Why You Shouldn't Believe the Hype That the GOP Fixed Its Data Problem

The Republican National Committee has decided that Tuesday's win in Florida's special election, once expected to be an object lesson in the damage Obamacare will do to Democrats, instead proves that its new suite of data tools finally levels the playing field with the opposition.

This article is from the archive of our partner .

The Republican National Committee has decided that Tuesday's win in Florida's special election, once expected to be an object lesson in the damage Obamacare will do to Democrats, instead proves that its new suite of data tools finally levels the playing field with the opposition. Be skeptical.

After Mitt Romney lost in 2012, the Republican Party was pilloried for getting beaten the day of the election — in large part because its vaunted turnout tool, dubbed "Orca," failed in spectacular fashion. Now, the RNC suggests, it has turned that problem around. The conservative Washington Examiner details the five tools the RNC used to help elect David Jolly to Congress on Tuesday. They are:

  • A canvassing app, allowing people in the field to log visits with voters.
  • Application programming interface, which the Examiner says is a "fancy term the RNC is using" to describe how it shares data.
  • The email list, which is now synced with voter rolls.
  • Voter scoring, which we will get to below.
  • Dashboard, a real-time display of voter contact information.

This suite of tools, the RNC implies, is what made the difference in FL-13, getting people to the polls. In The Wall Street Journal, no less a figure than Karl Rove thinks that the RNC cracked the code on voter turnout.

Republicans also substantially erased the Democratic edge in get-out-the-vote. Ms. Sink had Team Obama and its Florida ground game, which delivered a 2,988 edge among the 131,713 voters who cast an early ballot. But Mr. Jolly crushed her by 6,445 of the 52,565 who turned on Election Day. Mr. Jolly, the Republican National Committee and the Florida Republican Party took on the volunteer-intensive portions of the ground game; outside groups took on the expensive, mechanical parts.

And as added proof, the party's chief digital officer, Chuck DeFeo, described to The Washington Post how close they got to their vote goals.

"One thing we haven’t released yet — the RNC set a vote goal" with the model, DeFeo said. "We felt Jolly needed 89,514 votes. He won with 89,099 votes. ... The power here is, we were using multiple methods of communicating to a specific universe of people we knew would help reach our goal.”

So is this legit? Has the RNC put together a digital package that will level the playing field with Democrats? It's hard to say, but there's good reason to be skeptical and to think that the RNC is hoping reporters don't know much about field or technology. (For example: "application programming interface" is not a fancy term the RNC uses. It is a generic term everyone uses.)

Here's how field campaigns work. Looking at poll numbers and working with campaign strategists, estimates are made of how many voters are likely to turn out and how they are likely to vote. Obviously a candidate needs the plurality of votes, so a strategy is developed, including digital tools, TV ads, direct mail, and direct contact to get to that magic number of votes. For the RNC, it was apparently 89,514. That's not a data breakthrough, it's spin. It's how campaigns develop targets.

Once that magic number is set, each part of the campaign moves into operation. The job of field is to get volunteers or paid canvassers on phones and in neighborhoods making specific pitches to voters who meet one of two criteria: they're likely to vote and could be persuaded to vote for your candidate, or they support your candidate and are only somewhat likely to vote. The canvassers make the case for why the person should go vote for the candidate, and move on.

Now here's the thing about FL-13. This was a special election in which, as The Wire noted on Wednesday, turnout was much lower than normal. That's naturally advantageous to Republicans, whose generally older voting base tends to turn out much more frequently. (Which, by the way, is one of the reasons the RNC has historically not worried much about turnout tools.) In an election like Tuesday's, no field campaign would worry about people who rarely vote. Instead, it would be about tracking the probable voters and just making sure they turn out.

A key way to get people to turn out is to take advantage of early voting. The more people you get to turn in early ballots, the less work you have to do on election day. Which suggests an alternate theory: that Jolly's Democratic opponent did a better job on field, getting that 3,000-vote advantage Rove mentions in the weeks leading up to the election. Then, the natural advantage Republicans have in low turnout elections kicked in on election day.

In part, that's based on the fact that contacting voters is very, very time intensive. On phones, you can get maybe 15 people an hour to talk. In person, knocking on doors, it's far less than that— say, 6. Meaning that it would take over 1,000 man-hours to turn out the 6,445-vote difference on Election Day, assuming none of the people would have otherwise voted. It doesn't matter what tools you have; lots of people aren't home or don't care about your visit.

The Examiner describes the "voter scoring" mentioned above:

The RNC used to rank voters on a simple one-to-five scale, one being the least likely to vote for the Republican candidate, five being the most likely. The committee now ranks prospective voters on a complex scale of one to 100, with each data point representing a piece of information about a voter that helps determine how best to talk to him or her and motivate him or her to vote.

In some ways, that's like Spinal Tap bragging about how their speakers go to 11. Yes, a deeper database of voter information is more helpful over the long term. In Florida, though, it is hard to believe that there existed more than five distinct scripts for canvassers to use on the doors to try and persuade people to vote — iPhone app notwithstanding. In a GOTV effort such as you'd see on election day, your script is this: "Go vote." Maybe only people rated 72 or higher got that push, but, then, how's that different than giving it to people rated 4 or higher? And it doesn't make people more likely to be home when you knock.

Coincidentally, there's a big piece in Time describing "Conservatives, Inc." — the massive network of consultants and marketing firms that have sprung up around the conservative movement hoping to wring out a little cash.

In the Obama era, lucrative outfits have sprung up to spread the falsehood that the President was born on foreign soil, seek his impeachment and investigate the attack on the U.S. diplomatic post in Benghazi, Libya. Winning doesn’t always matter; in fact, a campaign that has no measurable impact on elections or public opinion can be as remunerative as one that does. Activists make their appeals, collect the checks and move on.

Part of that involves convincing people that you have the ability to do what others haven't been able to: undermine Obama or whatever.

That's the business the RNC (and DNC!) is also in. They need people to give money; to do so, they need to show that they can win elections. Maybe this new set of tools will bear fruit over the long term. Maybe it turned businessman David Jolly into Rep. Jolly (FL-13). Or maybe it's marketing.

This article is from the archive of our partner The Wire.