Coverage of prediction platform Polymarket has recently converged on a single statistic, delivered with the cadence of a verdict: Most users lose money. The top 1% of accounts capture roughly three-quarters of the gains, while most traders since 2022 are underwater.
These numbers are accurate — and meaningless. They describe a general property of retail derivatives venues and betting markets. Some 1.3% of daily fantasy sports players took 91% of the profits in 2015. Of Brazilian retail futures traders who persisted past 300 days, 97% lost money. Retail equity options traders lose 5% to 9% per earnings-announcement trade. A Washington Post analysis concluded that Polymarket users lost at roughly the same rate as British online sports bettors.
Every futures market was attacked on its founding by people who assumed gambling was evil, winners were cheats, and losers were dupes. The Chicago Board of Trade spent its first half-century defending grain futures against state laws and federal bills that treated them as illegal gambling. The grain options market on futures was effectively banned for decades after the 1936 Commodity Exchange Act on the theory that farmers were being fleeced by speculators. The introduction of stock index futures in the 1980s produced editorials indistinguishable from what is now being written about Polymarket.
The critics counted losers in each case, declared the practice antisocial, and were eventually overruled when the markets in question turned out to do something useful — pool dispersed information into a public price signal that other people could use for free.
Economist Fischer Black explained this in his 1986 American Finance Association presidential address. Markets need participants who trade on what they think is information but isn’t, because two people with the same information have no reason to transact. These “noise” traders allow a market to exist and produce a price.
Information traders profit from them, but not so reliably that the noise traders disappear, because there is always a fresh supply of people who believe their own forecasts. The equilibrium is uncomfortable to look at if you start from the premise that everyone should win, and unremarkable if you start from the premise that there are two sides to every trade.
A paper from London Business School and Yale University researchers shows the architecture explicitly: Roughly 3% of Polymarket accounts generate the bulk of price discovery, and the remaining 97% funds that accuracy, just as Black described.
The public probability estimates the Wall Street Journal now embeds alongside its political coverage exist because a large group of less-informed traders is willing to pay for the privilege of discovering whether their own models of the world are any good.
The loss-distribution framing misses this last part entirely. The retail participant who funds a Polymarket account and loses $300 over a year of trading on elections, geopolitics and Federal Reserve decisions is purchasing a piece of information about herself that she cannot obtain in any other way: Whether her judgment, the thing she has been operating on her whole life, is better or worse than the wisdom of crowds. For almost everyone, the answer turns out to be worse, and the rational response is to listen to the crowd. For a small number of people, the answer turns out to be better, and the rational response is to keep going. Either lesson is immensely valuable.
This discovery is not transferable. It cannot be obtained by reading about other people’s Polymarket losses. That some small fraction of participants discover they have a real edge — and unlock a lifetime of being able to trust their own judgment in venues with far more capacity and social importance than Polymarket — cannot exist without the larger population of people discovering otherwise. You must open a lot of oysters to find a pearl.
Other participants are there for other reasons. Some are hedging, like an Airbnb host whose nightly rates depend on her city’s team making the playoffs buying a contract that pays off if they don’t. Many more are paying for entertainment in the same way one pays for a baseball ticket or a video game. The dollars lost in pursuit of an evening’s engagement with the news cycle are not obviously worse spent than the dollars lost at a movie theater, and the activity produces a useful public signal as a byproduct.
The implicit standard behind the recent coverage — that a market is suspect unless most participants win — is one nobody applies anywhere else. We do not declare marathons rigged because most entrants do not finish in the money, or universities fraudulent because most applicants are rejected. The premise that every participant must succeed for the activity to be legitimate is the economic policy equivalent of giving every runner a medal and every student an A.
There are real questions about Polymarket and rival Kalshi Inc. The platforms’ dispute-resolution mechanism is in the hands of a small group of token holders or employees; insider trading has produced a federal indictment and a congressional probe; and the state-by-state fight over sports contracts will determine whether the largest growth category survives in its current form. These are worth covering carefully — but independent of the question of how the wins and losses are distributed among ordinary participants.
That question has an answer, and the answer is: The way they have always been distributed in every market like this one. The story is not that retail is losing. The story is what retail is buying with the loss.
A message from Advisor Perspectives and VettaFi: Discover something new! Click here to register for our upcoming webcasts.
Bloomberg News provided this article. For more articles like this please visit
bloomberg.com.
Read more articles by Aaron Brown
