Dissecting Polymarket's Top 10 Whales' 27,000 Transactions: The Illusion of "Smart Money" Win Rate and Survival Strategies

Author: Frank, PANews

Recently, the popularity of prediction markets continues to rise, especially with smart money’s arbitrage strategies being regarded as the gold standard. Many have started to imitate and experiment, as if a new gold rush has begun.

But behind the hype, how effective are these seemingly clever and reasonable strategies in reality? How exactly are they executed? PANews conducted an in-depth analysis of 27,000 trades made by the top ten profit-generating whales on Polymarket in December, exploring the truth behind their profits. After analysis, PANews found that although many of these “smart money” operations involved hedging arbitrage strategies, this hedging differs significantly from the simple hedging described on social media. The actual strategies are much more complex, involving not just straightforward “yes” or “no” combinations, but also fully utilizing rules like “over/under” and “win/lose” in sports betting. Another key discovery is that the extremely high win rates observed in historical holdings are largely due to a large number of “zombie orders” that remain unsettled, giving a false impression of success; the real win rate is much lower than the historical average. Next, PANews reveals the true operations of these “smart money” whales through actual cases.

1. SeriouslySirius: 73% win rate masked by “zombie orders” and complex quantitative hedging SeriouslySirius is the top address in December, with a profit of about $3.29 million and a total historical profit of $2.94 million. Based solely on completed orders, his win rate is as high as 73.7%. However, the reality is that this address still holds 2,369 orders, with 4,690 orders settled so far. Among these, 1,791 open positions have actually failed completely, but the user has not closed them individually. On one hand, this saves a lot of effort and transaction fees. On the other hand, since most of his closed orders are profitable, the historical settled data shows an extremely high win rate. When considering these unsettled “zombie orders,” the true win rate drops to 53.3%, only slightly above a coin flip. In actual trading, about 40% of his orders are multiple hedging bets on the same event. However, this hedging is not just simple “YES” + “NO.” For example, in an NBA game between the 76ers and Mavericks, he bought 11 different directions, including Under, Over, 76ers (home team), Mavericks (away team), and others, ultimately earning $1,611. During this process, he also employed arbitrage strategies with insufficient probabilities, such as buying the 76ers’ victory at a 56.8% chance and Mavericks at 39.37%, with a combined cost of about 0.962, ensuring profit regardless of the outcome. Ultimately, he made a profit of $17,000 on this game. However, this strategy does not always work; for example, in a Celtics vs Kings game, he participated in nine directions and ended up losing $2,900. Additionally, many trades show severe imbalance in capital allocation, such as placing bets on two directions but with more than tenfold difference in invested funds. This likely results from insufficient market liquidity, indicating that while arbitrage strategies look promising, liquidity can be a major obstacle in practice. Opportunities may appear, but they do not necessarily allow for perfect hedging on both sides. Moreover, since these are automated trades, buy and sell operations under such conditions are very likely to result in significant losses. Nevertheless, the reason SeriouslySirius can achieve large profits with this strategy is primarily due to proper position management, with a profit-loss ratio of about 2.52. This is the main reason he can profit despite a relatively low actual win rate. Furthermore, this strategy is not always profitable; before December, this address’s profit and loss situation was not optimistic, with a long period near break-even, and a maximum loss once reaching around $1.8 million. Now, with a more mature strategy, it remains uncertain whether such profits can be sustained.
2. DrPufferfish: Turning small probabilities into large probabilities, an art of “profit-loss ratio” management DrPufferfish is the second most profitable address in December, with a monthly profit of about $2.06 million, and an even more impressive historical win rate of 83.5%. However, considering the large number of “zombie orders,” his actual win rate has reverted to 50.9%. The strategy here differs significantly from SeriouslySirius. Although about 25% of his orders are hedging, this is not opposite hedging but rather diversified betting. For example, in a US professional baseball championship, he bought 27 teams with low probabilities, with the combined probabilities exceeding 54%. This approach turns low-probability events into high-probability outcomes. The main reason for his huge gains is his ability to control the profit-loss ratio. Take Liverpool, for example, a team he favors; he has predicted their results 123 times, ultimately earning about $1.6 million. Among profitable predictions, the average profit is about $37,200, while the average loss on unsuccessful predictions is about $11,000. Most of these losing bets are sold early to control position losses. This operational approach results in an overall profit-loss ratio of 8.62, with high profit expectations. However, overall, his strategy is not just simple arbitrage hedging but involves professional prediction analysis and strict position management. Also, most of his hedging trades are in a loss state, with total unrealized losses of about $2.09 million. It appears that these hedging trades are mainly used as insurance.
3. gmanas: High-frequency automated pipeline operations The third-ranked address, gmanas, has a similar style to DrPufferfish, achieving a total profit of about $1.97 million in December. His true win rate is close to 51.8%. He executes trades at a higher frequency, with over 2,400 predictions completed, clearly relying on automated processes. His betting style is similar to the previous address, so no further details are necessary.
4. Hunter simonbanza: Treating probability predictions as “K-line” swing trading Fourth place goes to simonbanza, a professional prediction hunter. Unlike the previous addresses, he does not use hedging orders at all. His realized profit is about $1.04 million, with unrealized zombie order losses of only $130,000. Although his capital and trading volume are not large, his true win rate is the highest among these whales, at about 57.6%. The average profit per settled order is about $32,000, with an average loss of $36,500. The profit-loss ratio is not high, but his high win rate ultimately yields good returns. Additionally, he has very few zombie orders—only six—because he usually does not wait for the event to conclude but instead profits from probability fluctuations. In simple terms, he takes profits when he is ahead and does not hold out for the final outcome. This represents a unique prediction market investment approach, where probability changes are akin to stock market rises and falls. The specific logic behind his high win rate remains unknown, as it is his exclusive secret.
5. Whale gmpm: Asymmetric hedging strategy, using “large positions” to secure certainty Fifth place, gmpm, although ranked fifth in December profit and loss, has a higher total historical profit of $2.93 million. His true win rate is about 56.16%, also relatively high. His approach is similar to the fourth address but with a unique core strategy. For example, he often places bets on both sides of the same event, but his strategy seems to involve investing more capital on the side with higher probability and less on the lower probability side. This allows him to have larger positions when the probability is high, but also limits losses when low-probability events occur, achieving a hedging effect. In practice, this is a more advanced hedging strategy, not relying solely on “yes” + “no” <1 mathematical arbitrage, but combining comprehensive event judgment and hedging to reduce losses.
6. Workaholic swisstony: “Ants Moving” high-frequency arbitrage Sixth, swisstony is a super high-frequency arbitrage address, with the highest trading frequency among these whales, executing 5,527 trades in total. Despite earning over $860,000, the average profit per trade is only $156. His strategy resembles “ants moving,” typically buying all the odds of a single event. For example, in a Jazz vs Clippers game, he bet on 23 different markets. Due to the small investment size, his capital distribution is relatively balanced, providing some hedging effect. However, this strategy heavily tests the details of entry, such as ensuring “yes” + “no” <1. Interestingly, his hedging orders often have total amounts exceeding 1, which ultimately leads to losses regardless of the outcome. Nevertheless, with reasonable profit-loss ratios and win rates, his overall profit expectation remains positive.
7. Outlier 0xafEe: The “pop culture prophet” with unconventional tactics Seventh, 0xafEe is a low-frequency, high-win-rate trader. His trading frequency is very low, averaging only 0.4 trades per day, with a real win rate of 69.5%. Among his completed orders, he earned about $929,000, with very few zombie orders—only about $8,800 unrealized loss. He also never uses hedging orders, focusing solely on predictions. His predictions mainly target Google search trends and pop culture topics, such as “Will Pope Leo XIV become the most searched person on Google this year?” or “Will Gemini 3.0 be released before October 31?” He seems to have a unique analytical approach for these topics, resulting in a very high success rate, making him an outlier among top whales—an address outside the sports category.
8. Manual hedging player 0x006cc: From simple to complex hedging strategies Eighth place, 0x006cc, is similar to the previous complex hedging addresses, with a total net profit of about $1.27 million and a real win rate of about 54%. Compared to other automated addresses, his trading frequency is low, averaging only 0.7 trades per day. Early on, he likely used a simple manual hedging strategy. Since December, this simple hedging has evolved into more complex strategies. His trading history shows that as more people understand hedging, he has gradually upgraded his approach.
9. Cautionary example RN1: When “hedging” becomes a “loss formula” Ninth, RN1 is among the top ten profit addresses in December but is overall in loss. His realized profit is about $1.76 million, but unrealized losses reach $2.68 million, resulting in a total loss of about $920,000. As a cautionary example, there are many lessons to learn from RN1. First, his true win rate is only 42%, the lowest among these addresses, and his profit-loss ratio is only 1.62. Combining these metrics indicates his profit expectation is negative; overall, this strategy is not profitable. A closer look reveals that RN1 is also a clear arbitrage strategy address. However, many of his hedging trades, while satisfying the “yes” + “no” <1 condition, tend to involve more capital on the low-probability side and less on the high-probability side, leading to less favorable outcomes when high-probability events occur, resulting in actual losses.
10. Gambler Cavs2: Unilateral heavy positions in hockey, luck over strategy Tenth, Cavs2 is also a gambler who prefers unilateral heavy positions, mainly in NHL hockey. Overall, his profit is about $630,000, with a real win rate of about 50.43% and a low hedge ratio of 6.6%. The data is average, with luck playing a significant role, as he successfully predicted some high-yield single-game results. His overall strategy offers limited reference value. The 5 Harsh Truths After Disenchanting “Smart Money” After an in-depth analysis of these “smart money” trades, PANews summarizes the reality behind the “wealth stories” of prediction markets.
11. “Hedging arbitrage strategies” are far from simple probability conditions; under fierce market competition and liquidity constraints, they can easily turn into counterproductive loss formulas. Blind imitation is not advisable.
12. “Copy trading” in prediction markets also seems ineffective, mainly because: first, the rankings or win rates seen are based on historical settled profit data, which are “distorted” figures; behind such data, many “smart money” are not truly “smart.” The actual win rate exceeding 70% is rare, and most are similar to flipping a coin. Additionally, the trading depth in prediction markets is currently limited, so arbitrage opportunities may only accommodate small capital entries, and copy traders might be squeezed out.
13. Managing profit-loss ratios and position sizes is more important than chasing win rates. Among the top-performing addresses, a common trait is excellent management of profit-loss ratios. For example, gmpm and DrPufferfish often exit positions based on probability trends to reduce losses and improve ratios.
14. The real secret lies beyond “mathematical formulas.” Currently, many social media interpretations of “arbitrage formulas” seem reasonable at first glance, but in practice, these “smart money” rely on skills beyond formulas—either strong judgment of certain events or unique cultural analysis models. These unseen decision algorithms are key to their success. For users lacking such algorithms, prediction markets remain a cold “dark forest.”
15. The profit scale in prediction markets is still small. For these top whales in December, the largest total profit is only around $3 million. Compared to the profit potential in the crypto derivatives market, the space appears limited. For those dreaming of overnight riches, the market size is still insufficient. Such a niche, highly specialized, and small-scale market is unlikely to attract institutional players, which may be a key reason for the limited growth of prediction markets. In the seemingly golden prediction market of Polymarket, the so-called “god-level whales” are mostly surviving gamblers or diligent arbitrageurs. The true secret to wealth is not hidden in those inflated win rate rankings but in the algorithms used by a few top players who bet with real money after filtering out noise.