Bankless: How has Polymarket performed since the US election?

Author: David C, Bankless; Translator: Wuzhu, Golden Finance

The 2024 US election season has sparked a real craze for prediction markets, with Polymarket catapulting to the undisputed center of betting activity. Bloomberg has integrated Polymarket on its terminal. Reports say the Trump campaign is actively monitoring Polymarket’s odds in real time. Throughout the year, the entire market traded about $3.7 billion in election contracts.

This success has sparked a gold rush, with entrepreneurs and developers launching their own prediction market clones in hopes of getting a piece of the action. However, despite the influx of new entrants, none has been able to make a substantial breakthrough in front of the mainstream platforms.

Of the $162 million in total locked value (TVL) in prediction markets currently tracked by DefiLlama DefiLlama, Polymarket still dominates, accounting for about 85% of the total locked value (TVL) in prediction markets currently tracked by DefiLlama. However, its total value locked (TVL) has fallen from its peak during the election cycle, when daily active wallets reached 50,000, daily transactions exceeded 500,000, and on November 6 alone, transactions reached $350 million. Faced with such staggering numbers, many observers questioned whether Polymarket could maintain the momentum after the votes were counted.

Now, six months after the election, Polymarket’s trajectory has defied those skeptics and demonstrated the resilience to lay the foundation for the next phase of growth.

How has Polymarket performed since the election?

2025 is approaching, and a dark cloud hangs over the Augur prediction market platform. For those with shorter memories, Augur was once in the limelight during the 2020 election, but suffered a catastrophic crash after the votes were counted. As users left en masse, Augur's total value in ETH plummeted 20 times - from 19,000 ETH to 700 ETH. Worse, market resolutions have been repeatedly delayed due to legal disputes and design flaws, leaving the few remaining users frustrated and watching the platform exit.

But Polymarket has never had such a problem.

Rather than watching its user base evaporate, Polymarket has achieved something far more impressive: a true transformation. Users who were initially drawn to the Trump-Harris predictions have smoothly moved on to Federal Reserve rate decisions, geopolitical conflicts, token price movements, and—perhaps most importantly—sports betting markets.

Especially notable is the sharp increase in the number of daily wallets that have never participated in U.S.-related markets. These users now account for about a quarter of all Polymarket activity—a significant and healthy shift in user mix that has helped the platform maintain surprising stability amid a natural decline in bettors paying attention to the election.

Other interesting stats include:

• Cumulative user counts are roughly the same as they were in mid- to late October last year, when election activity was heating up.

• The number of monthly active traders as of April 1 was the same as last November.

Trading volume shows a similar picture, with daily trading volume remaining between $20 million and $30 million over the past three months. While it has declined since the highs, it is still roughly the same level as last September.

How accurate is Polymarket's prediction?

In addition to retaining its user base after the election, Polymarket has also recently received attention for its prediction accuracy, according to analysis by data scientist Alex McCullough. McCullough claims Polymarket has an accuracy rate of 94 percent, which excludes bets that are almost certain (more than 90 percent likely) or highly unlikely (less than 10 percent likely) because the outcomes of these bets are known but have not yet been officially announced. His findings, all documented here, include:

• Markets overestimate event probabilities:The platform sometimes overestimates event probabilities, which could be due to group bias, illiquidity, or speculative risk appetite, raising questions about its reliability.

• Sports betting is most accurate:Sports markets are the platform’s fastest-growing sector and are Polymarket’s most stable predictor, thanks to balanced betting, clear criteria, and nearly $4.5 billion in trading volume. This could skew the platform’s overall predictive power for other markets.

• Long-term predictions appear more accurate than they actually are:Long-term markets often include outcomes where the answer is clearly “no” over time. In an interview with Polymarket, McCullough cited the example of a market such as “Will Gavin Newsom be president in 2024?” that ran for months but eventually became increasingly clear that it would not happen. These extreme cases make Polymarket’s overall accuracy look better than it actually is, as these near-certain outcomes are factored into the platform’s prediction history.

Overall, Polymarket’s continued activity suggests it is not a flash in the pan, but its reliability as a predictor warrants cautious optimism pending deeper scrutiny.

Are these insights actionable?

While we may need more analysis to truly confirm Polymarket’s accuracy, McCullough’s analysis reveals some interesting and actionable trends in the workings of these markets that anyone can use to improve their own betting strategies. Here are some key takeaways from his research that hopefully will be of some use to Polymarket’s active users:

• Most markets end with a “No.”About three-quarters of the questions you ask will be answered with a “No.” Keep this in mind when looking for opportunities in “No” stocks that may be undervalued.

• The best price to bet on a “No.”History shows that the true opportunity is lower when the price of a “Yes” is between $0.20 and $0.50. Buying a “No” stock at this time can yield the greatest gains. This edge shrinks but still exists until around $0.80.

• Prices close to the edge are reasonable.If the price of a “Yes” is below $0.10 or above $0.95, you are close to the right price. Any gains are small once the fees are paid.

• Timing can work to your advantage.

• More than four hours before market close: Yes is usually overvalued → Buy No.

• Last four hours: Prices tighten and converge to accuracy → Buy now only if you are bullish on Yes in this direction.

• If you hold No, consider selling before the last hour - prices may rebound at the last moment.

• Data from sports markets may vary.

• The day before a game: The accuracy rate is about 66%. This is a good time to buy No.

• After the contest begins: Accuracy increases, reaching about 94% after two hours.

• Prices typically move every 30 minutes (30, 60, 90, 120 minutes), which means these times can be good times to adjust or place bets.

• Traders like to bet on "yes". Traders like to bet on what's going to happen, follow the hype, and the market can be quiet. Placing limit orders carefully can help you avoid paying too much.