Most people's misconceptions about prediction markets: it's not excessive financialization, but subjectivity and truth discovery.

Source: Jeff Park, Bitwise Advisor; Translation: Golden Finance Claw

Last week, both Axios and More Perfect Union (MPU) stepped forward to explain to the public what prediction markets are. Although Axios’s Dan Primack tried to offer a neutral platform for a debate with Kalshi’s founder (even if his bias was fairly transparent), MPU’s Trevor Hayes took a more direct stance, portraying prediction markets as a “social cancer.”

Honestly, I sympathize with parts of both viewpoints. As someone whose career sits at the intersection of Wall Street and crypto, I understand society’s growing concern about “over-financialization,” a trend that is fueling a culture of “public health crises caused by gambling.” But at the same time, a common error these journalists make is: they pre-assume the conclusion, then search backward for “culprits,” often lumping multiple issues together in an overly simplified narrative. One moment we’re talking about “insider trading,” and the next it’s “online casinos,” and finally it all boils down to “gambling addiction.”

That, however, is exactly the misunderstanding most people have about prediction markets: no matter how you view the downsides of over-financialization (through 0DTE options, swap-based ETFs, Meme stocks, and more), the story of prediction markets should be celebrated as a driver of high agency (High Agency), truth discovery (Truth Discovery), and decentralized moral rights.

The article below is my attempt to break down this viewpoint in greater depth.

The Blurred Line Between “Investment” and “Gambling”

Whether something is “investment” or “gambling” depends entirely on whether you believe the behavior has a “positive expected value” (+EV), not on whether the system is deterministic or random. In other words, it’s defined by the players—not by the game.

Let’s unpack this. The first thing I noticed in MPU’s coverage is that Trevor Hayes often opens questions with “Since prediction markets are obviously gambling…” as if that were a settled fact. That foundational assumption needs to be challenged first.

The biggest trend in finance over the past twenty years is that the clear boundary between “investment” and “gambling” has become increasingly blurred. Consider the following facts: 1) 60% of U.S. stock trading volume is high-frequency trading (HFT), monopolized by oligopolies such as Jane Street and Citadel; 2) passive ETFs account for more than 90% of assets under management in ETFs (even though active strategies are starting to mount a belated rebound); 3) the average holding period for U.S. stocks has shrunk from 9 years in the mid-1970s to only about 6 months in 2025! At the same time, driven by algorithmic trading, average daily trading volume has more than doubled over the past decade. On top of all this data, there is an unstoppable trend: retail investors’ trading activity in 2025 exceeded $5 trillion, up about 50% from 2023.

Yet you won’t find many experts coming out to blame “stock trading” as gambling. Why? Because most people agree that stock picking isn’t gambling—presumably because it requires skill. This is a key insight: the reason the term “gambling” has become descriptively unfair is that it conflates “technical skill games” with “pure probability games.” For example, both slot machines and poker are called gambling, but many people can intuitively recognize the unfairness—slot machines are purely luck-based strategies with negative expected value (-EV), while poker can be a skill-based strategy with positive expected value (+EV).

Put plainly, whether something is “investment” or “gambling” mainly comes down to whether a person believes the strategy allows for positive expected returns. It has nothing to do with the game itself—whether it’s deterministic (pure risk-value arbitrage and slot machines are) or random (stock picking and poker are).

Like poker, prediction markets are a stochastic game with deterministic components. Whether you see them as “gambling” or “investment” depends entirely on the player—that is, on you. It depends on whether you’re a high-agency, high-skill participant or a low-agency, low-skill participant. This brings us to a second question: if we think of gambling as “player-driven speculation,” how do such markets actually operate? And who provides liquidity?

“Speculation’s Other Side Is Insurance”

All financial innovations initially look like gambling. The earliest stock markets were like that (teeming with frenzied insider trading), so were the futures markets (European dollars being the earliest political “insider trading” tool used by government officials), and of course modern commodity markets are no different (where classic insider trading is nearly impossible to define). Strictly speaking, this is because the other side of speculation is insurance. They’re two sides of the same coin, a zero-sum game defined by strict synthetic risk transfer. And not all “information” naturally originates in private enterprises.

This leads to the next question prediction market critics often raise: “Some markets are, in function, purely speculative; since they create no value for society, they shouldn’t exist.” The most common target is sports betting. Because sports are entertainment, betting on entertainment is seen as fundamentally unproductive.

But this view is wrong. Entertainment is social consumption. Some would even argue that entertainment is a fundamental reason humans find life fulfilling. More importantly, entertainment itself is economic consumption, which means it has a two-sided market. The sports industry produces more than $50 billion in revenue; if you add the surrounding ecosystem (media, equipment, apparel, nutrition products, and so on), the estimated figure reaches over $1 trillion. Take Nike as an example: it pays millions of dollars in sponsorship fees to players and teams, who have real economic interests in how capital is allocated (and how risks are hedged)—all based on the outcomes of sporting events and the performance of players. Today, society has been widely brainwashed into thinking sports betting is purely “casino” behavior simply because legitimate federal markets didn’t exist before; this entirely misses the kind of possibility that prediction markets can offer beyond imagination.

Derivatives are useful because they allow risk transfer. This is the foundational principle behind all insurance models (and securitization). To have insurance, you need a counterparty—a speculator on the other side; in transparent, open markets without government interference, there’s no other way. In fact, insurance most often fails when government intervention distorts real market prices. Insurance and securitization remain among the greatest financial innovations for releasing capital efficiency.

However, the problem of “events” still remains: in what circumstances does an event truly become a social cancer rather than a naturally useful financial service? How do we develop an “event taxonomy”? This leads to my final point.

The Difference Between Prediction Markets and Other Derivatives

“The difference between prediction markets and other derivatives lies in two features: 1) they are precise (Precise), and 2) they have limited expiry (Expiry).”

To understand what that means, let’s go back to the “Market Maker 101” course. In most financial markets, the role of the central limit order book (CLOB) is to measure and provide liquidity, because assets often have perpetual value. But prediction markets are different: once an event catalyst occurs, liquidity collapses to zero, leaving no buyers or sellers on the other side. This creates an enormous challenge for liquidity providers, because the binary outcome of 0 or 1 destroys the assumption of continuous dynamic hedging.

More importantly, prediction markets are markets based on “odds,” not “prices.” That means liquidity around 2 points near the midpoint (50%) will be far higher than liquidity around 2 points near 98%, because the odds payout in the latter case is exponentially heavier per point. In other words, liquidity cannot be sustained by bid-ask spreads alone. Fixed-income derivatives traders understand this very well—for example, a 10 basis point move when rates are at 4% is completely different from a 10 basis point move when rates are at 0.5%.

All of this implies that in markets with extreme information asymmetry and outcomes that can be precisely predicted, professional market makers are unlikely to provide large amounts of liquidity. It also means that most assumptions about insiders “cashing out” with insider information ultimately involve very small amounts. The market ultimately decides what people care about. Yes, I possess secret information about whether Jeff Park’s next recording will involve a Bitwise sweater, but the chance for liquidity in that market is negligible. Most “anti-insider trading” viewpoints assume insiders make a lot of money—but that isn’t true in most markets. In short, irrelevant markets don’t generate natural liquidity. In fact, I’d go so far as to bet that liquidity itself will be priced precisely according to the value of that information. That’s how an “event taxonomy” develops organically.

So why are prediction markets useful enough that their benefits outweigh potential costs?

I mentioned earlier that they’re precise. This is one of the most virtuous aspects prediction markets must emphasize. In a world where over-financialization causes asset prices to be driven more by technicals and capital flows than by fundamental analysis, prediction markets uniquely restore the cleanest basis risk for “truth”. In the future, if you believe you have fundamental alpha on whether Tesla’s revenue will beat expectations, you should consider placing a bet in a prediction market rather than buying the stock—because the stock price may be influenced by other external factors and behave abnormally. If you believe you have an edge on non-farm payroll data, you should bet on that data rather than trading European dollars or E-mini futures. In other words, greater precision rewards genuine excess returns, real research, and real skill.

Many people think prediction markets “plunder financial illiteracy” by assuming that “gamblers” lose money, making it a social vice. But in fact, prediction markets have the fairest mechanism for rewarding the real skills of high-agency investors. Even more importantly, prediction markets have no “house.” Unlike Las Vegas casinos that kick out positive-EV players, prediction markets welcome your participation.

Citadel Securities and Charles Schwab have announced that they’re exploring entry into prediction markets. Are they “exploiting economically vulnerable groups”? I’m highly skeptical. They simply understand better than most that “the other side of speculation is insurance”—meaning your convexity (passively bearing risk) is my concavity (actively hedging risk).

Why “Gray Ladies” Fear Truth Markets

This leads to my final note. If you’ve read what’s above, you may at least be starting to appreciate the power of prediction markets under proper regulation. If we believe benefits outweigh costs, we can address “gambling problems” and social ills in a variety of ways. However, there’s one issue you might have noticed we skipped: “What about insider trading in markets tied to major public interests? Isn’t that about privatizing profits?”

That is still a complex question, and I plan to answer it in another article. But I want to leave you with an idea—and a book I recently read, Ashley Rindsberg’s The Gray Lady Winked. It documents decades of media institutions failing not by accident: Duranti’s suppression of Stalin’s Great Famine, Castro’s strange rise in Cuba, the push regarding Iraq’s weapons of mass destruction, and the systematic whitewashing of Hitler’s rise. In all these cases, The New York Times (the “Gray Lady”) has always been there—leveraging access, ideology, and institutional self-preservation to muddy the public’s demand for truth.

If you’ve read the book, you’ll understand how it reframes “media bias” from left/right arguments into a more interesting structural problem: how prestige institutions manufacture consensus, and then retrospectively launder their mistakes. In fact, it brings us back to where we started: Axios and More Perfect Union are not unbiased actors in this space. For these reasons, you will continue to see many media criticisms of prediction markets. But don’t mistake it: the reason they don’t like it is precisely why you should support it.

Information has a price. There’s no controversy about that. I often say the opposite of misleading information isn’t necessarily the truth; the opposite of misleading information is actually “state-controlled information.”

The focus of this debate is: who has the right to set prices, who profits from them, and whether all of this has already happened before you ever see it. When insiders hoard asymmetrical information, monetary incentives take a back seat to exchanges of power. By taxing others’ ignorance, this information can be weaponized to sway sentiment or spread falsehoods, and even the truth markets themselves can be captured.

So the real reason to oppose insider trading is not so much about economic efficiency as it is about access rights. The fact is, some people trade based on what they know, and the rest of us trade based on what we’re allowed to know.

Once you see this clearly, you won’t be cynical about prediction markets. You’ll just become more precise about the world. That’s why I firmly believe that maintaining optimism about prediction markets is one of the most democratic values a person can hold.