The midterms are over, and Kalshi has plenty to show for it- the platform said that 479 out of 487 predictions by traders on its platforms were correct – and they were correct even prior to any votes being cast, once again demonstrating these platforms’ ability to call future outcomes and highlighting the danger of third-parties abusing the information for their own gain.
Prediction Markets Tend to Get It Right – and Wrong
However, there is an important caveat when talking about prediction market results at face value. The platforms are often wrong. Take Polymarket, for example. The platform had attributed a 70% chance of winning to Aleksandar Vučić, the ousted prime minister of Serbia, who was defeated by the Student List, formed after two years of mass protests.
Vučić ended up losing that race, securing only 39.2% of the vote, far below the 70% chance of winning that was attributed to him, and certainly a huge flip on the statistical expectation based on traders’ insights, as these platforms call them. The race wasn’t even close, in other words.
Both Kalshi and Polymarket gave Francesca Hong a 95% chance to win in the Wisconsin Democratic governor’s primary, and they were utterly wrong. While true that David Crowley won by less than a percentage point in the end, the fact remains that these platforms have made boastful claims about their accuracy that has been proven wrong time and again.
There is ample evidence that prediction markets are inconsistent and behave illogically, not least because they are driven by traders’ own expectations, and those who trade on the platforms are not necessarily representative of every voter out there, for example.
One study that analyzed 2,500 political prediction markets concluded that prediction markets were behaving inconsistently with rational information aggregation, accentuating a possible bias in the trades placed on prediction markets. In other words, platforms see a lot of people who wish a certain outcome to be true.
Prediction Markets’ Collective Wisdom – Down to a Few People
Prediction market platforms argue that they are representative of the “wisdom of the crowds,” basing this on the argument that the many and their opinions of an event would easily outperform individual experts.
However, a closer look at prediction markets suggests that, in fact, only 1% of traders hold three-quarters of all the profits. The fears of insider trading are also important, because sometimes traders would follow a whale – a person who bets big on certain outcomes and bets large amounts.
So, are prediction markets a good weather vane? Probably. Are they the infallible scrying ball well-timed marketing makes them out to be – definitely not.