In six weeks, Americans will vote in the midterms. The entire House of Representatives and a third of the Senate are up for grabs.
Forget the daily (polling) noise. Here are three numbers that tell the story.
40%
That’s roughly where Donald Trump’s job approval sits, depending on the poll.
I’ve been through midterm elections in presidential systems myself—not in the US, but elsewhere. The lesson is clear: the president may not be on the ballot, but the election is still largely about him.
At around 40% approval, Trump is a significant factor Republicans must contend with. 40% is better than George W. Bush at the time of the midterms in his second term, but worse than Barack Obama, Bill Clinton, and Ronald Reagan at the moment of their respective second midterm election.
$11.8 billion
That’s the estimated amount that will be spent on these midterms. It will likely be the most expensive congressional election ever.
More money. More consultants. More polling. More ads.
But does more money mean more effective campaigns? I’m skeptical.
Polling in the U.S. is getting sloppier by the cycle. And much political advertising remains painfully obvious political advertising—the kind voters recognize and tune out within seconds.
Which makes the third number particularly striking.
7,800
Models of the last House election suggest that roughly 7,800 voters switching from Republican to Democratic candidates across key districts could have changed control of the Lower Chamber.
That is the absurd mathematics of modern American politics.
More than $10 billion USD spent. More than 200 million voters. And potentially a few thousand voters separating one party from control of the House.
The political industry is enormous. The margin that matters can be microscopic.