Who will win the White House? The betting markets have already picked their winner

Analysis shows whether the bookies are rallying around Donald Trump or Kamala Harris as election day looms

Donald Trump will be the next president of the United States, the betting markets believe.
The bookies have rallied around the Republican candidate with the former president receiving his best odds since the end of July, analysis shows.
On Friday, the former president’s implied odds hit 59 per cent, the highest figure since July 30 – just two days before Democrat delegates began the formal process to endorse Kamala Harris.
Trump’s sudden takeover of Harris in the betting markets reflects a similar direction of polling of the election, although the latter continues to offer stubbornly inconclusive answers.
Nationally, Trump is within just two points of Harris, having been four points behind the vice-president just one month ago, Telegraph analysis of the polls shows.
In the seven key swing states, no candidate has more than a 1.5-point lead, although Trump has overtaken Harris in both Michigan and Pennsylvania with the slimmest of margins, enough to comfortably take the presidency.
Trump’s growing betting odds of winning the presidency therefore reflect an exaggerated version of what is happening in the polls, as well as gloomy commentary on the state of the Harris campaign.
Betting odds have long been used as a key indicator of the state of play in election campaigns, as punters make judgment calls on a variety of factors impacting the campaign.
Trump’s odds crossed the 50 per cent marker on Oct 6, Telegraph analysis of Betfair Exchange data shows, roughly the same day pollsters started showing a sustained, albeit tiny, lead for the former president in Pennsylvania. On the same day, Trump visited the site of his July assassination attempt in Pennsylvania which led to a high point in his polling.
Trump now has a 53 per cent chance of winning the state which, with its 19 electoral college votes, is seen as the key to winning the White House for both candidates.
Betting markets also have Trump winning Arizona, North Carolina and Georgia by odds of up to 70 per cent.
However, these “implied odds” are out-of-step with state-level polling leading by just one point in all three, easily enough to be considered a rounding error.
Political betting markets, in particular betting exchanges, are heavily influenced by where money is being placed.
This means any shift in direction from polls or media narrative which influences gamblers will see sudden movements in where people place their bets as they try to capitalise on seemingly low odds. This will quickly result in a movement of odds in favour of that candidate.
Polling and election narratives, however, have on notable occasions been incorrect, which in turn led to betting markets providing incredibly extreme odds in the wrong direction.
In 2016, polling greatly underestimated Trump’s chances in most swing states.
In October 2016, polls had Hillary Clinton six points ahead in Wisconsin and Pennsylvania, and two ahead in North Carolina. Trump went on to win all three states.
Incorrect polling, and a general assumption by the media that Trump’s campaign was doomed to fail, therefore heavily influenced betting. Ms Clinton went into the 2016 election with an “implied probability” of an 80 per cent chance of winning, the bookies said.
It wasn’t until 9pm on election night, two hours after polling closed in the first handful of states, that odds started shifting to Trump.
Betting odds therefore heavily influenced by incorrect polling and in the past two elections, that polling has been biassed against Trump.
The main issue is that certain groups are far easier for pollsters to reach, in particular college-educated voters, who were more likely to vote for Clinton. Republicans in general were less likely to reply to surveys.
Pollsters can correct these biases by weighting responses differently, effectively making up for the lack of responses by portions of society.
In this election, pollsters will likely be using two elections worth of data to determine how to factor in any lack of survey responsiveness amongst his base. This could, in turn, over-egg his chances if the MAGA-vote is less enthusiastic this time.
Pollsters are always clear of the margin of error in their surveys, although they seldom make headlines. For a well-run survey, a result can be up to six points plus or minus the best estimate, according to Pew Research.