile exhibitions of bilingualism — statistically speaking, he is no more likely than Clinton to campaign in areas with large Latino populations. Nor, for that matter, does the empirical model provide evidence that, proportional to his total number of solo appearances, he campaigned more heavily in the South, or in more competitive battleground states, compared to Clinton. But there are some significant differences. First, in keeping with his centrist reputation, Kaine is significantly more likely than Clinton to campaign in less liberal congressional districts. Also, Kaine’s visits have been targeted toward areas with more college-educated adults and, oddly, toward Clinton’s home region of the Northeast.

Now what about the Republican ticket? Pence is, in fact, significantly more likely than Trump to campaign in the Midwest and among more college-educated populations. However, contrary to widespread perceptions of his electoral value, Pence is not campaigning in more ideologically conservative or rural areas than Trump. Nor is he visiting more competitive battleground states than Trump. Also, there is weak evidence (that is, the evidence is at marginal levels of statistical significance) to suggest that Trump is more like than his running mate to visit areas with older populations and higher income levels, as well as areas where the population includes higher percentages of African-Americans.

So, what types of voters are the vice presidential candidates targeting in 2016? Are these the same voters that Kaine and Pence were supposed to appeal to? The results are mixed. On the one hand, Kaine and Pence seem to be playing to their strengths by targeting moderate and Midwestern voters, respectively. On the other hand, they are no more likely than their presidential counterparts to campaign among the groups of voters — such as Latinos in Kaine’s case and rural residents in Pence’s — with which many expected them to have a particularly strong rapport.

These results help to further illustrate an important lesson from the 2016 campaign: While presidential and vice presidential candidates differ in some respects, ultimately they are part of the same team. Just as each ticket puts aside its differences to unite around a common policy agenda and campaign message (at least, they usually do), a party’s nominees are also mostly united on the campaign trail — even when they appear separately, they typically visit the same states and speak to similar voters. Indeed, our empirical models suggest that there are not many differences in the types of locations that presidential and vice presidential candidates tend to visit. And that makes sense. After all, there are no split decisions in presidential elections; come November, the Clinton-Kaine and Trump-Pence tickets will win together or lose together. Either way, both tickets seem to believe that they are — to borrow a phrase — stronger together.


  1. For purposes of this analysis, we have excluded events in which both candidates on a given ticket appeared together. Also, we exclude press conferences, national conferences and conventions, and other events that are not clearly organized or selected by a campaign for the purpose of appealing to a localized concentration of voters. We do include unscheduled events, such as visits to local businesses and restaurants. Each event’s occurrence is documented using multiple media sources.

  2. We estimated two logistic regression models — one for the Democratic ticket (N=121) and one for the Republican ticket (N=198); we dropped two Republican visits to Washington, D.C., because the city doesn’t have data for all the variables we examined) — to predict whether a campaign was more likely to send the presidential or vice presidential candidate to visit a given locale. Variables we considered include: median age; median income; population per square mile; percentage of the population that is African-American; percentage of the population that is Latino; percentage of the population that is college-educated; congressional district ideology, based on the DW-NOMINATE score for its Representative in the House; state margin of victory in the 2012 presidential election using Dave Leip’s Election Atlas (squared to measure the differences between swing states and states with a strong partisan lean); and whether the state is located in the presidential or vice presidential candidate’s home region. Demographic data come from the U.S. Census Bureau’s “American FactFinder” and “QuickFacts” online resources. Regions are defined by Census designations. In a small number of cases, we used county-level data because city- or town-level data were unavailable.

Christopher J. Devine is an assistant professor of political science at the University of Dayton.

Kyle C. Kopko is assistant dean for academic achievement and engagement and associate professor of political science at Elizabethtown College.