The mood about who will win the 2024 presidential election may have shifted — but with just two weeks until Election Day, the polls are as inconclusive as ever.

The polling averages show a contest with Kamala Harris and Donald Trump almost tied in most key swing states. And although Trump's poll numbers have improved slightly in recent weeks, it wasn't enough to give him a clear lead. At least not yet.

On average, Trump has a small lead in Arizona, Georgia and North Carolina, generally between 1 and 2 points. To win the presidency, however, he must break the “blue wall” by winning either Michigan, Pennsylvania or Wisconsin. As of midday on Monday, these states are still about the same on average.

There haven't been many high-quality swing state polls in recent weeks, and the most respected pollsters are all likely to release new numbers toward the end of the campaign. Once that happens, we might get a better sense of where the breed stands.

If the polling averages continue to show an extremely close race after a recent spate of high-quality polls, that will only increase the uncertainty about what will happen. Polling errors – where a candidate or party is systematically underestimated – on the order of a few points are common.

The polls could again underestimate Trump, as they did in swing states in 2016 and 2020. Alternatively, the pollsters may have corrected — or even overcorrected — their earlier bias, risking an error that underestimates Harris. We just won't know until the votes are counted.

To recap, there are seven swing states that will most likely determine the outcome of the election (with every other state expected to solidly support either Trump or Harris). Listed in order of votes, these are:

If Harris can garner 44 electoral votes in these swing states, she will win the presidency. Trump would need 51 electoral votes from the same states to win. The polls say these swing states are – get ready – very close!

Pennsylvania, Wisconsin, Michigan, Nevada and North Carolina are all almost exactly tied in the New York Times polling averages. These show Trump leading by 1 in Georgia and 2 in Arizona. Nate Silver's poll averages come to similar conclusions.

The Washington Post averages show Harris ahead by two points in Pennsylvania, Michigan and Wisconsin. On the other hand, RealClearPolitics averages show Trump leading by 2 points or less in all swing states.

But these differences are hair-splitting. None of these results are one-sided enough to inspire real confidence about which direction the outcome will go.

The mood, however, points more clearly in one direction: Democrats are increasingly concerned about Harris' chances, while Republicans seem quite confident about Trump's chances.

But the mood before the election would have told you that Republicans were headed for a landslide victory in 2022, that Biden would win overwhelmingly (and not narrowly) in 2020, and that Trump had no chance of winning in 2016. These intangible hunches and assumptions simply aren't worth much.

Some point to alleged clues about the outcome in non-survey metrics, such as: B. early voting numbers or prediction markets. But early voter numbers are a notoriously poor indicator of what will happen on Election Day. Prediction markets generally reflect conventional wisdom — and this year in particular, they could be skewed by some Trump fans betting big.

Once we know the outcome, hindsight will be 20/20. We can all look back at the clues that were supposedly hidden from public view as we tell ourselves stories about how fate had been all along. At the moment, however, the race is simply too close to decide.