--- title: "Sharp Money in Polymarket Sports Markets: 155,832 Graded Bets, Checked Against the Result" canonical: "https://0xinsider.com/research/sharp-money-polymarket-sports" datePublished: "2026-09-12" dateModified: "2026-09-12" description: "Across 155,832 Polymarket sports buys of $1,000 or more, wallets graded S, A or B beat the price they paid by 1.25 points. D and F wallets lost 1.21. The whole gap opens after kickoff." --- # Sharp Money in Polymarket Sports Markets: 155,832 Graded Bets, Checked Against the Result Last updated: 2026-09-12 Canonical: https://0xinsider.com/research/sharp-money-polymarket-sports Sports prices on Polymarket match results on average. This study asks who is on each side of that average. Every sports buy of $1,000 or more between June 1 to September 11, 2026, 155,832 of them, is scored against the grade the wallet already held on the day it traded. Wallets graded S, A or B won 65.9% of the time at an average price of 64.7 cents, beating the price by 1.25 points. Wallets graded D or F lost to it by 1.21. The gap holds in every price bucket and every market type, and almost all of it opens after kickoff. ## Sharp money is a claim about people, not prices The calibration study at 0xinsider.com/research/favorite-longshot-bias-polymarket-sports found that large buyers on Polymarket sports markets paid, on average, what the result was worth. An average like that hides who is on each side of it. If some wallets are consistently right and others consistently wrong, the market can be calibrated and still have winners. 0xinsider grades every Polymarket wallet from S to F on its own settled record. The grade study at 0xinsider.com/research/do-wallet-grades-predict-outcomes tested that grade against every large buy on the platform, 67,531 buys of $10,000 or more, and found S, A and B wallets +1.57 points ahead of the price and D and F wallets -2.06 behind it. Sixty-three percent of that sample was soccer, so the question of whether the grade works in sports was already half answered. This is the other half. The same method, restricted to sports markets, with the floor lowered to $1,000 so that tennis, esports and baseball have samples worth reading, and with two splits the earlier study did not make: by market type, and by whether the buy landed before or after kickoff. ## 155,832 graded sports buys The sample is every buy of $1,000 or more that 0xinsider recorded on a Polymarket sports market between June 1 to September 11, 2026, on a market that settled after the trade, at a price between 2 and 98 cents. 155,832 buys, 7,488 wallets, 15,270 settled markets, $2.50 billion in stakes. Grade assignment is point-in-time. For each buy the query takes the most recent daily ranking dated on or before the day of the trade, so a grade computed after the game settled never scores the buy it would have predicted. The window starts on June 1 because grade coverage widened that month, from about 238 wallets scored per day in May to 18,464 in June. Earlier than that, no grade mostly means nobody had been scored yet. Wallets are pooled into four cohorts: S, A and B; C; D and F; and no grade on the trade date. The edge is the share of buys that won minus the average price paid, in percentage points. The dollar return is stake-weighted, held to settlement, before fees. Intervals resample markets, not trades, with 5,000 draws. ## S, A and B wallets beat the price in sports; D and F lose to it Wallets graded S, A or B made 48,255 of the buys, across 1,547 wallets and 8,316 settled markets, worth $818.1 million. They won 65.9% of the time at an average price of 64.7 cents: plus 1.25 points, and plus 2.04% on the dollar. The 95% interval runs from plus 0.20 to plus 2.31 and clears zero. Wallets graded D or F made 43,615 buys across 3,241 wallets and 7,057 markets, worth $650.8 million. They won 58.3% at an average price of 59.5 cents: minus 1.21 points, and minus 0.05% on the dollar. The interval runs from minus 2.72 to plus 0.33. It touches zero, so the D and F loss is the likely reading rather than the established one. C wallets came in at plus 0.63 points and wallets with no grade yet at minus 0.62, both with intervals well across zero. The picture is the same as the all-category study: the grade separates the top three letters from the bottom two and says little in the middle. Restricted to buys of $10,000 or more, for a direct comparison with the earlier study, S, A and B wallets came in at plus 1.41 points over 23,613 buys and D and F at minus 1.83 over 16,633. The all-category study reported +1.57 and -2.06 on a sample that was mostly sports, so the two agree. ## Letter by letter, from S to F S wallets made 26,743 buys from 128 wallets, worth $391.7 million, and came in at plus 1.13 points. A wallets, 9,050 buys from 494 wallets, plus 1.08. B wallets, 12,462 buys from 1,259 wallets, plus 1.63. C, 37,712 buys from 3,112 wallets, plus 0.63. D, 10,600 buys from 1,767 wallets, minus 1.22. F, 33,015 buys from 2,114 wallets, minus 1.21. B ahead of S is not evidence that B is the better grade, and D level with F is not evidence the two are the same. At these sample sizes the order inside each group is noise. What the letters show is a step: three above zero, one near it, two below. Dollar returns tell the size story. S wallets returned plus 0.66% on $391.7 million, less than A or B on far more money, because S wallets concentrate stakes in fewer, larger positions where one settled game moves the dollar figure more than the per-buy edge. ## Not a favorites effect The obvious objection is that well-graded wallets buy favorites and favorites win. Split the same buys into five bands by the price paid and the objection fails in all five. Under 20 cents, S, A and B came in at plus 4.98 points against plus 2.33 for D and F. At 20 to 40 cents, plus 0.43 against minus 1.53. At 40 to 60, plus 0.36 against minus 0.85. At 60 to 80, plus 2.38 against minus 0.13. At 80 cents and up, plus 1.10 against minus 3.73. The gap is widest at the top, as it was in the all-category study. A well-graded wallet paying 87 cents for a favorite was right more often than a poorly graded wallet paying the same 87 cents, by nearly five points. Expensive contracts are where a bad read costs the most. ## Soccer, esports and tennis separate; baseball does not Soccer is 35,083 of the 48,255 S, A and B buys, so the cohort headline is mostly a soccer number. In soccer, S, A and B wallets won 66.9% at 65.4 cents, plus 1.54 points, while D and F wallets won 59.4% at 60.1 cents, minus 0.66. Esports separates more sharply on fewer buys: plus 1.77 for S, A and B against minus 2.44 for D and F, with the graded cohort returning plus 6.13% on the dollar and the poorly graded one minus 7.45%. Tennis has the widest D and F loss of any sport, minus 4.46 points over 3,251 buys, against plus 0.72 for S, A and B. Baseball is the exception. Every cohort lost to the price: S, A and B at minus 1.22, D and F at minus 1.52, no grade at minus 3.01. The grade does not separate baseball buyers in this window, and nothing in the data says why. The honest reading is that a grade earned on soccer and esports carries into tennis and does not obviously carry into baseball. Wallets with no grade yet look like C wallets in soccer and esports and like D and F wallets in tennis and baseball. A new wallet is a blank, not a bad sign. ## Spreads and totals show the widest gap By Polymarket's market type, S, A and B wallets beat D and F wallets on all four with a sample worth reading. Moneylines, the bulk of the buys, are the narrowest: plus 0.73 against minus 0.50. Child moneylines, which cover one map, set or period, plus 1.94 against minus 1.32. Spreads are where the graded cohort did best: plus 2.06 points and plus 10.72% on the dollar across 4,941 buys, against minus 0.60 and plus 0.74% for D and F. Totals are where the poorly graded cohort did worst: minus 3.62 points across 7,037 buys, against plus 0.72 for S, A and B. A moneyline is a pick. A spread or a total is a number, and a wallet that has been right about numbers before is more likely to be right about the next one than a wallet that has not. That is the pattern a grade built on settled history should produce, and here it does. ## The whole edge opens after kickoff Split every buy by whether it landed before or after the market's kickoff time and the cohorts look different before the game and different again during it. Before kickoff, S, A and B wallets came in at plus 0.27 points, D and F at minus 0.52, and no grade at minus 0.33. Three cohorts within a point of each other and within a point of zero. Pre-game, in this window, the grade barely separated anyone. In-play, S, A and B wallets came in at plus 1.95 points and plus 4.48% on the dollar across 24,090 buys. D and F wallets came in at minus 2.56 points and minus 7.66% on the dollar across 15,473. No grade, minus 1.06. The four and a half points between the two graded cohorts is the whole finding of this study, and it is a live-betting finding. The timing study at 0xinsider.com/research/when-large-sports-bets-land goes further into this: S, A and B wallets place 51% of their sports buys in-play against 37% for D and F, and the separation is widest in the first hour of play, plus 3.07 points against minus 3.03. Whatever the graded wallets know, they act on it while the game is being played. ## Ten wallets carry half the cohort The S, A and B cohort is 1,547 wallets, and it is top-heavy. The ten most active carry 53.8% of the cohort's buys and 50.4% of its dollars. The 128 S wallets alone made 26,743 of the 48,255 buys. That cuts two ways. The cohort edge is, in large part, the edge of a small group of wallets that bet sports at size and often, and a study of 1,547 wallets is partly a study of ten. It also means the grade is doing what a grade is for: those ten wallets are S, A or B because their record earned it, and the record kept predicting. It is the reason to read the cohort number as a property of the grade and not of a typical graded wallet. Most S, A and B wallets made a handful of sports buys in the window. The ones that made thousands are the ones that move the figure. ## Month by month The separation holds in all four months. In June, S, A and B came in at plus 0.79 and D and F at minus 1.10. July, plus 2.10 against minus 1.19. August, plus 1.62 against minus 2.83. September, through the 11th, minus 0.53 against minus 4.64. The later months are small. The alert floor rose from $1,000 to $10,000 in August, so August and September record only the largest bets, and the graded population thinned from about 18,500 wallets a day in June and July to about 3,900 after. September is 774 S, A and B buys and 359 D and F buys. Read it as consistent with the rest and not as a fifth data point. ## What this study does not show The edge is measured at each buy's own price against the settled result. It ignores fees and slippage, treats every position as held to settlement, and scores a wallet that exited early on a result it never waited for. It is a test of whether the grade carries information, not of whether copying a graded wallet's trade is profitable after you have paid to get in. Grades are computed from a wallet's own earlier settled trades. The lookup is point-in-time, so no buy is ever scored by a grade that saw its own game resolve, but the same wallets whose history built the grade are the ones being tested. This is not a held-out universe. The D and F interval touches zero. The S, A and B result is established at the 95% level; the D and F loss is likely. The C and no-grade cohorts are indistinguishable from the market. And the in-play edge may be speed. A wallet that is right about a live price two seconds before everyone else is right for a reason the grade cannot see. Whether the graded wallets are reading the game better or reading the feed faster is a question this data cannot answer. Either way they were right. ## How to use it on 0xinsider Every Polymarket wallet has a grade on 0xinsider, and every grade has the record behind it. The leaderboard at 0xinsider.com/leaderboard ranks wallets by that record and each profile shows the calibration edge, the same win-rate-minus-price measure this study uses, alongside realized P&L and the positions the wallet holds now. On each sports board, 0xinsider.com/sports/soccer, 0xinsider.com/sports/tennis, 0xinsider.com/sports/esports, 0xinsider.com/sports/nba and the rest, the sharp money strip on a game shows which side the profitable wallets are on and how much they have on it. That is this study's S, A and B cohort, per game, before kickoff and during play. The pick of the day at 0xinsider.com/pick-of-the-day is one position a day that S, A and B wallets are already holding, published with the entry price and the closing price so every pick's closing line value is on the record. The whale alerts feed at 0xinsider.com/whale-alerts shows each large buy as it lands with the buyer's grade beside it, which is the in-play signal this study found. ## How to check it yourself Every number on this page comes from one read-only run of a set of queries against the production database at 03:27 UTC on September 12, 2026. The query text, the raw output and the bootstrap script are public at github.com/0xinsider/research, under polymarket-sports-markets, in the files whose names carry the words sharp-money. The bootstrap reads market-level exports the queries write, so reproducing the intervals needs a database; the committed output is the run these figures come from. The universe is every row of the large-trade alert table for a Polymarket buy on a sports-category market from June 1, 2026, joined to the settled outcome of its market, with the resolution timestamp after the trade and the price between 2 and 98 cents. The grade is the most recent daily ranking dated on or before the trade day. The grading method is documented at 0xinsider.com/transparency and explained at 0xinsider.com/learn/prediction-market-trader-grades-explained. The window cannot be extended backward. Before April 2, 2026 the alert table's outcome index was a defaulted zero for a large share of buys, and before June 1 the graded population was a few hundred wallets a day. Both boundaries are described in the calibration study. Expect your counts to grow. Every buy sits on a settled market, and markets keep settling. The cohort edges should hold within their intervals; the September row will fill in. ## FAQ ### Do 0xinsider wallet grades predict sports outcomes on Polymarket? Across 155,832 Polymarket sports buys of $1,000 or more from June 1 to September 11, 2026, wallets graded S, A or B won 65.9% of the time at an average price of 64.7 cents, beating the price by 1.25 points with a market-clustered 95% interval of plus 0.20 to plus 2.31. Wallets graded D or F lost to the price by 1.21 points. ### Is the sports edge just graded wallets buying favorites? No. Split by the price paid, S, A and B wallets beat D and F wallets in all five bands, from under 20 cents to 80 cents and up. The gap is widest above 80 cents, where the graded cohort came in at plus 1.10 points and the poorly graded one at minus 3.73. ### Where does the sharp money edge in sports come from? From in-play buys. Before kickoff, S, A and B wallets came in at plus 0.27 points and D and F at minus 0.52, within a point of each other. After kickoff, S, A and B came in at plus 1.95 and D and F at minus 2.56, four and a half points apart. ### Does the grade work in every sport? It separated the cohorts in soccer, esports and tennis. In baseball every cohort lost to the price, including S, A and B at minus 1.22 points, so the grade did not separate baseball buyers in this window. ### Can I see which wallets these are? Every Polymarket wallet's grade and record is public on 0xinsider. The leaderboard at 0xinsider.com/leaderboard ranks them, each profile shows the calibration edge behind the grade, and the sharp money strip on each sports board shows which side the profitable wallets are on for a given game.