The Betfair Prof: "It's been a dog's life for followers of the prediction markets!"

The Betfair Prof / Leighton Vaughan Williams / 17 March 2008 / 1 Comments

Prediction market guru Leighton Vaughan Williams tells us why the money didn't talk at Crufts and why favourites with the bookies are distinctly different from favourites on Betfair...

It's been a bit of a dog's life for advocates of the forecasting powers of the betting markets ever since Philippe the giant schnauzer walked off the other day with this year's best in show trophy at Crufts. Having extolled to all who would listen the likelihood that a so-called "super-poodle" from the utility dog class would cruise to victory on the strength of a flood of money which caused at least one bookmaker to close its books, followers of the markets were left as bemused as the bookie by the strange case of the money that definitely didn't talk.

Either way, it was always going to be a story to tell in the bars at Warwick University, which is hosting the Royal Economic Society conference this Easter week and where I'm contributing a paper about what we can learn from the ebbs and flows of money in the prediction markets. The real story, though, is not about the giant schnauzer but about the research paper which examines whether the bias in favour of favourites (i.e. level-stakes bets on favourites tend to produce a better return over time than equal-sized bets on long-shots) is as evident on the exchanges as it is with bookmakers.

Why should there be a difference? Perhaps because trades placed on the exchanges tend to reflect more informed, sophisticated money than that traded with bookmakers, perhaps because the margins are generally lower on the exchanges, perhaps something between the two. That's a question for another day, but whatever the reason it does indeed seem to be the case that while bets placed at shorter odds with bookmakers tend to yield a better expected return than bets placed at longer odds, this is less evidently so on the exchanges.

It was also less evidently so at this year's Cheltenham Festival where only two favourites obliged in one of the biggest Festival calamities for favourite-backers since 1962, when not a single favourite won. It was another long-shot that the roof of my hospitality box would blow away, but it did! Still, there's always the election betting markets.

Indeed, in 2004 the shorter-priced Presidential candidate in each individual state won each and every one of those contested seats, a feat reproduced in 2006 when the Senate candidate favoured on the exchanges won every contested Senate seat. Admittedly, the New Hampshire Democratic primary threw a spanner into the works when Hillary Clinton, the long-shot in spades, stormed home, but generally the weight of evidence does point to an extreme bias in favour of betting on favourites in election markets.

So what makes elections so different to dog shows and even to the odd racing festival? Who knows? Perhaps it's just the run of the cards and this year we'll see the little-favoured underdogs sweep the board in the US Presidential and congressional elections. Maybe so, but I won't be betting on it. Any more than I'll be betting that we'll be watching some super-poodle from the utility class being held aloft at the end of Crufts dog show next year - whatever the money says!

Professor Leighton Vaughan Williams is the Director of the Political Forecasting Unit and Betting Research Unit of Nottingham Business School, Nottingham Trent University, and Editor of the Journal of Prediction Markets

Betfair Politics Zone

Comments (1)

1. James | 24 March 2008

   Really like the column keep it up!