Yves here. Rajiv Sethi focuses on the likelihood that PredictIt will have to wind down operations, and the messiness and gaming opportunities that will result from having to close out contracts that expire after PredictIt ends operations. He mentions only in passing that this event would also shutter the last significant on-line election prediction market. There’s a lot of enthusiasm for prediction markets as soothsayers, but that may be as much due to the disastrous state of polls as the supposed virtues of prediction markets (pollsters have a great deal of difficulty getting anyone to answer questions, and their calling lists are skewed towards landlines, as in old farts).
Prediction markets work best when they represent the electorate well. The reason the Brexit betting markets were so wrong (they predicted a Remain win by 6 points the night before the vote) is that in gambling, the “votes” are weighted by