Should traders sell on the prediction market before the event?

Find out when it’s more profitable for traders on prediction markets to lock in their profits: before the event occurs or after the outcome is known. We analyze risks and liquidity.

Traders on prediction markets who find themselves in profit shortly before a major event, such as an election, face a tough choice: lock in their profit at the current price or wait for the outcome and try to maximize their gains? At first glance, it all comes down to the forecast: how likely is each result, and how well does the contract price reflect that probability? However, in practice, a second aspect is just as important—the market’s liquidity at the moment the trader decides to exit their position. Let’s look at how liquidity changes using the example of the 2026 Wisconsin Democratic primary on Kalshi, and what conclusions can be drawn from this. Liquidity Experiment on Election Night: The Kalshi Market Case The focus of the analysis is the Kalshi contract KXGOVWINOMD-26-DCRO, which pays $1 if David Crowley, the Milwaukee County Executive, wins the Democratic nomination for Governor of Wisconsin. When the market opened, Crowley was considered an underdog, but he ultimately made a sensational comeback and narrowly defeated Francesca Hong. According to Wisconsin Public Radio, the Associated Press called Crowley the winner at 2:34 a.m. Central Time on August 12. The analysis used Kalshi’s order book history from Predexon, accurate to fractions of a cent: 27,848 snapshots from 8:00 p.m. on August 10 (exactly 24 hours before polls closed) to 2:45 a.m. on August 12. To avoid statistical distortion from frequent quote updates, the analysis used the last order book snapshot of each minute. The key liquidity metrics were the bid-ask spread and the order book depth for “YES” buy orders within 1 and 5 cents of the best price. These figures show how much a trader could sell without moving the price too far from the current level. Market Dynamics: Price Barely Changed, but Liquidity Vanished Twenty-four hours before polls closed, the contract on Crowley’s win was trading at about 5.05 cents. The spread was just 0.10 cents—meaning the difference between the best bid and ask was minimal. There were 29,854 contracts to buy within 1 cent of the best price, and 338,477 contracts within 5 cents. By the time polls closed, the average price had dropped to 4.50 cents, so the market’s assessment of Crowley’s chances barely changed. But liquidity plummeted: the spread widened tenfold to 1 cent, and “YES” depth within 1 cent shrank by 93% to 2,200 contracts. Within 5 cents, depth fell by 83% to 56,555 contracts. During vote counting, the situation remained tense. The median depth for orders within 1 cent was 8,667 contracts—43% less than a day earlier. In the 5-cent range, the median dropped from 312,693 to 20,118 contracts, a 94% decrease. The most critical moment came around 8:34 p.m. Central Time, when Crowley’s price hit 55 cents. There were only 102 “YES” buy contracts within 1 cent of the best price—and the same within 5 cents. The spread widened to 6 cents. A trader who waited for this dramatic price jump could find themselves forced to sell at a poor price or leave a limit order and hope it would be filled as the market shifted. After the Associated Press called Crowley’s victory, the average price reached 99.55 cents, the spread narrowed again to 0.10 cents, and “YES” depth within 1 cent exceeded 272,000 contracts. But this was already “settlement-stage liquidity”—when the outcome is almost certain, and traders no longer need flexibility to manage event risk. Why Liquidity Disappears When New Information Arrives This market behavior matches classic microeconomic models of financial markets. According to the model by Lawrence Glosten and Paul Milgrom, the bid-ask spread compensates market makers for the risk of trading with better-informed participants. When the chance of facing someone with insider information rises, market makers widen the spread or reduce order sizes. (Glosten & Milgrom, 1985) In Albert Kyle’s model, market depth depends on the share of uninformed trades and the informational advantage in the order flow. When information arrives quickly, even a small trade can move the price significantly. (Kyle, 1985) Election prediction markets illustrate these effects especially clearly. As soon as vote counting begins, some traders gain an edge: faster data sources, local insights, automated analysis systems. For them, a limit order becomes a free option: they can execute it when the price is outdated and ignore it otherwise. In response, less-informed participants either pull their orders, reduce size, or demand a wider spread. This explains why “headline” liquidity can be misleading. For example, at poll closing, Crowley’s order book showed over 1.1 million contracts, but in reality, only 2,200 could be sold near the market price. Large orders at prices near 0 or 100 cents create an illusion of depth, but actual liquidity for exiting at a fair price is minimal. Arguments for Selling Before the Event A practical rule: you should only wait if the expected benefit from additional information outweighs the expected costs—wider spreads, slippage, and the risk your order won’t be filled. Selling before the event can make sense even if you believe the contract has more upside. In this case, you’re not just giving up some profit—you’re buying yourself a guarantee that your order will be executed. Your Edge Has Already Played Out Often, a position is opened based on the market underestimating certain factors: polls, support, fundraising, or early voting. If the market has already moved in your favor, holding the position further turns a successful trade into a new bet—now on the event’s outcome itself, with a less favorable risk/reward and in worse liquidity conditions. Your Position Size Matches Working Liquidity A 20,000-contract position may seem small compared to a million-contract order book, but if only 2,200 contracts are actually available within 1 cent—and just 102 during volatility—your position is large relative to real liquidity. Selling in advance helps you avoid being forced to exit at a bad price at the worst possible moment. You Need Execution Certainty A limit order can protect your price, but not guarantee execution. A market order guarantees execution, but not price. Before the event, you have time to break up your order, wait for counterparties, and work in parts. Once new data arrives, that flexibility disappears. Partial Exit—A Compromise Solution Selling part of your position before the event (for example, to recoup your investment or lock in a target profit) reduces pressure on future decisions. You can hold the rest until settlement, without depending on liquidity at the critical moment. When Holding a Position Makes Sense This analysis doesn’t mean everyone should always exit their position. If your position is small and you’re willing to hold it to settlement, you don’t need exit liquidity. If you have a real informational edge during vote counting, you can take advantage of volatility that others avoid. Also, on some markets, liquidity may actually increase before major events as new market makers arrive. The Crowley market is just one case, not a universal rule. Here, high volatility, an unexpected twist, and a razor-thin margin all came together. The main takeaway is that liquidity can be variable and may disappear during information shocks—not that this will always happen everywhere. How to Plan Your Exit Strategy Wisely Prediction market traders usually record their entry price and probability estimate. It’s just as important to plan your exit strategy in advance. Before a major catalyst, compare your position size to real working order book depth, monitor the spread (not just headline liquidity), and decide in advance how much to sell before the event. The main lesson from the Wisconsin case: being right about the event and being able to monetize your insight are two different things. Before polls closed, Crowley traders had access to a deep and tight market, but no certainty about the outcome. During vote counting, certainty increased, but liquidity vanished. When liquidity returned, the price was already near its peak. Sometimes the best sale isn’t the one at the highest theoretical price, but the one you can make while there are still real counterparties in the market. Methodology and Limitations of the Analysis Predexon’s Kalshi data includes full order book snapshots accurate to tenths of a cent and fractional contract sizes. The analysis window is from 8:00 p.m. on August 10, 2026, to 2:45 a.m. on August 12, Central Time. For key moments, exact snapshots are used; for phases, median values from the last order book each minute. The analysis only considers visible “YES” buy orders and cannot reflect hidden interest, new orders in response to market orders, or execution quality for a specific trader. This is descriptive statistics, not proof that all price changes were caused solely by vote counting. Prediction markets involve risk and are not suitable for everyone. Even the best platforms do not guarantee outcomes, and you should never risk more than you can afford to lose. Always check the platform’s legality in your region before participating. Predexon: Kalshi Orderbook History (Sub-Cent) Wisconsin Public Radio: In Shocking Comeback, David Crowley Wins Democratic Primary for Governor Glosten, L. R., and P. R. Milgrom. “Bid, Ask and Transaction Prices in a Specialist Market with Heterogeneously Informed Traders.” Journal of Financial Economics 14 (1985): 71–100. Kyle, A. S. “Continuous Auctions and Insider Trading.” Econometrica 53 (1985): 1315–1335. Kalshi public market record: KXGOVWINOMD-26-DCRO What This Means for PPPoker Club Players For club game participants on PPPoker, this case from the prediction market world illustrates an important principle: liquidity and the ability to exit a hand or tournament profitably do not always coincide with the moment when you have the greatest edge. In poker, this shows up when the pot gets big and opponents become cautious, or when the tournament payout structure suddenly changes player motivation. Practical takeaway: plan your exit strategy in advance—whether it’s a timely fold, partial chip removal, or locking in winnings at the club. Don’t count on always being able to realize your edge at the optimal price or under optimal conditions at the most critical moment. Sometimes it’s better to lock in part of your profit early than risk everything for a maximum, but not guaranteed, result. Finally, analyze not just the size of the pot or prize pool, but also real “liquidity”—how many opponents are willing to play big pots, how quickly the game dynamics change, and how much you can influence the outcome of a hand at critical moments. This approach will help you manage risk more effectively and maximize profits in PPPoker clubs.