---
title: "Favorite-Longshot Bias on Polymarket Sports Markets: 411,770 Large Bets, Priced Against the Result"
canonical: "https://0xinsider.com/research/favorite-longshot-bias-polymarket-sports"
datePublished: "2026-09-12"
dateModified: "2026-09-12"
description: "Across 411,770 Polymarket sports buys of $1,000 or more, prices matched results to within 0.02 points. The longshot penalty exists only under 10 cents, where buyers lost 61% of their stake."
---

# Favorite-Longshot Bias on Polymarket Sports Markets: 411,770 Large Bets, Priced Against the Result

Last updated: 2026-09-12
Canonical: https://0xinsider.com/research/favorite-longshot-bias-polymarket-sports

Every sportsbook dataset ever published shows the same thing: longshots return less than favorites because the public overpays for a big payout. This is the same test on Polymarket sports markets, where the crowd sets the price on an order book. Across 411,770 buys of $1,000 or more between April 2 to September 11, 2026, buyers paid 60.6 cents on average and won 60.7% of the time. The bias survives in one place: under 10 cents, where buyers lost 61 cents of every dollar.

## The bias every sportsbook study finds

The favorite-longshot bias is the oldest result in betting research. Richard Griffith found it at American racetracks in 1949: horses at long odds were bet more heavily than their win rate justified, and favorites less. Every large sportsbook dataset published since has shown the same shape. A bet at +900 returns less per dollar than a bet at -300, and both lose to the house, but the longshot loses far more.

The usual explanation has two parts. Bettors overestimate small probabilities and enjoy the size of a possible payout, so they overpay for it. And a bookmaker who knows that shades the longshot price further, because the demand is there. The bias is a fact about the customers and about the margin built on top of them.

Polymarket sports markets have no bookmaker. A contract on a team pays $1 if the team wins and nothing if it does not, and the price is whatever the last buyer and seller agreed on. If the bias is a fact about bettors, it should show up here too. If it depends on a bookmaker shading the line, it should not.

## 411,770 sports buys, scored against the price paid

The sample is every buy of $1,000 or more that 0xinsider recorded on a Polymarket sports market between April 2 to September 11, 2026, on a market that settled after the trade cleared, at a price between 2 and 98 cents. That is 411,770 buys by 16,575 wallets across 34,120 settled markets, carrying $8.21 billion in stakes. Soccer, esports, the NBA, tennis, baseball, hockey, cricket, MMA and the WNBA are all in it.

Buys only, because a buy at a price on a named outcome is a clean bet on that outcome. A sell can be a hedge, an exit or a market-making fill, and its price says nothing about what the seller believed.

Each buy is scored two ways. The edge is the share of buys that won minus the average price paid, in percentage points: buy at 60 cents and win 61% of the time and the edge is plus one point. The dollar return is what a dollar staked came back as, held to settlement, before fees. A bucket of 10-cent contracts that pays out 8% of the time returns 80 cents on the dollar, a 20% loss, even though the edge is only minus two points. Both numbers are on every chart below because they answer different questions.

Confidence intervals resample markets rather than trades. Fifty buys of the same side of the same game are one bet on that game, and treating them as fifty is how a study manufactures precision it does not have. 5,000 resamples, seed 20260912.

## Prices matched results to within a fiftieth of a point

Across the whole sample the average price paid was 60.6 cents and the share of buys that won was 60.7%. The edge is +0.02 points. The dollar return, before fees, is +0.46%. On average, a large buyer on a Polymarket sports market paid what the outcome was worth.

Split into ten price buckets, every bucket from 20 cents to 98 cents sits within a point and a half of zero, and every one of those intervals straddles zero. Buyers at 40 to 50 cents paid 45.4 cents and won 45.8%. Buyers at 50 to 60 cents paid 54.1 and won 53.4%. Buyers at 80 to 90 cents paid 84.7 and won 84.3%. Buyers at 90 to 98 cents paid 93.5 and won 92.8%.

The one bucket whose interval clears zero is the cheapest. Buys under 10 cents paid 6.49 cents on average and won 2.97% of the time, an edge of minus 3.52 points with a 95% interval from minus 5.50 to minus 0.28. Everything the textbook says about longshots is true on Polymarket, and it is true only below a dime.

## Under 10 cents: the one place the bias shows

In five-cent steps the picture is sharper. Buys between 2 and 5 cents paid 3.19 cents and won 0.97% of the time: a dollar staked came back as 65 cents. Buys between 5 and 10 cents paid 7.62 cents and won 3.66%: a dollar came back as 32 cents. Those 2,020 buys put $28.2 million into contracts that were worth, at the frequencies they actually paid, less than half the price.

This is the classic shape. A 7-cent contract is a bet on something the market says happens one time in fourteen. The people buying $1,000 or more of it are paying for a fourteen-to-one payout, and in this sample they got a payout that arrived one time in twenty-seven. The bias does not need a bookmaker. It needs a buyer who wants the big number.

Two things about the size of the effect. It is concentrated: those buys are half a percent of the sample by count and a third of a percent by dollars. And it is large where it lives: minus 61% on the dollar is the worst return of any bucket in any sport in this study by a wide margin. If you take one thing from this page, take this. On Polymarket sports markets, a large buy under 10 cents has been a bad bet.

## The 10 to 20 cent bucket is one market

Buys between 10 and 20 cents show a dollar return of plus 27.6% and an edge of plus 4.61 points. Read alone, that says the second-cheapest bucket was the best bet on the platform. It was not. The confidence interval runs from minus 5.64 to plus 19.10 points, the widest of any bucket, and the reason is a single market.

Will Spain win the 2026 FIFA World Cup? drew 490 buys of $1,000 or more at an average of 15.2 cents, $7.48 million in all. Spain won. Those buys paid out $41.31 million in profit. The whole 10-to-20-cent bucket, all 5,913 buys across 1,556 markets, shows a combined profit of $25.89 million. The other 1,555 markets in the bucket lost $15.4 million between them.

This is what market-clustered intervals are for. A study that treated those 490 buys as 490 independent observations would report a significant longshot edge that is one tournament. The honest reading of the 10-to-20-cent bucket is that it is indistinguishable from zero, and that the bias below 10 cents does not obviously extend above it.

## Favorites carry no penalty here

The other half of the textbook bias is that favorites are underbet. At a sportsbook that means a -400 favorite still loses money over time, because the house margin is built into the line, but it loses less than the longshot on the other side. The favorite is the better of two bad bets.

On Polymarket sports markets the favorite is not a bad bet at all before fees. Buyers at 80 to 90 cents came out at minus 0.44 points and plus 0.36% on the dollar. Buyers at 90 to 98 cents came out at minus 0.69 points and minus 0.60% on the dollar. Buyers at 70 to 80 cents came out ahead, plus 0.91 points and plus 3.58% on the dollar, across 49,123 buys.

Fees change this by a known amount. Polymarket's sports taker fee is 0.05 times the share count times the price times one minus the price, which at 90 cents is 0.5% of the stake and at 95 cents is 0.25%. A resting limit order pays no fee. The favorite that returns minus 0.6% before fees returns about minus 1.1% after them as a taker and minus 0.6% as a maker. Either way it is a long way from a sportsbook, where the same favorite carries the full hold.

## Sports against the rest of Polymarket

Run the same ten buckets over every large buy on politics, crypto, geopolitics, culture, finance, world and business markets in the same window and the picture changes completely. Those 69,403 buys miss the price by six to ten points in most buckets, always in the buyer's disfavor.

Buyers at 20 to 30 cents on those markets paid 25.1 cents and won 16.1% of the time, minus 9.0 points. Buyers at 40 to 50 cents paid 45.4 and won 39.3%, minus 6.1. Buyers at 90 to 98 cents paid 94.1 and won 84.4%, minus 9.7, across 25,076 buys. The only buckets near zero are 60 to 80 cents.

That is not a statement about which markets are mispriced in the abstract. Non-sports markets settle over months, the sample is thinner, and a run of surprising political outcomes in one summer can move every bucket at once. It is a statement about where a large buyer's price has matched the result. In sports it has. Elsewhere on the platform, in this window, it has not.

## Sport by sport

Soccer is the largest sample, 156,545 buys across 11,653 settled markets and $3.04 billion, and it is calibrated: buyers paid 61.0 cents and won 61.7%, plus 0.70 points and plus 1.44% on the dollar. Esports, 69,328 buys, came in at plus 0.07 points. The NBA, 61,913 buys across 1,813 games, minus 0.20. Tennis minus 0.57, baseball minus 0.52, hockey minus 0.64, all on tens of thousands of buys.

Cricket is small and positive, plus 1.22 points over 6,130 buys, most of them in-play. Basketball outside the NBA and WNBA is 1,971 buys at plus 0.59 points. The WNBA is 1,454 buys at minus 0.57.

MMA is the outlier. Across 5,350 buys, buyers paid 57.9 cents and won 52.2%, minus 5.71 points and minus 5.34% on the dollar. Nothing else in the study is more than a point and a quarter from zero.

## MMA favorites lose, and two fights explain most of it

Split every sport into buys under 30 cents, between 30 and 70, and at 70 and above, and the favorite band is where the sports differ. Soccer favorites at 70 cents and up paid 83.9 cents and won 83.9%, to the decimal. Tennis favorites paid 84.1 and won 85.4%. Esports favorites paid 81.7 and won 82.3%. NBA favorites paid 84.3 and won 83.2%. Baseball favorites paid 83.3 and won 80.9%, the second-worst band.

MMA favorites paid 80.4 cents and won 61.2% of the time. That is minus 19.1 points and minus 17.1% on the dollar across 1,488 buys. Two fights carry most of it. Khamzat Chimaev went off at 81 cents against Sean Strickland at UFC 328 and lost, with 222 large buys and $4.29 million on him. Ilia Topuria went off at 80 cents against Justin Gaethje at UFC Freedom 250 and lost, with 148 large buys and $1.57 million.

MMA has 344 settled markets in this sample against 11,653 for soccer, so one card moves the whole sport. Read the MMA number as a warning about small-sample sports rather than a finding that fight favorites are systematically overpriced. What it does establish is that the calibration seen in soccer, tennis and basketball is not a law of the platform. It is a property of deep, frequently traded markets, and a fight card is neither.

## Moneylines, spreads and totals

Polymarket tags each sports market with a type. Moneylines are 269,185 of the buys and $5.59 billion of the stakes, and they are calibrated to the hundredth: 61.2 cents paid, 61.2% won, plus 0.01 points. Totals, 45,585 buys, came in at minus 0.58 points and minus 0.85% on the dollar. Spreads, 30,094 buys, came in at plus 0.66 points and plus 2.24% on the dollar, the best of the three. Child moneylines, which cover a single map, set or period, came in at plus 0.54.

The soccer specials behave the same way. Team-to-advance markets, 4,828 buys, plus 0.05 points. Both-teams-to-score, 3,589 buys, plus 1.19. Exact-score markets are bought almost entirely as favorites at 86.8 cents on average, and they won 86.8% of the time.

Within each type the tails look like the whole. Moneyline buys at 70 cents and up paid 83.1 and won 82.9%. Spread buys at 70 and up paid 83.8 and won 86.3%. Total buys at 70 and up paid 84.0 and won 82.5%. None of the market types shows a favorite premium worth the name.

## What this study does not show

The edge and the dollar return are measured at each buy's own price against the settled outcome. They ignore fees and slippage, and they treat every position as held to settlement. A wallet that bought at 60 cents and sold at 75 before the game ended is scored on the final result, which is not what it made.

The sample is large buys 0xinsider recorded as alerts, not every fill on the platform. The alert floor was $1,000 through July 2026 and rose to $10,000 in August, so August and September are thinner and skew to bigger bets. Large buyers are not a random sample of bettors, and a market can be calibrated for them while retail buyers of $20 contracts pay a different price.

Only settled markets are in the sample, which tilts it toward the four months in the middle of the window. Pre-kickoff and in-play buys are mixed together here; the timing study at 0xinsider.com/research/when-large-sports-bets-land splits them, and the split matters for who is on each side.

And a calibrated market is not a market nobody beats. Prices that match results on average mean the winners and losers among large buyers net to zero, not that there are no winners. The sharp money study at 0xinsider.com/research/sharp-money-polymarket-sports takes the same sports buys, from June onward, and splits them by the grade the wallet held on the day it traded.

## What it means for a sports bettor

On a Polymarket sports market, the price is a fair estimate of the outcome. Not a house line with a margin folded in, and not a number the public has pushed off its value except at the cheapest end. That changes what analysis is for. There is no vig to beat and no systematic mispricing to harvest, so a bettor's edge has to come from knowing something the price does not yet reflect, or from knowing who is on the other side.

The second of those is what 0xinsider shows. Every wallet on the platform has a public record, and 0xinsider grades it from S to F on its settled history. On each sports board, 0xinsider.com/sports/soccer, 0xinsider.com/sports/nba, 0xinsider.com/sports/nfl and the rest, the sharp money strip on each game shows which side the profitable wallets are on and how much they have on it. The whale alerts feed at 0xinsider.com/whale-alerts shows every large buy as it lands with the buyer's grade beside it. The pick of the day at 0xinsider.com/pick-of-the-day is built from positions those wallets are already holding, with the entry price and closing price recorded for every pick.

And the one rule this study supports on its own: do not buy the 5-cent contract because the payout is twenty to one. In this sample it paid out one time in twenty-seven.

## 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 committed in the 0xinsider repository under docs/research-articles/verification with the date 2026-09-12 and the word calibration in the file names.

The universe is every row of the large-trade alert table for a Polymarket buy on a sports-category market, joined to the settled outcome of its market, with the resolution timestamp after the trade timestamp and the price between 2 and 98 cents. The edge is the mean of a won-or-lost indicator minus the mean price. The dollar return is the stake-weighted sum of won divided by price minus one.

The window starts on April 2, 2026 for a reason worth knowing before you run anything earlier. Before that day the alert table's outcome index was a defaulted zero for a large share of buys. In February, buys tagged outcome zero at 80 cents or more won 55.7% of the time while buys tagged outcome one won 88.5%. In March, 52.8% against 84.0%. From April 2 onward both sides win about 88%, as they should. A study that includes February and March reads a reverse favorite-longshot bias of 25 points that is a data defect, and an early pass of this analysis did exactly that.

Expect your counts to come out higher than these. Every buy in the sample sits on a market that has settled, so the universe grows as open markets resolve. The bucket-by-bucket picture should not move; if a rerun moves a bucket by more than its interval, that is a finding, not a reproduction.

## FAQ

### Does the favorite-longshot bias exist on Polymarket sports markets?

Only at the cheapest end. Across 411,770 Polymarket sports buys of $1,000 or more from April 2 to September 11, 2026, buys under 10 cents won 2.97% of the time at an average price of 6.49 cents and returned 39 cents on the dollar. Every bucket from 20 cents to 98 cents sat within a point and a half of its price, with confidence intervals that include zero.

### Are favorites a good bet on Polymarket sports markets?

Before fees, favorites at 80 cents and up returned between plus 0.4% and minus 0.6% on the dollar, and buys at 70 to 80 cents returned plus 3.6%. Favorites carry no premium and no penalty; a taker pays a fee of 0.25% to 1% of the stake at those prices, and a resting limit order pays none.

### Why does the study start on April 2, 2026?

Before that day the large-trade alert table recorded a defaulted outcome index for a large share of buys, so February and March data reads as favorites at 85 cents winning 59% of the time, which is a defect and not a market. From April 2 onward buys on either outcome win at their price.

### What does edge mean on this page?

The share of buys that won minus the average price paid, in percentage points. A bucket that paid 60 cents and won 61% has an edge of plus one point. It is the same measure as the calibration edge on every 0xinsider trader profile.

### If sports prices are calibrated, does anyone have an edge?

Calibration on average means winners and losers among large buyers net to zero, not that there are none. Splitting the same buys by the grade the wallet held on the day it traded, S, A and B wallets beat the price by 1.25 points from June to September 2026 while D and F wallets lost 1.21. That study is at 0xinsider.com/research/sharp-money-polymarket-sports.