Has Kamala Harris turned back into an underdog even during her own convention? Late yesterday, I stumbled across this tweet from Clay Travis, who watches the betting markets a little closer than I normally do. Earlier in the day, one of the major markets had suddenly begun betting on Red rather than Blue:

Good morning. After two days of the DNC, Trump has now opened up a 5% lead over Kamala in ⁦@Polymarket⁩ gambling odds. Kamala has gone from a +10 favorite to -5 underdog in less than a week. pic.twitter.com/ugMyHwoEuh

— Clay Travis (@ClayTravis) August 21, 2024

After tossing it into the Final Word post, I decided to take a look at RealClearPolitics' tracking of the betting markets. They aggregate those as well, and the data demonstrates a sort of immediacy that polling sometimes lacks. And sure enough, it wasn't just Polymarket. After only a short period of favoring Harris to win, the markets had locked into a toss-up, when aggregated.

As of late afternoon yesterday, that is. This morning, Donald Trump has pulled back into a narrow favorite in the aggregation:

In case it's tough to read, the "smart money" now has Trump as a favorite, 50.3 to Harris' 48.3. That's almost too narrow to differentiate, of course, but the trending at least is not. It's the first time that Trump has been ahead at all since August 6. Not only does Trump lead in the aggregate, he's also a narrow favorite in five of the six markets RCP tracks. Only PredictIt has Harris up (52/48). She's going in the wrong direction even during the DNC convention, and that trend started even before Robert Kennedy began hinting at an alliance with Trump.

What does this mean? Friends of mine put great store in these markets as predictive indicators, especially given the commitment necessary to "put your money where your mouth is." Bettors certainly captured the zeitgeist of the presidential debate outcome better than polls did, as we can see in this graph. It may not have shaken too much of Biden's polling support, but the betting markets look much more accurate as to what the debate did to Democrats' chances after Biden's debate collapse. They also anticipated Harris as a candidate; that's actually a separate track for her in a blue-gray than Biden's dark-blue track, and bettors began placing money on her within a couple of days after the debate. 

This doesn't exactly match up with polling, however. RCP's head-to-head poll tracking shows what we'd expect during the Democrat convention -- some gains by Harris on Trump as the glitz and glamour gooses enthusiasm:

Worth noting, however: the five-candidate RCP track shows very little of this movement, although Harris maintains about the same gap in the lead. 

However, this takes me back to the immediacy of the betting markets and the longer roll time of traditional polling. It took a week or longer for Biden to bottom out in the polling, but that happened nearly overnight in the betting markets. It may well be that the bettors have a week or so lead time on changes in the public perception of the race, and if so, then the gap we're seeing now in the polling may be gone by this time next week. 

Or it could be that the bettors are wrong, too. Long shots win at the track on occasion, for instance. 

Still, this trend back to Trump seems to contradict expectations, especially this week, while the media keeps trying to bathe Kamala Harris in Joy and Light ... or perhaps it's just Joy Lite, and bettors are seeing through it. It bears watching, as do the polls after the conventions have packed up and the candidates have to start working seriously on the hustings.