Prediction markets are not gambling but misunderstood as the "truth machine"

Author: Jeff Park Source: X, @dgt10011 Translation: Shano Opa, Golden Finance

Last week, two media outlets, Axios and More Perfect US (MPU), respectively introduced the public to what prediction markets are. Axios’s Dan Primack tried to build a neutral platform and got into a fierce debate with the founder of Kalshi (though its position was already fairly clear); meanwhile, Trevor Hayes of MPU was more direct, turning prediction markets into a kind of social ill.

To be honest, I partly agree with both sides’ viewpoints. As someone who has worked for years at the intersection of Wall Street and cryptocurrency, I understand people’s growing unease about the increasing financialization—this trend is creating a social atmosphere like a “gambling-style public health crisis.” But the mistake these media people often make is that they first draw a conclusion, then work backward to ask “who is fueling this phenomenon,” and ultimately distort and confuse multiple distinct problems with an overly simplified narrative. One moment they’re talking about “insider trading,” the next they’re talking about “online casinos,” and in the end, everything is attributed to “gambling addiction.”

And that is the core misunderstanding most people have about prediction markets: regardless of how you view the downsides of excessive financialization (whether it’s zero-to-expiration options, swap-based ETFs, or internet-celebrity concept stocks), prediction markets should be praised as tools for enhancing individual autonomy, uncovering the truth, and practicing decentralized moral rights.

In what follows, I will break down this issue more rationally.

The Blurred Boundary Between “Investment” and “Gambling”

The only standard for distinguishing “investment” from “gambling” is whether the behavior has a positive expected value (+EV), not whether the system itself is deterministic or random. In other words, what defines the boundary is the participant—not the game itself.

Let’s go into more detail. In MPU’s coverage, Trevor Hayes’s questions always begin with “Obviously, prediction markets are gambling…” as if that were a self-evident conclusion. Yet this fundamental premise is precisely what needs to be examined first.

Over the past twenty years, the biggest trend in finance has been the increasingly blurred line between “investment” and “gambling”:

1. 60% of trading volume in the U.S. stock market comes from high-frequency trading, monopolized by oligopolies such as Citadel and Castle Investment;

2. Passive ETFs account for more than 90% of the total assets under management of ETFs (even though active strategies have finally begun to make a slow recovery);

3. The average holding period for U.S. stocks has dropped from about 9 years in the mid-1970s to only about 6 months in 2025.

Meanwhile, over the past decade, daily trading volume in U.S. stocks has more than tripled, driven as well by algorithmic trading. An even more noteworthy trend is that in 2025, the trading volume of retail investors has already exceeded $5 trillion, up roughly 50% from 2023.

But you rarely see commentators accuse “stock trading is gambling.” Why? Because most people take it for granted that stock picking is not gambling—it requires skill. That’s the key insight: labeling something as “gambling” is unfair, because it mixes up skill-based games with pure probability-based games.

For example: slot machines and poker are both called gambling, but they are fundamentally different—slot machines are purely negative-expected-value behavior driven by luck, while poker is an activity that achieves positive returns through real skill.

To put it plainly, the definition of “investment” versus “gambling” basically depends on whether people believe the strategy has positive returns, and has nothing to do with whether the “game” itself is deterministic (riskless arbitrage, slot machines) or stochastic (stock picking, poker).

Like poker, prediction markets are stochastic games with a deterministic component. Whether you see them as “gambling” or “investment” depends entirely on the participant—whether you are someone with strong autonomous decision-making and high skill, or the opposite, or somewhere in between.

This leads to the second question: if we treat gambling as “speculation” driven by participants, then how do these markets actually operate? And where does liquidity come from?

The other side of speculation is insurance

When all financial innovations are first created, they look like gambling. Early stock market insider trading was rampant; the futures market (Eurodollar was the first tool used by government officials for political insider trading); and the modern large commodities market (traditionally, it was almost impossible to define insider trading in the usual sense)—all followed this pattern.

The reason is simple: the other side of speculation is insurance. The two are two sides of the same coin; the essence of these zero-sum games is standardized risk transfer. And not all “information” is naturally generated by private institutions.

Critics of prediction markets often ask: “Some markets are purely speculative and can’t create social value, so they shouldn’t exist in the first place.”

The most typical example of this view is sports betting. Many people think sports are entertainment, and betting for entertainment is inherently unproductive.

But this understanding is wrong. Entertainment is social consumption, and arguably one of the core sources of human happiness in life. More importantly, entertainment itself is economic consumption and has the nature of a two-sided market. The global sports industry generates annual revenue of over $50 billion, and when you include surrounding ecosystems such as media, equipment, apparel, and sports nutrition, the estimated scale exceeds $1 trillion. For example, Nike pays huge sponsorship fees to athletes and teams; its allocation of capital and risk hedging are naturally closely tied to game outcomes and players’ conditions.

Today, society often equates sports betting with “a casino” simply because the federal level has not allowed legitimate markets to exist. This completely ignores its potential value that has not yet been uncovered.

The value of derivatives lies in risk transfer, which is the core principle behind all insurance models and asset securitization. The existence of an insurance market necessarily requires speculators as counterparties; in transparent, open markets, absent government intervention, this is the only feasible model. In fact, most failures of insurance systems stem from government intervention distorting the true market pricing of risk carried by risk bearers. Insurance and asset securitization remain among the greatest financial innovations for improving capital efficiency.

But the controversy remains: at what point does the development of an event turn from normal financial services into a social ill? And how should we establish “event classification standards”? This is the last point I want to discuss.

Two key features that distinguish prediction markets from other derivatives

Prediction markets are different from other derivatives, mainly in two aspects:

1. Precise outcomes

2. Clear expiration timelines

A brief recap of the basic market-making logic: in most financial markets, the central limit order book (CLOB) is used to measure and provide liquidity because assets typically have perpetual value. But prediction markets are different: once the outcome of an event is realized, liquidity instantly drops to zero, with no remaining bids or asks. This is extremely unfriendly to liquidity providers—the binary 0 or 1 payoff result makes the assumption of continuous dynamic hedging completely invalid.

More importantly, prediction markets are probability-based markets, not price-based markets. Contracts with odds around 50% have far higher liquidity than contracts with 98% probability, because for the latter, every 1 percentage point change leads to an exponential increase in payout costs. In other words, liquidity can’t be sustained by bid-ask spreads alone. This is something fixed-income derivatives traders understand far better than stock traders (for example, when interest rates move between 4% and 0.5%, a 10 basis point fluctuation means something completely different).

This means that in markets with severe information asymmetry and outcomes that can be precisely predicted, professional market makers are unlikely to provide large amounts of liquidity. It also explains why the assumption that “insiders massively profit by using information” only yields very small gains in the vast majority of scenarios. In the end, the market reflects what people truly care about. Even if I know for sure whether “Jeff Park’s next podcast will feature him wearing a Bitwise sweater,” the liquidity of that market is basically zero.

Most arguments against insider trading assume insiders can extract huge profits, but that’s not the case. A worthless market will have no natural liquidity; liquidity itself precisely prices the true value of information. From that, event classification standards will naturally form.

Why prediction markets are worth far more than the potential risks

As mentioned earlier, one major advantage of prediction markets is their precision, which is also their most valuable highlight.

In today’s overly financialized environment, asset prices are determined more by technical factors and capital flows than by fundamental analysis and real intrinsic value. Prediction markets, uniquely, bring the purest form of “basis risk” back to the facts themselves.

In the future, if you think Tesla’s revenue will exceed expectations, rather than buying stocks whose price is interfered with by external factors, you could bet in a prediction market. If you have a unique view on non-farm payroll data, you don’t need to trade Eurodollar or mini S&P futures—you can simply bet on the data itself. In short, precision will truly reward excess returns, deep research, and real ability.

Many people criticize prediction markets as “harvesting the financial knowledge of those who are not well-informed,” assuming that “gamblers” will inevitably lose money, and therefore that it is a social ill. But in fact, prediction markets have the fairest mechanism, able to provide positive expected returns for investors with independent judgment. More importantly, prediction markets have no “house”—unlike Las Vegas casinos that eject high-yield players, prediction markets welcome truly capable people.

Castle Securities and Charles Schwab have both announced plans to enter prediction markets. Are they “extracting value from economically vulnerable groups”? Obviously not. They understand better than most: the other side of speculation is insurance, and your risk convexity is precisely my profit convexity.

Why do authoritative media fear the truth markets?

One last addition. After reading the above, you will at least agree on this: as long as regulation is appropriate, prediction markets have enormous value. We can fully address “gambling problems” and “social ills” on the premise that benefits exceed costs. But there is one key question we have overlooked: “What if insider trading occurs in markets involving major public interests? Will it become a tool for private profit?”

This question is very complex, and I will answer it in detail in another article.

Here, I’d like to share a book I recently read—Ashley Linsberg’s The Gray Lady Winked. The book documents decades of systemic failures at The New York Times that were never accidental: concealing the Great Famine during Stalin’s era, producing abnormal reporting on Castro’s rise, hyping Iraq’s weapons of mass destruction, downplaying the dangers of Hitler’s rise… This authoritative media outlet has consistently relied on its information channels, its ideology, and its instinct for institutional self-preservation to hide the truth, manufacture consensus, and launder its own mistakes afterward.

This book helps us re-understand “media bias”: it’s not simply a debate between left and right, but a structural issue where authoritative institutions manufacture consensus and whitewash errors.

Returning to the beginning: Axios and MPU are also not neutral parties in this debate. In the future, you will see more media outlets attacking prediction markets—and the reasons they oppose it are precisely the reasons you should support it.

Information is valuable—there’s no doubt about that. I also often say: the opposite of false information isn’t necessarily the truth; it’s information controlled by the state.

The real controversy lies in: who has the right to price information? Who can profit from it? And does all of this happen before the public knows?

When insiders hoard asymmetrical information, the motive of money is far less important than power struggles. By exploiting the public’s ignorance, information becomes weaponized: it manipulates public opinion, spreads false information, and ultimately allows the truth markets to be controlled. Therefore, the core issue in opposing insider trading is not economic efficiency, but the right to access information—some people trade based on the information they know, while most people can only trade based on the information they are allowed to know.

Once you understand this, you won’t feel pessimistic about prediction markets; instead, you’ll see the world more clearly and more precisely. And this is why I’ve always believed: supporting prediction markets is one of the most democratic value systems.