Prediction markets and North Korea
Prediction markets do not uncover intelligence on North Korea—they price America’s misinformed expectations.
With the Trump Administration, oil markets, and Iran, Polymarket is in the news near every day now. Prediction markets, like Polymarket, turn questions into tradable contracts. Traders buy and sell based on their beliefs, and the resulting price reflects a probability. The mechanism depends on incentives: those with better information or better judgment profit, and their trades move the market.
Their appeal is simple. In domains like elections, prediction markets often outperform polls and experts. They reward accuracy, aggregate diverse inputs, and operate continuously—and in certain circumstances can bring out fragments of intelligence that would otherwise remain scattered or unarticulated. From an intelligence point of view, they collect information from what are often inaccessible sources, and they reward and crowd-source analysis (meaning analysis of share trading is also valuable). For the individual, in an era of skepticism toward institutions, they also offer a wee skerrick of decentralized truth.
In theory, this produces a real-time synthesis of distributed knowledge. Applied to North Korea—the archetype of opacity—it seems almost ideal. Unfortunately, this intuition is flawed. The conditions that make North Korea analytically difficult also make it resistant to market-based intelligence.
The problem is not that prediction markets are flawed in general. It is that North Korea violates the assumptions that make them work.
First, there is no meaningful insider participation. Prediction markets derive much of their power from participants who possess privileged or semi-privileged knowledge. In open systems, this includes bureaucrats, contractors, journalists, and industry actors. North Korea offers none of this. The regime is sealed, information is tightly compartmentalized, and even elites operate within narrow informational silos.
More importantly, those with genuine proximity to decision-making cannot participate in global financial platforms (although, it is conceivable that this opens opportunities for coordinated insider trading as a revenue source for North Korea’s revenue hungry entrepreneurs?). The result is a market often composed almost entirely of external observers - often largely American-based. It aggregates misinterpretation—not access. This is intelligence analysis crowd sourcing at best.
Intelligence crowd sourcing can be useful, but only when interpretation is tracked, tested, and refined over time—when analysts build a record, good judgment is identifiable, and reliable signals can be separated from noise through repetition and feedback. In the North Korean case, this process breaks down. The signals are too sparse, outcomes too infrequent or ambiguous, and feedback too weak to clearly distinguish good analysis from lucky guesses.
Second, the informational baseline is too weak to sustain convergence.
Markets work by processing signals. North Korea produces very few, and those it does produce are often ambiguous. Satellite imagery, state media releases, and indirect economic indicators provide only partial and highly contestable insights.
In such an environment, price does not converge toward accuracy; it drifts. Small pieces of information—often no more than anomalies—take on exaggerated significance. A missed public appearance by Kim Jong-un becomes a proxy for regime instability. Activity near the Yongbyon Nuclear Scientific Research Center is read as strategic intent. With no dense stream of data to discipline interpretation, the market cannot stabilize around a reliable estimate.
Third, narrative replaces information as the dominant driver. In high-information environments, markets can correct for bias because new data constantly challenges prevailing views. In the North Korean case, data is sparse, so narratives fill the void. Once a dominant storyline emerges—imminent escalation, leadership instability, internal crisis—it begins to drive trading behavior. Prices move not because new information has arrived, but because participants align with or react to the narrative. This creates reflexivity: the market reinforces the story, and the story reinforces the market. Rather than aggregating independent signals, the system amplifies shared assumptions.
Fourth, thin markets are structurally vulnerable to distortion. Prediction markets on North Korea are unlikely to attract deep liquidity. This matters. In thin markets, relatively small trades can move prices significantly. This introduces noise and opens the door to manipulation. A trader—or coordinated group—does not need vast resources to shift probabilities in a visible way. In a strategic context where perception itself can matter, this is not trivial. Whether or not state actors actively intervene, the structural vulnerability remains: price can be moved without corresponding information.
Fifth, markets struggle with the distinction between capability and intent. Even well-functioning prediction markets are better at forecasting observable events than interpreting underlying motivations. North Korea magnifies this gap. A market can price the likelihood of a missile launch, but it cannot see the internal deliberations that produce it. The result is a familiar slippage: patterns of capability are read as signals of intent.
Over time, this creates a deeper distortion. Trading begins to reflect not North Korean decision-making, but expectations about how the United States and its allies will interpret North Korea. Prices move on anticipated reactions—what Washington thinks Pyongyang might do—rather than on direct insight into Pyongyang itself. The market, in effect, becomes a mirror of external strategic imagination, not a window into the regime.
Sixth, opacity creates asymmetry, not just uncertainty. In most markets, uncertainty is broadly shared. In the North Korean case, uncertainty is asymmetric. Some actors—state intelligence agencies, for instance—may possess significantly better information than the general trading population, but they are unlikely to participate directly. Their knowledge does not enter the market efficiently. Instead, the market is dominated by second-order interpretations of what better-informed actors might know. This produces a recursive structure of inference, rather than a direct aggregation of primary information.
Finally, the absence of feedback loops limits learning. Prediction markets improve over time when outcomes provide clear feedback. In North Korea, many of the most important questions—about regime stability, internal power dynamics, or long-term strategy—do not resolve cleanly or quickly. Even when events occur, their meaning is often contested. This weakens the market’s ability to learn and recalibrate. Errors are not easily identified, and correct predictions may be indistinguishable from lucky guesses.
None of this means money cannot be made. If anything, North Korea markets can be profitable precisely because they are so prone to misinterpretation—especially the kind that circulates through Western media. Coverage routinely inflates weak or ambiguous signals into coherent narratives, and those narratives drive pricing.
The opportunity, then, is not in superior analysis of North Korea itself, but in recognizing when the market has absorbed and amplified these distortions. Profit comes from trading against consensus when it is built on thin, exaggerated, or poorly grounded interpretations.
In this sense, Polymarket does reward “better analysis”. But crucially, this is not insider intelligence in any meaningful sense. It’s an arbitrage on misinterpretation. The profits come not from knowing what North Korea will do, but from recognizing when the market is overconfident in what it thinks it knows.
Prediction markets function best where information is dispersed but available, where insiders can participate, and where outcomes provide clear feedback. North Korea satisfies none of these conditions.
In this context, prediction markets do not uncover intelligence on North Korea—they price America’s misinformed expectations.