If you're trying to handicap November's presidential election eight months out, don't just look at the polls. Following the betting.
And as President Joe Biden gears up for Thursday night's State of the Union speech, the news for him is doubly grim.
First, he is now badly trailing Donald Trump, who has emerged as the clear betting favorite to win in November and be elected to a second term as president.
Second, the betting markets are now starting to price in a non-trivial chance that the situation for Biden is so bad that he will end up dropping out of the race altogether, for political or other reasons.
Trump leads Biden in the betting by a wide margin, with a 55% chance of winning compared to Biden's 33% chance. [You can see all the odds here]
(The percentages don't add to 100% in part because of the bookmakers' profit margins. For legal reasons, most of the betting on our elections has to take place with bookmakers based overseas, typically in London).
Biden fares better in the main, U.S.-based market, Predictit. But he still trails Trump there, 48% to 43% - exactly matching the two candidates' polling numbers in seven battleground states, according to a recent Bloomberg/Morning Consult poll.
Predictit is a small-odds, peer-to-peer betting market run as an experiment by a university (weirdly, New Zealand's Victoria University) under a federal waiver.
Meanwhile, Cambridge University professor Tom Auld, an expert in political betting, points out that gamblers now give the Democrats a much better chance of holding onto the White House than they are giving to Biden personally.
"Although Trump is well ahead of Biden, the Republicans are only slightly ahead of the Democrats," he tells me. The market, he says, is pricing in a significant and rising chance that Biden drops out. This could be for medical or political reasons. (He adds that the market also sees a smaller chance that Trump drops out for some reason).
It's an open question whether anyone cares much about the annual State of the Union speech outside of Washington and the media. But the grim political outlook for Biden raises the stakes for the address.
Based on comments from White House press secretary Karine Jean-Pierre, it's likely to be the usual exercise in political box-ticking. Jean-Pierre mentioned 17 different items that Biden plans to mention in his speech, from tackling "junk fees" to "ending cancer as we know it." Of more interest to most observers will be his performance. His key challenge is to reassure voters he will be able to serve as president for another four years.
Trump, meanwhile, is continuing to attack Biden over immigration, crime and inflation.
A Trump victory in November would be among the biggest comebacks in U.S. political history - surely matching or perhaps dwarfing those by, say, Richard Nixon in 1968, or Grover Cleveland in 1892. And it would be a result that was almost inconceivable three years ago, in the aftermath of the Jan. 6, 2021, riot by his supporters on Capitol Hill.
Betting is no more infallible than opinion polls, but as a political barometer it has several advantages. First, it's an attempt to predict a result, whereas what people tell pollsters right now is just about how they feel. Second, the people making bets are risking real money. It costs you nothing to say something to a pollster.
Still, this gloomy picture for Democrats among the bookies reflects the latest polling data. There was the survey in battleground states just mentioned. And a recent Wall Street Journal poll of nationwide registered voters gives Trump the edge, 47% to 45%.
Even worse for Democrats: In the past, Trump's actual vote share has significantly beaten his poll numbers. In the 2016 election, he beat his poll averages by about 2 percentage points. (He got 46% of the vote in the election. But in the polls conducted in the final days of the campaign he averaged just 44%. And this was in the polls conducted after FBI Director James Comey's October surprise, where he announced he was reopening his investigation into Hillary Clinton's emails).
In the 2020 election the gap was even bigger, three percentage points or more. Trump ended up with 46.8% of the vote. But the final polls just before the election gave him only 43.5%. (And for those polls in the previous week he averaged about 44%.)
The polls in 2020 were so off that they generated a lot of news coverage.
The explanation is almost certainly that a certain segment of Trump voters are either reluctant to admit it to pollsters, or just refuse to participate.
"Don't look at the polls, because nobody's going to admit they're going to vote for Donald Trump," a smart old friend told me in the summer of 2020. "If you really want to know who people are going to vote for, don't ask them who they're going to vote for. Ask them who they think their neighbors are going to vote for." A few polls which did this at the time showed much stronger support for Trump.
So, right now, the race is down to Trump-Biden. Trump is leading in polls which traditionally understate his true support, and he is way ahead at the bookies, where people are betting real money on what they think is going to happen.
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.