As the contentious U.S. presidential election continues the biggest winner of this week might be prediction markets, which have provided real-time assessments of who might win. Supported by the efficient market hypothesis, prediction markets, in theory, should be the best predictor outcomes based on the wisdom of the crowd. As long as the marginal trade is motivated by economically rational traders, free markets theoretically should be better predictors than professional pollers given they aggregate information from a wide base of independent and diverse individuals rather than relying on just a handful of models. Below we compare centralized to decentralized marketplaces and statistical models to prediction models.
In February 2020, FTX announced a new product, the Trump futures contract to bet on the US elections. For their contract, if you think there's a 72% chance of Trump winning, then it makes sense to buy $TRUMPWIN below $0.72 and sell above $0.72. The futures contract expires at US$1 if Trump wins and $0 otherwise.
Here, FTX may choose to settle the contract according to the unanimous election results of The New York Times, FiveThirtyEight, 270toWin, Fox, and CNN. If it’s impossible to declare a winner by November 5, 2020, $TRUMPWIN will settle to $1 if Donald Trump is still president on February 1, 2021.
The 2020 election was the most bet-on event in history in both the traditional and crypto markets. Augur open interest went from $17,000 at the start of August 2020 to $4 million by the elections. Also, in a month, the trading volume on decentralized betting platform Polymarket went from zero to $3 million.

DeFi prediction market platforms, where users can transact on the blockchain without a broker, benefited from this wave of new interest:
Mira was a Senior Research Analyst at Messari. Prior to joining Messari, Mira was a Senior Portfolio Manager for a US$6 billion Asia Pacific equities fund at APG Asset Management. Mira received a BA in Economics and Mathematical Methods in the Social Sciences from Northwestern University.