Monday 30 August 2010

When's an arbitrage not an arbitrage?

I'm beginning to hate Australia.

To recap: I've been playing the price spread between BetFair, iPredict and CentreBet on the Australian election.

If Labor wins, I'll be up $191, with lots of long positions at BetFair and CentreBet, where Labor has been cheap relative to iPredict. If the Coalition wins, I'll be up $280 at iPredict, where the Liberals have been cheap relative to the other two markets. I win in both states of the world.

But the states of the world now don't span the space. The third alternative is no result and a second election. In that case, BetFair unwinds its bets and gives you your money back, iPredict holds positions pending the results of the second election, and CentreBet is still deciding what to do.

If one of the Australian readers would kindly go and box the independents about the ears and make them get on with it already, I'd be much obliged.

BetFair has a very very thin book on whether there'll be a second election. I need a market in "There'll be a second election and Labor will win". Working out and implementing the right hedge here is going to be tricky.


  1. I don't want sound unsympathetic or anything to your money making efforts, but, um, arbitrage is supposed to be hard. You're earning a normal return to an investment of time, effort and perhaps some angst.

  2. Oh, I'm earning nowhere near a normal rate of return if you count time. I'm mostly interested to see how much of a pain it actually is to run the thing.