An exploitable arbitrage opportunity emerged over the course of the recent US presidential election. During the election season, I primarily rely on Nate Silver's blog, FiveThirtyEight for odds because he is the best statistician in terms of both the quality of his work and the pace at which he updates his numbers such that he is both accurate and useful for a market participant. So those are the odds as far as I am concerned, but where can we go for an edge over those odds?
Happily, there are multiple online prediction markets and their odds vary wildly. Additionally, for added liquidity, there are private derivative transactions based on the contract price on the main exchanges. There are many but I focused on InTrade, Betfair, Pinnacle, and Paddy Power, four of the main ones by volume.
In mid-September, Obama's odds of winning the Electoral College were approximately 75%. At the same time, InTrade was at 66% and Betfair was at 71%. This offered a relatively scalable bet in which one could bet in favor of Obama on InTrade and against him on Betfair, capturing a substantial riskless spread. Costs were next to nothing based upon how the respective sites set their fees. InTrade charged a flat $5 fee, a de minimis amount of a scaled up bet. Betfair charged 2% of winnings. However, the Betfair side was the losing bet, so the cost was ultimately zero. Also, one could choose to undersize that hedge to size it at 25% of the InTrade bet, proportionate to the insurance needs in the 25% chance of the Obama loss. Spreads widened and tightened throughout the election, but averaged over 8% between various market pairs.
By late October, Obama's odds of winning were at 71%, InTrade was at 55% and the average of European markets was 65%. This was one of the better opportunities presented; in fact, it was such a wide spread that it looked as if American Romney supporters might have been betting in favor of their candidate to strategically improve his appearances on InTrade, instead to betting with the correct odds. But for whatever reason, the US-legal market was the outlier, far from the odds and from the other markets.
This was worth taking advantage of: the available return on investment was over 18% and risk was essentially zero. This opportunity arose through a combination of biased wagers and cross-border regulatory issues. However, while this is a worthy personal account activity, it is not scalable enough to have any professional merit.
In conclusion, trying to exploit mispricing is easier and better than trying to predict the future. Constrained counterparties, such as those who are under specific regulatory burdens, offer some of the best potential mispricing. When parties cannot act freely to place and scale their investments, the pricing system fails. And such pricing failures are just what we look for.
If you would like to better understand American politics, read The Almanac of American Politics 2012. If you want to learn more from the great Nate Silver, I would heartily recommend reading The Signal and the Noise: Why So Many Predictions Fail - but Some Don't.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
Additional disclosure: Chris DeMuth Jr is a portfolio manager at Rangeley Capital, a partnership that invests with a margin of safety by buying securities at deep discounts to their intrinsic value and unlocking that value through corporate events. In order to maximize total returns for our partners, we reserve the right to make investment decisions regarding any security without further notification except where such notification is required by law.