Some Republicans may still harbor fantasies about replacing their presidential nominee, but ballots have been printed, absentee and early voting has started in many states, and hundreds of thousands of people have already voted.
Early voting tea leaves will be a major part of the media coverage over the next month. But it will be important to resist making big predictions and reaching hasty conclusions from the available information. There are two major reasons for this.
First, context matters immensely. Raw numbers are misleading absent an awareness of historical patterns, legal reforms, and organizational factors, such as knowing which party tends to prefer which mode of voting in each state.
Second, even documented changes from one cycle to the next only tell us so much. For instance, what does it say if the number of registered Florida Democrats voting early soars compared to 2012? It could mean that Democrats have really expanded the voting universe in a way that will swing the state to them, or it could simply mean that voters who would have voted anyway are voting earlier, perhaps as a result of changes in Democrats’ turnout strategy.
Early voting reports can, however, be very instructive when read with caution and context. They can provide valuable clues as to which parties are struggling to turn out their voters, and which parties are seeing signs that their mobilization efforts are working.
The best way to put it is that early voting statistics will never tell you if a candidate is winning, but they can tell you whether a party’s base is showing up in sufficient numbers to make victory plausible. For instance, getting Democratic voters to turn out strong during early voting will not clinch North Carolina’s Senate race for Deborah Ross if she cannot also pull the same feat on Election Day, or if she cannot win swing voters by a sufficient margin. But if we see weak early voting by Democratic voters amidst the party’s usual efforts to get them to the polls, it would raise concerns as to whether Ross (and other state Democrats) can be victorious.
Let’s take a specific example. Most polls had Harry Reid losing Nevada in 2010. But unusually impressive early voting by registered Democrats made it plausible to think that the polls’ likely voter screens were too tight and that Reid’s ground game would carry him across the finish line. By contrast, four years later Nevada’s early voting statistics were dismal, as registered Democrats were evidently not showing up; Nevada journalist Jon Ralston warned that Democrats could not expect to win with such low turnout among their electorate, and indeed 2014 was a wipe-out for the party.
Here are some indications of what to look for in seven competitive states with major early voting operations. And make sure to check out at Michael McDonald’s detailed early voting database to track the latest information for each state.
Some states provide no information besides the raw number of voters as to who is early voting; other states detail the early voting electorate’s splits by party registration or race.
Colorado
Two years ago, the state of Colorado began sending every registered voter a ballot by mail, a drastic change in the state’s voting procedures. This makes comparisons to the last presidential election hazardous.
What we saw in 2014 is that Colorado Republicans returned their ballots at a far quicker pace, while Democrats picked up the rhythm later on. That led to major early anxieties for Democrats that were somewhat alleviated in the final days of the campaign when the gap between Democratic and Republican voters’ return rate narrowed. It was not enough to save Sen. Mark Udall, but the party avoided the wipe-out scenario it feared as Gov. John Hickenlooper won a second term.
Florida
Historically, registered Republicans have favored mail voting, while registered Democrats have made up the deficit when in-person early voting has started. Expect the same pattern this year, so Democrats should not be alarmed when they hear that registered Republicans have amassed a large lead in absentee voting. (In 2010, for instance, that lead was 21 percentage points.)
Barack Obama's 2012 campaign made an effort to encourage its base to cast ballots by mail rather than vote early in person, so the party has made some effort to change its usual strategy. Another factor that complicates interpreting this year’s data is a new state law that came in effect for the first time in 2014. It requires that anyone who requested an absentee ballot in the preceding two federal election years be sent one automatically this year. This law may change voters’ habits in a way that will make it difficult to compare this year’s patterns to past cycles.
Georgia
Georgia may not be the center of attention this year, but early voting statistics (which are broken down by race) could offer an indication of the degree of mobilization of African-Americans voters. In both 2012 and 2014, 33 percent of early voters were African-American.
Iowa
Unlike in other states like Colorado or Florida, Iowa Democrats have historically been far more mobilized to vote early than Republicans in Iowa.
By Oct. 10 in 2012, 54 percent of all early votes cast were from registered Democrats while just 27 percent were from registered Republicans. That huge margin narrowed, though, to 11 percentage points by the election. By contrast, more registered Republicans voted on Election Day itself.
So there will be two ways of assessing Iowa's daily absentee ballot reports this year. First: Are registered Democrats dominating as a share of the total vote cast? For now, the answer is a clear yes. Second: Are enough registered Democrats showing up to dispel polling-fueled fears that the party is in trouble in this state? At of now, we’ve seen a great drop in the number of Democrats who have requested an absentee ballot compared to the equivalent 2012 point, far greater in fact than among Republicans. But does this signal that Iowa registered Democrats are less eager to turn out this year, or does it signal that the Democratic Party has adopted a different schedule to mobilize its voters? We can’t know.
Note that the gap between the two parties in 2014 was far smaller than it had been in past cycles, spurring concerns throughout October about lagging Democratic turnout. These concerns turned out to have been largely justified on Election Day, when we learned that 68 percent of registered Republican had voted, versus just 57 percent of registered Democrats, and Republican Joni Ernst went on to win that year’s Senate race.
Maine
Maine has never split its electoral votes since a law passed giving an electoral vote to the winner of each congressional district. Polls have suggested that 2016 may break that pattern, with Trump better positioned in the northern 2nd District. So what will be very interesting to keep track of is whether there are signs of turnout divergence between the state’s two districts.
Looking at the available information, Democrats' turnout picture (in terms of the relative increase compared to 2012) in the 1st District is indeed slightly rosier than in the 2nd District. Will this hold as more voters request their ballots? We’ll have to wait and see.
Nevada
All eyes will be on Clark County, home to Las Vegas and almost three-quarters of the state's residents. For much of the two-week early voting period in 2014, Democrats seemed to have disappeared, and it was clear by Election Day that the party’s base was not turning out. The question this month is whether Clark County Democrats can build the sort of large early voting lead that they enjoyed in earlier cycles heading into Election Day.
North Carolina
North Carolina is another state with a history of GOP-friendly mail voting and Democratic-friendly in-person voting. That dynamic was dampened somewhat in 2014, when Democrats mobilized around the mail option more than they had in the past, and it appears that this change is holding true this cycle as well. Predictably, Republicans have worked to cut the availability of early voting locations since they took control of the state government. Which is to say: Early voting matters a great deal!
All eyes will be on the relative mobilization of white voters and African-American voters. The early voting electorate was 67 percent white in 2012, when Barack Obama lost the state by 2 percentage points; it was 71 percent in 2014, when Sen. Kay Hagan lost her re-election bid, also by 2 points. One further factor to keep an eye on is the reported differential in the parties’ ground game preparation. If Democrats are indeed out-organizing Republicans, this could create a greater-than-usual edge for Democrats among early voters; the information we have so far suggests that this may be occurring. But what will this mean? Will registered Republicans simply go to the polls on Election Day rather than mobilize earlier, or will the GOP suffer a significant turnout drop to the end?
We likely won’t know the answer to this—and to many other questions—until Nov. 8, but they are questions worth asking nevertheless.