Early as it is in the 2024 presidential election cycle, it’s good to pay attention to general election trial heats, which I did at New York:

Despite some Republicans’ concerns about Donald Trump’s 91 felony indictment counts, and Democrats’ worries about Joe Biden’s age and job approval ratings, these two men are steadily heading towards a 2024 rematch. Fairly regular polling of a prospective Biden-Trump general election has consistently shown a close race: At the moment, the two are actually tied in the RealClearPolitics polling averages.

But as we should be constantly reminded, presidential elections are determined by the Electoral College, not the national popular vote (if they were, neither George W. Bush nor Donald Trump would have been president). So as the two parties begin to formally nominate their 2024 candidates, attention will shift to how well they might fare in the “battleground states” where the fight for 270 electoral votes will be waged.

We’ve now gotten a taste of that landscape via two big batches of state general-election trial heats from Emerson and Bloomberg/Morning Consult. Emerson seems to have focused much of its polling on states with down-ballot races, so the firm is mostly confirming that Trump is predictably far ahead of Biden in red states like Montana (by 21 points), Tennessee (by 33 points), and Wyoming (53 points). But Emerson also has polls of Wisconsin (Trump 42 percent, Biden 40 percent), Michigan (Biden 44 percent, Trump 43 percent), and a Pennsylvania shocker (Trump 45 percent, Biden 36 percent).

The new Bloomberg/Morning Consult polls are squarely centered on the seven states that were closest in 2020 (Arizona, Georgia, Michigan, Nevada, North Carolina, Pennsylvania, and Wisconsin). The results show Trump leading among registered voters in Arizona (47 percent to 43 percent), Georgia (48 percent to 43 percent), North Carolina (47 percent to 43 percent), Pennsylvania (46 percent to 45 percent), and Wisconsin (46 percent to 44 percent). Biden leads in Nevada (46 percent to 43 percent), and the two candidates are tied in Michigan (at 44 percent). Taking the margins of error into account, Trump has small leads in Arizona, Georgia, and North Carolina, and the candidates are statistically tied in Michigan, Nevada, Pennsylvania, and Wisconsin.

Aside from the rather important fact that the 2024 general election is more than a year away, there are a lot of variables to keep in mind when looking at such early data. For one thing, both Emerson and Bloomberg/Morning Consult are measuring sentiment among registered voters; you can make arguments for either candidate having an advantage when a set of likely voters emerge (Trump because Republicans are historically the most likely to vote and Biden because he’s unusually strong among the college-educated voters most likely to vote). For another, in both 2016 and 2020, Trump over-performed his national numbers in battleground states, which is why he was able to win in 2016 while losing the national popular vote by 2.1 percent and come close in 2020 despite losing the national popular vote by 4.5 percent.

It’s pretty likely that neither candidate is going to be wildly popular by November 2024, so the ultimate deciding factors could be (a) how many voters who dislike both Trump and Biden decide to turn out and (b) which candidate they prefer. Trump actually won the “hate ‘em both” vote by similar margins in 2016 and 2020, but lost in part because this segment of the electorate was a lot smaller in the latter year.

The even bigger variables, of course, could be turnout levels within each party’s base, and the possible impact of nonmajor-party candidates such as Robert F. Kennedy Jr., Cornel West, and someone sponsored by No Labels. On this latter front, it will be important to see exactly where minor-party-indie candidates gain ballot access; nobody’s really going to care if they rack up significant percentages of the vote in states either Biden or Trump is carrying easily.

It’s not too soon to begin paying attention to the places where the presidency will actually be determined next year. In the end, a national popular vote win may be a nice bonus for the next president or another bitter reminder that our system isn’t fully democratic.