Why we cannot rely on bookmakers’ odds

News

This article is an online version of our Inside Politics newsletter. Sign up here to get the newsletter sent straight to your inbox every weekday.

Good morning. The topic of today’s newsletter is political betting.

I used to do a lot of political betting before joining the Financial Times, and you can find an awful lot of wisdom by talking to other political punters. (Fun fact: Aaron Bell, the Conservative MP for Newcastle-under-Lyme, and one of the first of the 2019 intake to go on the record criticising Boris Johnson over Partygate, was a regular on the blog-cum-website PoliticalBetting.com before he was elected).


Inside Politics is edited by Georgina Quach. Follow Stephen on Twitter @stephenkb and please send gossip, thoughts and feedback to insidepolitics@ft.com.


Betting the House

The bookmakers’ odds are a useful journalistic crutch when it comes to commentary on the Conservative leadership race, not least because it allows another way of writing about Liz Truss. (Instead of using the words “Truss” or “the foreign secretary”, we’re able to describe her as “the bookmakers’ favourite” or “the bookies’ frontrunner” or some other variation thereof.)

But how predictive are the bookmakers odds? Not very. This handy chart by the FT’s Martin Stabe and Oliver Hawkins shows you how much they have fluctuated during the Tory leadership election:

The academic literature is also very clear. Betting markets are less predictive than polling, write Robert Erikson, political scientist at Columbia University and Christopher Wlezien, Hogg Professor of Government at the University of Texas at Austin.

In many ways, using the bookmakers’ odds as a predictive measure is a category error. The aim of a betting market isn’t to predict a specific outcome, it’s to make money. A lot of political betting is done on betting exchanges, which, unlike a traditional bookmaker, allow users both to make and accept bets from other punters — or back and lay, in gambling parlance. So punters aren’t always betting on something they think will happen — they are betting on something they think they can cash out on later down the line.

I don’t bet on UK politics anymore, but if I had bet on this leadership election, I would have bet on Kemi Badenoch early in the contest. Why? Not because I thought she had a plausible path to the contest, but because I thought her odds would shorten after the televised debates and would therefore be able to make a quick profit “cashing out” on my bet.

It’s hard to tell whether the betting market briefly gave Truss worse odds than Tom Tugendhat, a candidate whose only path to the final ballot of members involved both Rishi Sunak and Penny Mordaunt deciding to eat a small baby live on television, because some punters seriously thought that he was better placed than Truss, or because some thought they could make money cashing out before Tugendhat’s odds collapsed.

So in this leadership election or in any other, don’t pay too much attention to what the headline odds are. However, political betting is a useful analytic tool for tracking movement. For example, if you want a sense of who is “winning” a televised debate, the direction the betting odds are shifting is a useful bit of additional information — albeit one that will most be rendered obsolete by post-debate polling.

Our latest stories

Now try this

I’m in Cannes (or, at least, fingers crossed, I will be when you’re reading this). My plans for the week include eating, drinking, and reading by the beach or in the hotel bar.

As it stands, at the moment I am juggling the tricky question of how many books to take — is it time to try Ducks, Newburyport, because we have a long train journey to Cannes, or should I be realistic and accept I will mostly be gazing out of the window as we have seats on the top deck of the train?

I always think a good approach is to take a mixture of books that you want to read, an old favourite and some books by authors you know but haven’t got around to reading.

In the “authors I love, but haven’t read this one yet” pile, I have:

  • Manhattan Beach by Jennifer Egan

  • Glow by Ned Beauman

While in the “to try” pile I have:

  • Ducks, Newburyport by Lucy Ellmann

  • All The Light We Cannot See by Anthony Doerr

Plus John le Carré’s Tinker Tailor Soldier Spy as the old favourite.

Brilliant novelists including Douglas Stuart and Jamaica Kincaid will be speaking at the seventh annual FTWeekend festival on September 3. Join us in Hampstead Heath, London, for debates, wine tastings, jazz and more. As a newsletter subscriber, you can get £20 off your festival pass using the promo code FTWFxNewsletters (here’s the link).