During the election season, voters are bombarded with poll after poll on who is leading. Now with early voting underway, Americans will be hearing additional poll numbers on "early voting voter turnout."
Over the course of the next several weeks, many eligible voters will cast their ballot early. They can do so during their state's designated early voting period by mail, physically voting at an early voting site or by requesting an absentee ballot, according to the National Conference of State Legislatures. Thirteen states do not allow early voting unless there is an approved excuse.
Pollsters from both parties will be reading the tea leaves of the early voter turnout, but are the early votes a guarantee of a win? If history is of any guide, the answer is no. So how should voters read those tea leaves?
CNBC asked voting turnout specialist Michael McDonald, an associate professor of political science at the University of Florida and a nonresident senior fellow at the Brookings Institution, what he will be looking for during the early voting polls. McDonald is no stranger to offering such guidance and expertise. He has worked for a national exit poll organization and has consulted to the U.S. Election Assistance Commission. Here are edited excerpts from that interview.
Who is the early voter?
There are two types of early voters, what I call situational and behavioral early voters.
Situational early voters are people who must vote early because they are away from their homes, like our military deployed abroad, or are too ill to make it to the polls.
Behavioral early voters are those who decide to vote early, if their state allows it. The earliest early voters tend to be highly informed about politics and the candidates and they cast their vote when they feel comfortable in their choice. We can see this the metrics — the first early voters tend to be older and more frequently registered with a political party. As the early vote season progresses, especially the last week before the election, we'll see more young people and people who do not register with a party in the mix, that is, people who are not as tied to the parties and tend to have less information about politics.
Voter mobilization is key for the early vote. What should Americans know about each party when it comes to early voting?
The early voting period provides campaigns with a longer period to make contacts with people to encourage them to vote. There is an important asymmetry between Democrats and Republicans.
Typically, Republicans tend to be older, more highly educated, and wealthier. They fit the profile of a likely voter, and an early voter. Republicans have thus felt that they do not need to invest as heavily as Democrats in voter registration and mobilization efforts, as they believe their voters will take care of themselves.
The Democrats have built more sophisticated voter targeting operations and do more physical contact due to the nature of their coalition of young people, who also as a consequence of the changing face of the nation, tend to be minorities. These efforts are effective and have been shown in numerous studies to increase voter turnout. The [Republican National Committee] has taken notice of the Democrats' efforts and have been investing more into voter mobilization. Republicans have fallen down more than once in the race for voter mobilization, but it appears that Republicans started getting their legs underneath themselves in some key 2014 Senate races. Republicans still need to catch up to the Democrats, but at least they are out of the starting block.
Donald Trump and the RNC have had very public riffs. Could this impact voter turnout for Trump?
It remains to be seen. An important part of Trump's coalition is rural, lower education whites. These are folks who have in the past been more a part of the Democratic coalition and needed voter mobilization. Trump is not investing heavily in traditional campaign ground activities. That is not necessarily a problem, since the RNC takes the lead on voter mobilization for their candidates whereas the [Democratic National Committee] lets the presidential candidates take the lead. However, there is widespread reporting on a rift between the RNC and the Trump campaign.
The Trump campaign might be hobbled if the RNC decides, for example, (Sen. Rob) Portman has the Ohio senate race sewn up and decides to shift their resources elsewhere. The wildcard is that Trump supporters are very enthusiastic. This shows up in the polls and was evident in the primaries, where Trump consistently won the early vote. It could be that his supporters will take care of themselves without needing a campaign contact. I'll be monitoring the individual level early vote data to see if this indeed happens.
You have said there are many pitfalls reading the early balloting tea leaves. Can you go through them?
There are many nuances and patterns to early voting that we need to recognize. Through provisions in federal law, the very first early votes will come predominantly from overseas civilian voters, a solidly Democratic voting block — there are some military voters, too, but there will be more overseas civilians.
After the initial rush of overseas voters, we'll see early voting by mostly mail ballots, which is Republicans' preferred voting method. So, for much of the end of September through the middle of October, early voting will look good for Trump and the Republicans. Once the in-persons early voting period starts about two weeks prior to the election, the early voting numbers will shift in favor of the Democrats, as this is their preferred voting method. These dynamics are likely how the election will unfold in a state that has mail balloting, in-person early voting, and Election Day voting, like Florida and North Carolina.
This is not universally true. A couple of examples illustrate. Pennsylvania only allows excuse-required absentee voting. Only a small percentage of people will vote early, and since they will vote a mail ballot, they will tend to look strongly Republican throughout the election. Colorado conducts all-mail ballot elections, and so there is no choice; that said, if 2014 is a guide, Republicans will vote earliest and Democrats will come roaring back in the waning days of the election. There are many nuances to early voting that I fully expect the campaigns to plaster over to make the case that they are winning.
Is there a voter demographic you are watching closely that could have an impact on this election?
There are two groups to track during the early voting period: Hispanics and rural whites. I've already discussed rural whites, so let me mention Hispanics. Aside from Cuban-Americans, Hispanics are a part of the Democratic coalition and their turnout rates have been low, about 20 points lower than whites or blacks in presidential elections. There is speculation that Trump's rhetoric could entice Hispanics to vote against him, and there is some evidence in polls that Hispanic voting enthusiasm is running higher than normal. In Georgia, Florida and North Carolina race and ethnicity are tracked on these states' voter files, so it is possible to look at the individual level data to see if Hispanics are showing up at unusually higher numbers than the past.
There has already been talk about voter fraud this election. Do you anticipate more?
Given the current state of the polls, which show a Clinton lead, I expect Trump will initially crow about the early voting numbers, holding them as evidence the polling is wrong. That is, until the in-person early voting starts. At that point, Trump will blame election "rigging" for the Democrats' early voting advantage. If you are serious about vote fraud, the truth is mail ballots are where most of the — extremely rare — election fraud occurs, rather than in-person voting. Yet, Republicans' vote fraud rhetoric and stringent voter ID laws are targeted at in-person voters and ignore mail ballots. It is perhaps not a coincidence, then, that mail balloting is Republican voters' favorite method of voting.
So when will we see the clearest picture of who is ahead in the polls based on early voting?
I generally believe that, properly interpreted, the early vote should confirm the polls. We should get a clearer picture of who is winning a week before the election by comparing the early vote this year to past elections.
That is, if the polling indicates a landslide election. If the polls are close, the early vote will likely not provide a strong signal as to who will win. There is collaboration between Slate and a new organization called VoteCastr that claims they will have accurate election forecasting the morning of the election based on modeling of the early vote in key battleground states. I'd caution about the precision of these estimates, as Election Day voters can be very different than the early vote. ... Unless the election is a blowout or early vote forecasts confirm the pre-election polling, I would take any forecast based on the early vote alone with a huge grain of salt.