If you look at the headlines, Vice President Kamala Harris and former President Donald Trump are in a dead heat, at least as far as the polls are concerned.

But as I pointed out last week, Kalshi’s markets give a clear edge to Trump in terms of his probability of winning. To be clear, that’s not necessarily contradictory to the polls (more on that later). But even so, our traders think Trump has solid momentum in the swing states.

His chances of winning North Carolina, Georgia, and Arizona are all approaching 70%, and his chances of winning Nevada are now over 60%.

He is also the favorite in Pennsylvania, Wisconsin, and Michigan, albeit by much slimmer odds.

As such, it’s no surprise that traders currently give him a 59% chance to win overall.

But it’s not time for MAGA to do any victory laps just yet.

Despite Trump’s momentum in the South and Southwest, Harris still has a narrowing, but viable, path to victory through the Rust Belt.

In the interest of providing Harris supporters with a little bit of copium, here’s a short breakdown of how (and why) she can still win.

^{This is an opinion, and is not financial advice. The author uses a pseudonym and cannot trade on Kalshi.}

Harris only needs 3 of the 7 swing states…

As mentioned above, Trump’s chances of winning in North Carolina, Georgia, and Arizona are nearing—and in some cases surpassing—70%.

But Harris doesn’t actually need any of those states to win. She doesn’t even need Nevada, where, thanks to recent mail-in voting data, Trump currently has a 62% chance of winning.

Assuming she carries Nebraska 2, which is very likely, Harris only needs Michigan, Wisconsin, and Pennsylvania to get to 270.

They call these states the “Blue Wall” for a reason. These are traditional Democratic strongholds.

Yes, Trump shocked the world by knocking down the wall in 2016, but Biden managed to narrowly rebuild it in 2020. In 2024, Trump is certainly capable of winning them back. In fact, he's currently ahead in these states. Fortunately for Harris, Trump’s odds aren’t insurmountable.

Wisconsin and Michigan are basically coin flips, and his chances in PA are still under 60%. So don’t expect the Dems to give up without fighting tooth and nail. If she can shift momentum in Pennsylvania, her path to 270 looks promising.

But with two weeks left, can she close such a large gap?

Yes, because…

A prediction market is not a poll…

That brings us to Pennsylvania, where our market shows Trump with a 57% chance of winning, leaving Harris at 43%.

Now, if this was a political poll, a 57% to 43% lead would be practically insurmountable, especially this late in the game. That’s because polls show (or attempt to show) the percentage of voters who plan to vote for each candidate.

So if Trump was polling at 57%, Harris would need to swing 14% of Pennsylvania voters in two weeks, which is damn near impossible.

But a prediction market is not attempting to show a candidate’s potential voter share. It is meant to show a candidate's current odds of winning.

For example, if a candidate was polling at 55%, that would translate to a somewhat comfortable lead, depending on the margin of error. But a 55% chance of winning is far from comfortable. In fact, it’s little better than a coin flip.

If you’re a veteran of prediction markets, this fact may seem obvious to you. But even some experts seem to confuse the two. For example, the New York Times recently ran a piece equating polling numbers with the probabilities generated by prediction markets.

As X user @AlecStapp points out, this article is from the outlet’s financial markets team, not the political team, which makes the error (which has since been corrected) even more surprising.

Confusion aside, what does this mean for Harris?

It means that while she is down in PA, she is certainly not out.

On October 22, 2020, Trump was the favorite to win Arizona with 62%. By election day, that number had climbed to 92%. But in the end, Biden won the state.

Again, I'm not trying to sugar-coat this. Clearly Trump is in a better position.

But a 58% chance of winning is far from a sure thing. Harris still has a better chance of winning Pennsylvania than Biden had of winning Arizona at this time in 2020, not to mention the Democrats' strong historical track record and infrastructure in PA.

Prediction markets are not infallible...

At Kalshi, we pride ourselves on the predictive power of our markets. They offer a real-time snapshot into the state of the election in a way that a poll simply can’t.

And based on historical data, election markets have a solid track record of accurately picking the winner.

As economists Paul W. Rhode and Koleman S. Strumpf have documented:

In the 15 elections between 1884 and 1940, the mid-October betting favorite won 11 times (73 percent), and the underdog won only once (when in 1916, Wilson upset Hughes on the west coast).

That said, most of the elections that were not accurately predicted were simply too close to call.

In the remaining three contests (1884–1892), the odds were essentially even throughout and the races very close.

Are the odds for this year's race “essentially even?”

Arguably, yes. Obviously, if things continue to trend much more in Trump's favor, it will be hard to make that case.

That said, even if Trump’s odds keep rising, a last-minute swing fueled by a turnaround in Pennsylvania is possible, as evidenced by Wilson’s 1916 comeback.

For a more recent example, let’s not forget what happened in 2016.

In mid-October of 2016, Kalshi did not exist. But on other platforms, traders priced Trump’s odds of winning the White House at just 18%. On the day before the election, Clinton’s chances were at 82%. And we all know what happened there.

And if you don't know what happened, I've got just the book for you!

A similar upset took place in 2022, when several prediction markets had the GOP favored to win the Senate. In the end, the Democrats greatly outperformed, and even added to their Senate majority.

So yes, election markets are a powerful predictive tool, and based on what they are currently showing, I’d much rather be in Donald Trump's shoes. But these markets reflect probabilities, not certainties. And anyone who is certain of a Trump victory might be in for a rude awakening in November.

Do you disagree? Feel free to yell at me on X. But before you do, click here to put your money where your mouth is.

Follow Terry Oldreal on X: @realoldterry

^{The opinions and perspectives presented in this article belong solely to the author, who is using a pseudonym and cannot trade on Kalshi. This is not financial advice. Trading on Kalshi involves risk and may not be appropriate for all. Members risk losing their cost to enter any transaction, including fees. You should carefully consider whether trading on Kalshi is appropriate for you in light of your investment experience and financial resources. Any trading decisions you make are solely your responsibility and at your own risk. Information is provided for convenience only on an "AS IS" basis. Past performance is not necessarily indicative of future results. Kalshi is subject to U.S. regulatory oversight by the CFTC.}