Early voters wait outside the Franklin County, Ohio, Board of Elections on Election Day. (John Minchillo/AP)

Horror stories about people standing in long lines to vote started even before Election Day this year, with reports of massive waits at early-voting locations.

But new technology and research could help give officials the information they need to figure out how to make elections run better next time and one day help them respond to problems at polling places as they happen.

There’s remarkably little detailed data about how long Americans wait to vote, according to electoral experts. They say that's a big problem because fixing long lines at the polls is practically impossible without knowing where they actually happen. Previous research has generally shown longer waits in urban areas and for minority voters. But much of that data comes from media reports or surveys, according to John Fortier, the director of the Bipartisan Policy Center’s Democracy Project.

“Even administrators that run large counties often don't have a handle on what's going on at all at their polling places,” he said. In fact, many precincts do not have systems to track long lines, let alone prevent them, Fortier and other election watchers said.

But that’s starting to change.

For example, poll workers in Georgia's Fulton County used a mapping platform called ArcGIS to organize estimated wait times and share them with voters on its website for early voting and Election Day this year. Fulton workers estimated wait times by handing select voters a time-stamped card when they got in line and collecting it once they reached the front, according to Rick Barron, director of Fulton County's Department of Registration and Elections. Once an hour, the workers entered the estimates into the system, which automatically updated the information online.

The same workers who handed out cards to Fulton voters also counted the total number of people in line for research that Fortier is working on, Barron said.

Fortier's group and academics from the Caltech/MIT Voting Technology Project teamed up for a more extensive project that analyzes areas with long lines. They worked with jurisdictions that represent more than 20 percent of all registered American voters this election to collect data, such as line length, total check-ins and the number of voting machines per location.

Election officials could use the data from the project, called the Election Day Line Data Collection Program, to determine which precincts are overwhelmed on Election Day and how best to deploy resources such as voting machines next time, according to MIT professor and Voting Technology Project director Charles Stewart III. The data could help spot potential problems before they balloon into catastrophes, he said.

The data may also give officials ammunition to argue for more financial support so that they can run more efficient elections, Stewart said. Local election operations rely on tight budgets, which is a major reason there’s so little data available to help fix long lines, Stewart said. They may not have the staffing or technical expertise.

The research project doesn't plan to provide real-time wait estimates for voters at this point. Still, Stewart said, sharing that information can be particularly beneficial during early voting. Most people must vote at an assigned precinct on Election Day, but they can often choose from several early-voting locations. If wait times for those places are listed online, voters can figure out the quickest way to cast a ballot.

But, some experts — including Stewart -- are concerned that sharing those wait estimates on Election Day may persuade voters to skip the polls if sites or apps show massive waits at precincts. But Barron did not seem very worried about people being scared off by the estimates.

“I think if people want to vote, they're going to go vote -- and they're going to go wait in line,” he said, speaking in the run-up to Election Day.