Betting on Harris

A week after Joe Biden announced his withdrawal from the race, no threat has emerged to Kamala Harris’s position as the presumptive Democrat nominee. The endorsement on Friday from Barack and Michelle Obama pretty much completes the roster of senior Democrats supporting her. An online vote will confirm her as the candidate within the next few days.

There’s no doubting the increased confidence and enthusiasm that the change of candidate has brought to the Democrat campaign. And at least some of that shift has flowed through to the betting markets. Sportsbet this morning quotes Harris at 13-10 against, and Maxim Lott and John Stossel’s odds aggregator gives the Democrats a 42.7% chance of victory, up about ten points in the last fortnight.

All the same, those still seem remarkably generous odds. Her opponent, Donald Trump, has been repeatedly rejected by the electorate, and his age and unpleasantness are now even more in the spotlight. Harris has already closed the gap in the polls – which was never very large in the first place – to almost nothing in her first week. You would naturally think that she should be, if not outright favorite, at least very close to Trump in the betting. Yet Sportsbet still puts him at an almost unbackable 7-4 on.

What’s especially interesting is that this has happened before. This time four years ago, the polls consistently gave Biden a substantial lead, but the betting market didn’t always reflect that. In fact by the end of August it gave Trump a 45.8% chance of winning: more than it now gives Harris, and with no corresponding shift in the polls. (I wrote about this at the time here.)

Nor was that the only piece of odd behavior. Normalcy reasserted itself for a while, but two months later, on the night of election day, the odds sharply reversed themselves: Trump suddenly became the favorite, with his implied chance of winning at one point reaching 78.6%. It’s true that early counting was much more favorable to Trump than had been expected, enough to give him a real chance of victory, but at no point did that look more likely than not – certainly there was nothing to justify odds of that sort.

So for whatever reason, it looks as if there is a tendency for the betting odds to unreasonably inflate Trump’s chances. Perhaps, as I’ve suggested before, the people with more money than sense are disproportionately likely to be Trump supporters. It doesn’t follow that the odds are wrong now, but it’s at least something to keep in mind.*

Also, a quick note on the age issue. Much has been made of the fact that the Republicans are now the ones with an unprecedentedly old candidate, and that although Harris at 59 is hardly young, the 18-year difference between her and Trump is very obvious. So it set me wondering about what the precedents are like for this sort of difference.

Here’s the resulting table, showing each election since the Second World War in which there has been a gap of more than five years between the two presidential candidates (ages as of election day):

There’s a slight overall advantage to the older candidates, with them winning seven out of twelve. But in the most recent times that’s been reversed: the last four cases, which include the three largest age gaps, have all gone to the younger candidate. Maybe the electorate is becoming more comfortable with youth, or maybe it’s just chance.

The most striking thing about the table is that it’s almost always the Republicans who choose the older candidate. Not since Truman-Dewey in 1948 have the Democrats run with a nominee more than five years senior to their opponent, and they’re certainly not doing it this time.

.

* If you choose to participate in the market yourself, perhaps with the idea of proving me either right or wrong, please bet responsibly.