How to bet on... Election MarketsRobert Blincoe gives a brilliant and potentially lucrative introduction to betting on world elections. There’s profit out there for those who know what they’re looking for
Political betting markets are a dream for punters who like access to lots of data and research to make informed decisions about where they’re putting their money. Polling information is prolific, sophisticated, and available.
The markets also reward those alert to market manipulation, scandal, gossip, policy changes and understand their impact.
And there’s always an election going on somewhere in the world. The US usually offers the most markets, volume of action, and market analysis, but there are always other opportunities the bookmakers might not be as clued up on as yourself.
At the time of writing the betting exchange Intrade had Hamid Karzai an 85 percent favourite to be sworn in as president of Afghanistan following the 2009 election. Whether you could lump on a sufficient sum to make use of this is another matter, but as you’ll see below, backing favourites is an excellent strategy.
Respect the favourite super bias
The key reason to be looking at political election markets is the presence of a super-bias towards favourites. The favourite-longshot bias was identified back in 1949 by a psychologist working at a veterans hospital in America’s horse racing heartland, Kentucky.
In a study of thousands of horses running at US racetracks, it was discovered that the shorter the odds at which a horse started a race, the better on average the value. Punters had been undervaluing favourites (or betting relatively too little on them), and overvaluing long shots (or betting relatively too much). This meant that by just backing favourites it was possible to earn above-average returns.
In election markets there’s a super-bias at work. According to professor Leighton Vaughan Williams, director of the political forecasting and betting research units at Nottingham Trent University, this was first highlighted in the 2004 US presidential election. “The favourite to win each of the states of the Union and the eventual winner was in every single case the same person. This is the equivalent of the favourite in 50 successive two-horse races winning.”
This kind of result just won't happen at the racetrack. But in the 2006 US Senate elections the feat was replicated, as every single contested seat fell to the polling day favourite.
In the 2008 presidential election it wasn’t quite the clean sweep – only 49 out of the 50 states returned the favourite.
This is not just an amazing set of coincidences – it’s an extreme bias. With hindsight, just backing the favourite in every general election race would have been a sure-fire strategy for winning. But once a bias is identified, the betting market tends to compensate for it, so you’ve got to combine knowledge of the bias, with a little extra research.
The market screams out for an accumulator approach. However there are always upsets. Back in the primaries to select the Democratic presidential candidate, Hilary Clinton beat Barack Obama in New Hampshire. Obama had been an 80 percent favourite to win, and with the bias, this would have been kicked up to more than 90 percent.
Do your homework
There’s a wealth of websites out there that are reporting the polls; averaging the findings of the polls; and those using sophisticated forecasting models to assign different weights to different polls according to their record of success in pinpointing the actual results. (See boxed text for Vaughan Williams recommendations).
Catch the arbs
Arbitrage is the practice of taking advantage of a price differential between two or more markets. In betting, these markets are known in slang as ‘arbs’, and bettors who take advantage of them are ‘arbers’. The views on an election can differ from within a country to those outside it. The classic example of this in political betting came during the 2000 US presidential election between Al Gore and George Bush. There was a moment when Bush was 6/4 in the US and Al Gore was 6/4 in the UK, which meant you could back both to the same stake and guarantee a 25 percent return, whatever the result. Opportunities should be watched for.
Exploit market manipulation
Because betting markets have earned a reputation as being the most accurate in elections, supporters, and campaign managers, of some candidates try and take advantage of this. If your guy becomes favourite, even if he’s not in ahead in the polls, it can change the result.
A terrific arb opportunity came during the 2008 US presidential election because of this idea. A John McCain supporter started pumping his own money into the Intrade market on McCain, making him favourite, and shoring up his support. This presented an almost ‘too good to be true’ opportunity for arbitrageurs. Lump on Obama on Intrade, and McCain on Betfair, say, and you were sure of a result.
Campaign managers also like to choose the best time to release the results of a poll they’ve commissioned to influence the markets, so beware of this.
If a price changes for no obvious reason, suspect that unpublished new information is about to go public, but hasn’t happened just yet. Make your move.
In some markets the volumes are so small that individual bets can change things. In the 2006 contest for the leadership of the UK’s Liberal Democrat party, such a thing happened. Chris Huhne, who’d only been an MP for eight months, had a starting price of 300-1, as a rank outsider. He became the bookies favourite to win, though it later emerged that the odds had changed largely as the result of one particularly large bet placed on Huhne to win, and another on eventual winner Menzies Campbell to lose. It got Huhne’s name known on the political landscape, and meant better value for those backing Campbell.
Understand when to trust the polls
Election markets are more accurate three or four days before an election than on the day itself. This is the intuition of Williams, backed by many hours of observing and crunching data. “There’s a momentum to the markets which end up overstating the outcome,” he says. He recommends using the information from three and four days back, and exploiting it on election day.
The betting markets can also be more accurate than polls. Take this example from the 1985 by-election, in the Welsh constituency of Brecon and Radnor, called to fill a vacant seat for the UK Parliament. The key players were the Labour and the Liberal candidates. On the day of the election, a poll conducted by the leading market expert MORI gave the Labour candidate a commanding 18 percent lead over the Liberal. Ladbrokes, however, were making the Liberal the odds-on favourite with the Labour candidate the relative longshot.
Professor Vaughan Williams takes up the story. “The late-breaking MORI poll did nothing to change the price. Who won? It was the Liberal, of course, and those who followed the money.”
|
