Political prediction markets have been around for decades, but rose to unprecedented visibility and prominence during the last election cycle in the United States.
I generally agree with your post, but (as I emailed you on February 21), my concern about the behavior of the prediction market prices in this particular contest go further than models of the polls. Cuomo was dominating the polls in part due to name recognition, so a market should have expected that he would go downwards in the polls while someone would go upwards. And, combined with rank choice voting, this made a polling lead less valuable than in a plurality vote situation. Thus, I found 70% in the market for mayor 4 months out, shockingly high, because of what the markets should price in about polling movement and voting type, that polling at any given time cannot express.
Sent on February 21 @ 10:28 AM ET (unedited email to Rajiv): "It is entirely possible that Andrew Cuomo could win the Democratic nomination for mayor of NYC, but due to (1) fact he left in scandal (2) is seen as centrist pro-Trump Democratic with a Democratic base very angry at centrist pro-Trump Democrats (3) rank-choice voting, this outcome is highly improbable. It is trading at 70% on both Kalshi and PolyMarket. My guess is that the markets are dominated by right-wing tech-bros who get their news from Twitter and Fox. But, still why are people not buying the other side in Kalshi? Also, there are just a ton of poorly defined contracts on both sites that are going to end up in disaster. In short, I really want markets to succeed, but I worry that Kaslshi and PolyMarket are the two worst to have the reins and it is all going to go terribly wrong."
I generally agree with your post, but (as I emailed you on February 21), my concern about the behavior of the prediction market prices in this particular contest go further than models of the polls. Cuomo was dominating the polls in part due to name recognition, so a market should have expected that he would go downwards in the polls while someone would go upwards. And, combined with rank choice voting, this made a polling lead less valuable than in a plurality vote situation. Thus, I found 70% in the market for mayor 4 months out, shockingly high, because of what the markets should price in about polling movement and voting type, that polling at any given time cannot express.
Yes, I remember your email! Very prescient. Let me know if you want me to post your original message? Or you could post it?
Sent on February 21 @ 10:28 AM ET (unedited email to Rajiv): "It is entirely possible that Andrew Cuomo could win the Democratic nomination for mayor of NYC, but due to (1) fact he left in scandal (2) is seen as centrist pro-Trump Democratic with a Democratic base very angry at centrist pro-Trump Democrats (3) rank-choice voting, this outcome is highly improbable. It is trading at 70% on both Kalshi and PolyMarket. My guess is that the markets are dominated by right-wing tech-bros who get their news from Twitter and Fox. But, still why are people not buying the other side in Kalshi? Also, there are just a ton of poorly defined contracts on both sites that are going to end up in disaster. In short, I really want markets to succeed, but I worry that Kaslshi and PolyMarket are the two worst to have the reins and it is all going to go terribly wrong."