Though it’s not an unusual situation. Whenever you get a situation where people are inclined to bet with their heart rather than their head. The dumb people end up subsidizing the smart people. The classic example of this is I’ve seen is England coming into major football tournament. Not so much in recent year’s when they’ve been actually quite fancied, but in their “being knocked out by Iceland in the group stages” period you’ll still see odds of England winning the world cup or euros that are clearly inflated. And it’s because English people (who make up a larger proportion of the betting public relative to their population) are betting on who they want to win, not who will likely win.
So where are you getting a guaranteed risk free 4% return on investment in 9 months? I need to join your bank
Yes, that’s exactly the kind of explanation for a distorted price I look for when taking the other side.
But I see no obvious parallel in vast numbers of people wanting to back Obama for President at any price for sentimental reasons. The only weird thing I’m aware of is right wing conspiracy theories about her running - scaremongering (to those who are horrified by the idea of a Black female President), thinking it will undermine unity around Biden, that kind of thing.
You also see this with other big betting nations. See here for the 2026 world cup.
I mean China having the same odds as Iran (1 appearance vs 6 appearances) ? IMO that’s more to do with the number Chinese punters (of whom there are a lot, even before the rise of China as a economic superpower) betting on their home nation than some secret info about the Chinese football federation.
And the US are the hosts but does that really justify having 2x the likelihood as 2016 finalists Croatia?
Association football gets more and more popular in the US each year, and the US still has the population and resources to be world leaders at pretty much anything they’re really interested in: I’ve always thought the US is underpriced, I wouln’t be surprised to see them winning a Mens World Cup soon.
Total sidetrack, but I’d still say the US being rated as twice as likely as Croatia who made the final in 2016 is more to do with US punters betting for their home team, than the home advantage, the rise of US soccer, and aging Croatian golden generation.
Did you have to commit the 1000 pounds in cash? In other words are you forfeiting the interest you would have earned had you put the money into a GIC or other interest bearing account? (I can currently get around 5%)
Yes, I have to have as much as my maximum liability in Betfair funds. I’m hoping the Obama nonsense will fade within a couple of months, then I can cash out at the much higher price rather than waiting til November and forfeiting the interest.
But that’s the point. Most foreigners are clueless about American politics because they don’t care about it. So among foreigners, you are not likely to have a lot of dumb price-insensitive punters backing Obama for sentimental reasons.
Whereas there are always dispassionate pros with a lot of smart money who will research the details of the election of the attorney general of Papua New Guinea if they think there’s an edge.
Prices are most likely to get out of whack when you have a very large number of sentimental punters relative to the dispassionate smart money.
What you’re seeing in an example of favorite-longshot bias. There’s no agreed upon reason why his occurs, but it is indisputable that it does. The paper I’m about to introduce includes this explanation, “Wolfers and Zitzewitz (2007) show that in addition to nonrisk neutral investing, the favorite-longshot bias inherent to prediction markets is caused by transaction costs and liquidity concerns.”
In a recent paper in this journal, Erikson and Wlezien (2008) advance the debate between polls and prediction markets when they argue that while raw prediction market prices may provide more accurate forecasts than raw polls, adjusting the polls for known biases reverses this result. Thus, they argue that “market prices contain little information of value for forecasting beyond the information already available in the polls” (2008, p. 24). The problem is that Erikson and Wlezien do not advance the literature far enough. Their paper is the first empirical comparison that includes debiased polls and it is the first to focus on probability of victory, rather than just expected vote share. But the authors treat the well-documented favorite-longshot bias in prediction markets as a weakness of the markets rather than a systematic bias that can easily be corrected. In their conclusion they note the persistent problem of “The winner-takes-all market … overvaluing longshot candidates’ chances of victory” (p. 24).
He gives the following formula for that correction: p = \Phi(1.64 \cdot\Phi^{-1}(p_i)), where \Phi is the standard normal cumulative distribution function, p_i is the implied probability of the market price, and 1.64 is a best fit constant determined with out of sample data.
Let’s see what this means with respect to Obama’s chances in 2024. I can’t see BetFair prices, but I can see Polymarket which seem to be in line with the BetFair numbers I’ve seen in recent posts.
Polymarket currently has Obama yes at 0.068 and Obama no at 0.940. These sum to 1.008 so we’ll normalize yes/no to get p_i = 0.068 / 1.008 \approx 0.0675.
Thus we get p = \Phi(1.64 \cdot\Phi^{-1}(0.0675)) = 0.0071 < 1\%.
Now the market doesn’t seem so crazy. It’s just doing something (mildly irrational at best) that markets normally do.
