There may be two weeks left before Election Day, but the 2024 election is well underway. Around 18 million Americans had cast their ballots, either by mail or in early in-person voting, by Tuesday. There have already been some startling numbers, with the Georgia secretary of state noting first-day voting totals in the state were more than double the record. North Carolina also set a first-day record, though by a much narrower margin. There is still an open question, however, as to whether the early voting totals in the end will eclipse the numbers seen in 2020, when the nation was in the grips of the COVID-19 pandemic. More than 100 million votes were cast early in 2020 — roughly two-thirds of all votes in the election overall. The widespread assumption is that votes by mail this year will be significantly lower than the 65 million-plus that, according to the U.S. Election Project website, were cast in 2020. But the balance could shift in favor of in-person early voting — a choice made by almost 36 million Americans four years ago. The tightness of the polls between Vice President Harris and former President Trump means the early voting returns are beings scrutinized by partisans on both sides. Here’s what we know so far — and what we don’t. Early voting numbers tell us very little about the final result This is perhaps the most important point of all about early voting numbers. Partisan backers of one candidate or the other will always seize on the early statistics to predict the election’s overall outcome. They are almost always evincing a misplaced confidence. There are simply too many variables to make projections of an election’s outcome on this basis. The central difficulty is, there is simply too much that we cannot know.  For whom are the voters who are not registered as either Democratic or Republican casting their ballots? Do increased early vote numbers indicate unusual voter enthusiasm or simply a general shift in voting habits? There are still some observations that can be made, though, aside from those caveats. Republicans are making up a larger share of early voters this year Across the key battleground states that will decide the election, registered Republicans are generally casting a bigger share of early ballots than they did four years ago. This gets complicated, because some states do not have party registration and others do not release early vote data showing those affiliations. But in those that do, the share of early ballots cast by Republicans is edging up. In North Carolina, registered Republicans have cast almost as many votes as registered Democrats so far. According to the U.S. Election Project website — maintained by early voting expert and University of Florida professor Michael McDonald — roughly 34 percent of early ballots in the Tar Heel State have been cast by Republicans— almost double the 18 percent cast by GOP voters during the early voting period in 2020. A much milder version of the same pattern is seen in the key state of Pennsylvania — the biggest battleground, with 19 Electoral College votes — where registered Republicans had cast about 28 percent of early ballots by Tuesday, up from about 24 percent in the overall early voting period in 2020. This begs an obvious question. Are these numbers evidence of a Trump edge or merely a consequence of the former president having become much more encouraging of early voting? Trump held a rally in Georgia last week to mark the first day of early voting in the state.   “So if you have a ballot, return it immediately. If not, go tomorrow or as soon as you can. Go to the polls and vote,” Trump told his supporters. “Then, for the next 21 days, get everyone you know to get out and vote.” Nevada numbers cause a stir The state that gives Democrats most concern — and Republicans most encouragement — so far is Nevada. Among the battleground states, Nevada is one of the states where Harris is polling most strongly. But the early vote so far shows strong Republican turnout. As of late Monday statistics, registered Republicans had cast roughly 38 percent of early votes against roughly 36 percent by registered Democrats. Jon Ralston, a veteran political journalist in the state, wrote on the Nevada Independent website that those numbers “could signal serious danger for the Dems and for Kamala Harris here.” Ralston further noted that registered Republicans had not outvoted registered Democrats in early voting in the state since at least 2008. That said, the numbers are in a constant state of flux, and about a quarter of all early votes in Nevada come from voters unaffiliated with the two major parties. So the initial numbers are worrisome for Democrats but far from determinative. Don’t lose sight of the unaffiliated voters Enormous media attention gets placed on the voter breakdowns by party — but the numbers of unaffiliated voters too often get overlooked. According to McDonald’s figures, voters who are unaffiliated or registered with a minor party have cast about 21 percent of the early votes so far in Arizona, 24 percent in Nevada, 31 percent in North Carolina and 10 percent in Pennsylvania. Those figures come in the context of a battleground picture where neither candidate has a lead greater than 2 points in any of the seven key states, according to the polling averages maintained by The Hill and Decision Desk HQ. That being so, the preferences of those unaffiliated voters will have a seismic impact on the election’s outcome — and we have no idea of knowing what way they are casting their ballots. The Memo is a reported column by Niall Stanage.