About one match in four. In our own measurement of 676 completed three-way soccer markets on Polymarket US, taken 2026-08-08 by our trading research, 24.56% of games ended in a draw, with a 95% confidence interval of [21.25, 27.86]. The interval is not decoration. A sample of 676 games is enough to say "roughly a quarter" with confidence and not one decimal more, and any page quoting a bare draw percentage without an interval is claiming a precision its sample cannot pay for. This page does three jobs in order: it shows you exactly what that number is and where it came from, it fences off what the number does not cover, and it explains the quietly interesting part, which is why a prediction market is the cleanest instrument anyone has for measuring the draw at all. That last part is a structural story about how these markets price soccer's third outcome, and it is the reason this measurement exists in the first place.
The Quick Answer
In our sample of 676 completed three-way soccer markets on Polymarket US, 24.56% of games ended in a draw. The 95% confidence interval runs from 21.25% to 27.86%, so call it "about one in four" rather than a fixed constant. The figure is venue-specific and not league-weighted, which is why it can differ from a number computed on, say, a single European league. The full provenance, the sample's honest limits, and why a three-way market makes the draw directly readable are all below.
The Number, And Exactly Where It Came From
Numbers about draw rates float around the internet with no parentage at all, so here is this one's, in full. Stokastic trades prediction markets and holds positions in them, and as part of that trading research we measured every completed three-way soccer market on Polymarket US in our observation window: 676 games, each with a home-win, away-win, and draw contract that settled. Of those 676, the draw contract settled as the winner 24.56% of the time. At that sample size the 95% confidence interval runs from 21.25% to 27.86%. The honest reading is "very likely somewhere from just over a fifth to a bit over a quarter," centered near one in four.
The measurement in one line: 676 completed three-way soccer markets on Polymarket US → draws won 24.56% of them → 95% confidence interval [21.25%, 27.86%] → "about one in four," not a constant.
One more line, because it is the quality check that makes the rest worth quoting: across those markets, exactly one YES contract settled per game, a YES-per-game rate of 1.0000. Every match resolved to exactly one of home, away, or draw, with no gaps and no double-counting. That is a structural audit of the dataset rather than a claim about prices. It says the three outcomes account for the whole event, so the draw share is measured against a complete accounting of what happened.
What This Sample Is, And What It Is Not
Now the fence, stated in the same breath as the number. Those 676 games are one venue's board over a limited window, and a venue's board is whatever happens to be listed on it, not a tidy cross-section of world soccer. Concretely, the sample is:
- Not League-Weighted. No adjustment for which leagues supplied the games; the mix follows the venue's listings, not global fixture volume.
- Not Competition-Weighted. League fixtures, where the draw stands as a final result, dominate; it says nothing about knockout formats.
- Not Era-Adjusted. It is one recent window, not a historical average.
It is a real measurement of a real board, which is more than most quoted draw rates can say, but it is not "soccer" in the abstract. The distinction matters because the underlying draw rate differs by league and by era. A league full of cautious, evenly matched sides settles level more often than one dominated by two superclubs; points incentives shift it; scoring environments drift across decades. So if you have seen a draw figure computed only on a top European league and it disagrees with ours by a couple of points, neither number is wrong. They measure different populations of matches. Treat 24.56% as what it is: the measured draw share of one large, complete, verifiable sample, carrying its own error bars. Which raises the better question: why is this measurable here at all, when it is so awkward to read anywhere else?
Why A Three-Way Market Is The Cleanest Ruler
Here is the structural story promised at the top. Soccer is unusual among major sports in that the draw is a full citizen: a three-way market lists home, away, and draw as separate outcomes. On a prediction market, that third outcome is its own separately priced contract (in the market data it is literally a designated drawable outcome) that settles to $1 if the match ends level and $0 if it does not. Two things follow that no sportsbook line gives you.
First, the draw's implied probability becomes directly readable. A prediction market price is a probability in cents: a draw contract trading at 26¢ is the market saying 26%, no conversion required. On a sportsbook, the draw's probability is buried inside a three-way moneyline quoted at American odds with the book's margin folded in, and you have to translate the price format and strip the margin before you can even see the number. Second, settlement is binary and auditable, since each contract resolves YES or NO on its own; that is what let us run the one-YES-per-game check above. An exchange is a different instrument from a sportsbook, and this is the difference doing useful work: the exchange's format happens to be a measuring device.
A Worked Example: Reading The Draw Both Ways
Take round, illustrative numbers, not live prices from any market. A sportsbook quotes a three-way line: home −105, draw +260, away +310. Converting to implied probabilities gives 51.2% home, 27.8% draw, and 24.4% away. Those sum to 103.4%, not 100%; the extra 3.4 points are the book's margin spread across the outcomes. To find the fair draw probability you divide its implied share by the total: 27.8 ÷ 103.4 ≈ 26.9%.
| Outcome | Sportsbook line | Implied % | Fair % (implied ÷ 103.4) |
|---|---|---|---|
| Home | −105 | 51.2% | 49.5% |
| Draw | +260 | 27.8% | 26.9% |
| Away | +310 | 24.4% | 23.6% |
On a three-way prediction market, the same information arrives already unpacked: a draw contract at 27¢ is 27%, and the three contracts sum to roughly a dollar. The row worth staring at is the draw's. Two arithmetic steps on one side, zero on the other, and that gap is the whole reason a base rate like ours can be computed cleanly from exchange data: every game contributes one directly priced, binary-settled draw probability, and the settlement record grades it. If you want the sportsbook side of that comparison in the wild, our live odds screen carries three-way soccer moneylines across every major book, and running the arithmetic above against a real board is the fastest way to make it stick.
What This Number Does Not Tell You
This is the section that earns the page its keep, so it gets said plainly: 24.56% is a base rate, not a trade. Nothing in our measurement tests whether draw contracts are priced well or badly. We did not compare settled draw rates against the prices those contracts traded at, and without that comparison there is no statement to make about pricing, in either direction. It would be easy, and wrong, to leap from "draws happen about a quarter of the time" to some actionable conclusion about buying or selling them. The leap is exactly the part that has not been done.
The hard line: a base rate tells you how often draws happen. It does not tell you whether any current price is right, and we have not tested that question. Nothing on this page recommends buying or selling anything.
The honest reason sits right in the sample description: the board our history covers is not the board trading now. At the time of our measurement, the venue's live listings had rotated to obscure summer-season leagues — Colombian, Icelandic, Estonian, Norwegian — that the settled 676-game history does not cover. A base rate measured on one population of matches says nothing tested about how a different population is being priced. Whether these markets price the draw accurately is simply unknown to us, and we would rather hand you a clean base rate with its interval than a dressed-up guess. A shape-of-risk point belongs in the same paragraph, because it is the failure mode we care most about. In event markets, selling an unlikely outcome collects a small premium while risking most of a dollar; roughly speaking, one loss erases the premiums from about 14 wins. That asymmetry, not any hit rate, is why position sizing decides everything for people who trade these markets, and it is why a base rate alone, this one included, is nowhere near enough to act on. We say this as participants rather than spectators: Stokastic trades these markets and holds positions in them, and that discipline is documented in our running log of how these markets work.
Reading Draw Prices Yourself
If the measurement side interests you, the reading side is learnable in an afternoon, and this article's own number is the tool for it. Polymarket US is the CFTC-regulated venue American users fund through the Polymarket US App, with broad, state-specific availability under federal oversight. Every listed soccer match there carries the three contracts. Pull up any live draw contract and place its price against the interval this sample supports for a generic fixture: our settled draw rate's 95% interval ran from about 21% to 28%, and on these markets a price in cents reads directly as a probability. A draw quoted far outside that range is not a signal of anything — our measurement never tested prices, and thin books drift on small money — but it does tell you the market judges that particular match very unlike the average fixture, and asking why (a heavy favorite? a must-win with no incentive to settle level?) is exactly the reading skill these markets teach. Context for that exercise: how Polymarket works covers the machinery from funding to settlement, liquidity decides how seriously to take a quote, and trading costs sit between the posted price and what you keep. For the standing question of whether the venue itself is trustworthy, our read on whether Polymarket is legit covers the regulatory picture, and the deposits and withdrawals guide covers which Polymarket you are actually using before a dollar moves.
If you decide to open an account, code OS4 gets new users a $20 trading bonus after a $10 deposit. That is an affiliate arrangement, and we may earn a commission if you sign up through it. The offer's exact terms: Deposit $10, get a $20 trading bonus. 18+ Only. Restrictions and eligibility requirements apply. Not available in all jurisdictions. Trading is risky. 100% loss can occur. See polymarket.com/tos for more information. The Polymarket US App serves as an independent software provider and affiliate of Polymarket US and Polymarket Clearing, the CFTC-regulated exchange and clearing organization.
Soccer Draw FAQ
What percentage of soccer games are draws? In our measured sample of 676 completed three-way markets on Polymarket US, 24.56% of games ended in a draw, with a 95% confidence interval of 21.25% to 27.86%. It is a venue-specific figure rather than a league-weighted average; individual leagues run higher or lower.
How often are soccer games tied at full time? About one in four, by our measurement. Note that "tied" here means level after regulation in a format where the draw stands as a result. In knockout competitions that go to extra time and penalties, a match can be level at 90 minutes without officially ending as a draw, which is one more reason quoted draw rates differ across sources.
Why do soccer draw odds vary so much between leagues? Because the underlying draw rate is not one number. League balance, scoring environment, and incentives all shift how often matches end level, and prices follow the population of matches being priced. A figure computed on one league is a statement about that league, which is exactly why we state our sample instead of calling it "soccer."
What is a three-way soccer market? A market with three mutually exclusive outcomes (home win, away win, draw) where each outcome is its own contract on a prediction market. Exactly one settles YES per game; in our 676-game sample that accounting came out to precisely 1.0000 YES per game, the structural check that makes the draw rate measurable.
Does 24.56% mean draw contracts are a good buy? No, and this page makes no such claim. A base rate says how often draws happen, not whether current prices reflect that well or badly. Our measurement never tested settled outcomes against traded prices, so pricing accuracy is an open question, untested in both directions.
The Third Outcome, Counted Honestly
Zoom back out. For as long as sportsbooks have quoted soccer, the draw's probability has arrived buried inside a margin-laden three-way line, two arithmetic steps from readable. A prediction market's one structural gift is that it prices the draw as its own contract, which turns "how often do soccer games end in a draw?" from folklore into a measurement: 24.56% with an interval around it, on a sample you can audit down to the last settled game. That is the standard this page holds any draw number to, including its own. If reading markets this closely appeals to you, our analysts publish free expert picks every day, open with no account. And the fuller toolkit, the same fair-price arithmetic this page ran by hand but live across every major sportsbook, comes with a free week of OddsShopper Pro to try. Code DRAWRATE20 takes 20% off your first payment of OS Pro or OS Core if you stay.
Disclosure and fine print. Stokastic trades prediction markets and holds positions in them, including the Kalshi weather markets covered in our public research log. We have no affiliate or commercial relationship with Kalshi; we do carry sign-up offers for some other prediction-market and betting platforms, including Polymarket US, and pages naming them should be read with that incentive in mind. Event contracts on CFTC-regulated exchanges are derivatives, not sportsbook wagers, and a position can lose its full value; selling an unlikely outcome collects a small premium and risks most of a dollar, and one losing sale can hand back the premiums from about 14 winning ones. 18+, available where the exchanges operate; eligibility and availability are state-specific and change. The draw-rate figure on this page is a measured base rate with a stated confidence interval, not a pricing judgment. Nothing on this page is trading advice, a pick, a play, or a recommendation to enter any market, and no price shown here is a live quote.
