By Liam Stack
Early voting for the midterm elections has begun in states across the country, and enthusiasm — and voter turnout — both appear to be high, with hundreds of thousands of mail-in ballots arriving in Florida and voters lining up around the block in Texas.
Turnout has surged among Republicans, Democrats and independents, according to poll data. As of Tuesday afternoon, more than seven million people had voted early, according to data compiled by Michael McDonald, a professor of political science at the University of Florida who studies elections.
“If these patterns persist, we could see a turnout rate at least equaling the turnout rate in 1966, which was 48 percent, and if we beat that then you have to go all the way back to 1914, when the turnout rate was 51 percent,” he said. “We could be looking at a turnout rate that virtually no one has ever experienced.”
Publicly available data on early voting suggests more likely Republican voters than likely Democrats have so far cast their ballots, but it is too early to draw any firm conclusions. Here is a guide to how early voting works and why the information we see now may not resemble the final result after Election Day.
How does early voting work?
Early voting works differently in each state, but it generally take two forms: in-person and by mail.
Generally speaking, mail-in votes tend to skew Republican, in part because mail-in voting is popular with older voters. Some states make it easy for older people to vote by mail but harder for younger people to do so: In Michigan, for example, anyone over 60 can get an absentee ballot. In-person early votes tend to favor Democrats.
How are early voting figures collected?
Early voting figures are gathered from state and local election authorities, who keep a record of who has voted. That data can be analyzed using computer models or publicly available information — like age, race or party registration — to make an informed guess about how someone voted.
But any analysis of early voting figures is missing the crucial evidence, because the available data does not include how a person actually voted. And it is important to keep in mind that not everyone who is a registered Democrat, say, always votes for Democratic candidates.
Which party is in the lead?
That’s a tricky question, in part because who you voted for is not public information, but also because the snapshot of early voting results on any specific day does not necessarily represent what the final result will be. Whoever is in the lead now may not be the party that actually wins.
Right now, Republican-seeming voters have cast more ballots than Democratic-seeming voters, but that’s largely because mail-in ballots are sent to voters weeks before Election Day (to give them time to cast a vote and mail it in.) That means when early voting first opens, mail-in ballots are usually the first votes to be counted.
“The ebb and flow of the election cycle generally is — and it’s not true in every single state but it is true in the national aggregate — Republicans tend to run up the score early with the absentee vote, Democrats come back strong with a pretty massive surge in in-person early voting, then on Election Day the vote tends to be closer to parity with a slight Republican lead,” said Tom Bonier, the chief executive of TargetSmart, a data analytics firm that has collected early voting figures. “Republicans tend to have to play catch up on Election Day.”
Age is also a factor, Mr. McDonald said.
“As we get closer to Election Day, you see that younger people tend to vote later than older people,” he said. “So younger people will enter the electorate as we get into next week.”
How does this compare with 2014 and 2016?
Early voting has become more popular in recent years, with more than 22 million people casting early votes during the 2016 election. And enthusiasm in the 2018 midterms is high. Voter turnout on Monday, the first day of early voting in Texas, was up 325 percent in Dallas County and 213 percent in Harris County, home to Houston, compared with the first day in 2014, according to figures provided by Texas Democrats.
Early voting turnout appears to have jumped considerably since 2014. But Mr. McDonald cautioned against making year-to-year comparisons because so many variables can change in each state from election to election.
“Even when we can look at prior data, the laws may have changed or the campaign strategies may have changed, so it may not reflect a true difference,” he said. “But my judgment looking at these numbers — and I’ve been doing this since 2008 — is that there is higher turnout.”
Stephanie Saul contributed reporting.