Skip to main content
Menu
Trump’s Scorning Of Data May Not Hurt Him, But It’ll Hurt The GOP

Data doesn’t win elections; candidates do. Presumptive Republican nominee Donald Trump bet on that idea last week when he announced his plan to rely on his personality and rallies in the general election instead of collecting data on voters. Trump has a point: The effect of “big data” and improved analytics on elections is often overhyped. Even David Plouffe — the architect of President Obama’s 2008 and 2012 campaigns, the most data-savvy in history — agreed that Obama’s “data processing machine” was not responsible for his wins.1

But Republicans are worried, and for good reason: Trump’s assumption that the sole value of data is to win more votes is too narrow. His decision to limit the role of data probably won’t be the deciding factor in the 2016 election, but data organization and access are an investment in the future of the party. A presidential campaign presents a rare opportunity to cultivate the next generation of talent and collect a ton of new data on voters, and Trump’s refusal to do so means that Republicans may need to wait until 2020 or beyond to even the playing field with Democrats.

Presidential elections, when turnout routinely exceeds midterm and off-cycle elections, produce an influx of data to public voter files and staffers who earn experience analyzing that data. Two examples of innovations in data-driven strategy — direct mail after 1964 and digital data analytics after 2004 — demonstrate that even losing presidential campaigns have paid huge dividends.

Trump’s candidacy is often compared to the 1964 campaign of Barry Goldwater, and generally not as a compliment. Goldwater lost to Lyndon Johnson by nearly 23 percentage points, receiving only 27 million votes out of 70 million cast. Despite this historic loss, Goldwater’s campaign was a launching point for the national conservative movement largely because of the data it produced and the innovations it enabled.

Direct mail was a new tool for campaigns at the time. It was used to market a message directly to voters while raising money, much as online advertisements and emails do for today’s candidates. Messages are only as useful as the audience they reach, however, and the Republican Party was unorganized in most of the South and West. The “Draft Goldwater” committee, comprised of conservative activists who wanted to steer the Republican Party rightward, sought out “silent supporters” to become delegates and donors for Goldwater — creating reliable national lists of conservative Republicans in the process.

For these conservative activists, the Goldwater campaign was not a failure — it identified 27 million Americans willing to support a solidly conservative candidate, and generated data on where to find them. One activist, Richard Viguerie, went to the clerk of the House of Representatives and copied (by hand) the names of donors who gave Goldwater $50 or more, forming the basis for his mailing list. In his book, Viguerie claims that he used this data to send 70 million letters per year and 1 billion pieces of conservative direct mail between 1974 and 1980, when conservative Ronald Reagan was elected president. Goldwater’s ideological legacy was secured using strategies that relied upon the data generated by his failed campaign.

Political data has moved far beyond card catalogues and mailing lists. Making sense of today’s data requires experience with database software and statistical modeling. The software used by campaigns cannot be bought “off the shelf” or assembled quickly because it relies on data controlled and managed by the parties.2

In 2005, after winning two consecutive presidential elections, the Republican Party’s data was more centralized and better organized than Democrats’. This success led to a sense of complacency among Republicans. Meanwhile, after two straight losses, Democrats were open to new campaigning techniques at the same time that online and digital tools were emerging.

Like conservatives post-Goldwater, the Democrats used their data effectively after a loss in a presidential election. In this case, failed 2004 Democratic candidate Howard Dean — whose campaign was known for its innovative use of the internet — became chair of the Democratic National Committee in 2005. He prioritized the creation of a national voter database with one company in charge (NGP VAN). This investment made it easier for future candidates to incorporate data analytics into their strategy, giving them incentive to hire employees who could innovate and build on each other’s work — even after elections ended. Since 2004, Democrats have founded 67 firms and organizations dedicated to digital campaigning and data analytics, while former Republican staffers founded 13.3 These firms apply their expertise down the ballot by providing services to Democratic gubernatorial, Senate and House candidates, all of whom work from the same dataset.

Democrats now hold a substantial expertise advantage in digital data-driven campaigning, and the GOP admitted as much in their 2012 election post-mortem. John McCain hired only 15 data staffers in 2008, compared with Obama’s 131. To his credit, Mitt Romney increased the number of data hires to 87 in 2012. (Obama had 342). In 2016, Republicans were positioned to build on this effort and narrow the analysis gap between the parties, pivoting off of two consecutive losses into an innovative data strategy — just like in 1964 and 2004.

But Republicans seem set to squander the opportunity. Trump currently employs as few as two staffers dedicated to data, according to reports. (The Trump campaign did not respond to a request to confirm the number of staffers it has devoted to data.) The Republican Party has not consolidated data as Democrats did, instead relying on “a jumble of firms not always working in concert.” Independent firms such as i360, funded by the Koch brothers, have not integrated with the party database.

Meanwhile, Democratic consolidation and expertise building continues. Hillary Clinton set out to assemble a data team three times the size of Obama’s formidable 2012 operation. Several members of her digital and analytics leadership team worked for Civis Analytics, BlueLabs, and Blue State Digital4 — all firms founded by former Obama or Dean employees. After this election, win or lose, Clinton’s data staff will be positioned to bring their expertise back to these companies or found new ones while Republicans continue to chase the data advantage they wasted after 2004.

Listen to our podcast special on the history of political data in U.S. elections.

Footnotes

  1. Plouffe did not agree fully with Trump, noting that “flying blind is nuts.”

  2. Daniel Kreiss’ forthcoming book, “Prototype Politics,” explores many of the same points here, as did his blog post last week. For more on the historical dynamics of political data expertise (or just for an excellent read), you should learn more about the book here (and pre-order it here).

  3. For a full list of these companies, see the Appendix of “The Tech Industry Meets Presidential Politics: Explaining the Democratic Party’s Technological Advantage in Electoral Campaigning, 2004-2012” by Daniel Kreiss and Christopher Jasinski.

  4. Analytics Communications Director Matt Dover, Director of Analytics Elan Kriegel and Director of Frontend Engineering and Optimization Kyle Rush, respectively.

Joshua Darr is an assistant professor of political communication in the Manship School of Mass Communication and the Department of Political Science at Louisiana State University. His research focuses on campaign strategy.

Comments