Every few weeks someone asks me where they can bet on esports in the US without a VPN and a prayer. The honest version is that esports betting here has quietly grown a second home, and it does not look like a sportsbook. Kalshi, a CFTC-regulated exchange, lists Counter-Strike 2, League of Legends, Dota 2, Valorant, Rainbow Six Siege and, when that circuit is running, Call of Duty matches as yes/no contracts priced in cents, across more than 100 esports market series. The board is deeper than most people searching for esports odds realize, and the single most interesting thing about it is not a price. The interesting thing is a closing mechanic. These markets can close the moment a winner is declared instead of sitting on a clock, which makes them behave differently from any bet slip you have ever held. That mechanic is where this page starts, because it explains almost everything else about how the board works, including why a tournament favorite's contract settled at zero three weeks before I wrote this. That story comes later.
The Quick Answer
Esports betting on a regulated exchange means buying yes/no contracts on match winners, map winners and total maps across CS2, LoL, Dota 2 and Valorant, plus Call of Duty and Rocket League when those circuits are running, with match and map markets on Rainbow Six, priced between 1 and 99 cents. The match-winner question is built from exactly two markets, one per team, and in a match that gets played, exactly one of those two settles at $1. The full list of what is traded, how a match market is built, how deep the board really is, and the early-close mechanic that separates this from a sportsbook are all below.
Stay ahead of the markets.
Daily insights and expert picks on Kalshi, Polymarket, and what's moving markets.
Free forever. Unsubscribe anytime.
The Mechanic To Understand First: These Markets Can Close The Moment A Winner Is Declared
Every esports match market on the Counter-Strike page I pulled from the exchange carries a field in its market data called can_close_early, and for these boards it is set to true. That is permission, not a promise: it means the market can close before its scheduled close time once the outcome is known, instead of running out a clock nobody needs. After the close, determination and settlement follow the exchange's normal process. A contract on the winning team pays $1, the other side goes to zero.
The rule text is shorter than most bet-slip terms. On a Counter-Strike qualifier match listed the morning I checked, the core rule reads: "If Finishers wins the PGL Masters Bucharest: Oceanic Qualifier 2026: Finishers vs. FlyQuest CS2 match originally scheduled for Sep 10, 2026 at 5:00 AM EDT, then the market resolves to Yes." Team wins, contract resolves Yes. No grading queue, no "pending" tab. There is a second block underneath it for the messy cases, and esports has plenty of them: a match postponed and not started within 48 hours, cancelled before play, or forfeited before any play occurs resolves to the fair market price rather than to a team, and a match that starts and is then forfeited resolves to the organizer's official result.
Compare that to a sportsbook. There, your wager is locked the moment you place it. You can sometimes cash out at whatever number the book offers you, but the position itself is frozen until the game resolves on the book's schedule. On the exchange, the contract is a live instrument the whole way through: you can buy before the match, sell during it, add during a map, or exit entirely, right up until the moment the winner is declared and the market snaps shut.
It also explains why the book looks so different before and during a match. Pre-match, prices sit wide apart and drift as opinions form. Once the series is live, every round and every objective is public information arriving in real time, so the trading tightens and the price starts converging toward $1 or zero. Late in a one-sided series the headline number stops being the interesting part, and the spread, the fee and the size actually resting at your price matter far more. This is not a strategy and not a signal. That is simply how a continuously traded market on a fast-resolving event behaves, and nobody tells you before your first trade.
The one sentence to remember: an exchange esports position is not frozen the way a bet slip is. These boards trade live, with open matches quoting live bids and asks, so while the market is open you can exit at whatever is actually bid instead of holding to settlement, right up until the winner is declared. On a headline match that bid sits close to the price. On a fresh qualifier it can be a fraction of it.
How A Match Market Is Built: Two Markets, One Winner
Understanding that mechanic is easier once you see how small the building blocks are. For a standard two-team match, the winner question on the exchange is not one market with two sides. It is two markets, one per team, grouped as a single mutually exclusive event. Pull up a live one and you can see it: the Counter-Strike event KXCS2GAME-26SEP100500FINFLY, listed on the board at 2:50 AM Eastern on September 10, 2026 for a 5:00 AM match, contains exactly two markets. One asks whether Finishers wins, one asks whether FlyQuest wins. Once the match is played, exactly one of them settles yes; the postponement and forfeit exceptions are in the rule text quoted above.
Each contract trades independently between 1 and 99 cents, and the price is a rough implied probability before you account for the spread, the fee and the size actually available: a team trading at 62 cents is being priced at roughly a 62% chance, which in American odds terms works out to about -163. Its opponent will usually be trading somewhere near the mirror image, around 38 cents (a 38% chance, or roughly +163), because if the pair could be bought for meaningfully less than a dollar, traders would simply buy both sides. A 90-cent contract reads as a 90% chance, an 8-cent long shot as 8%, a coin flip as 50%. You can take either side of either market, which means you can effectively back a team or fade it without any spread-betting vocabulary in the way.
A real pair from that same morning shows the mirror in practice. In the CCT Europe match between UNiTY esports and Leo Team, the Leo Team contract had last traded at 53 cents on about 2,300 contracts of volume, and the UNiTY contract at 45 cents on about 880. Fifty-three and 45 sum to 98, which is the mirror doing its job. But look at where the actual quotes sat: 30 bid against 69 ask on Leo Team, 50 bid against 83 ask on UNiTY. Buying both sides at the offer that morning would have cost $1.52, not 98 cents. The last trade is where the market was. The ask is what you pay.
On top of the match winner, the bigger titles carry two more market classes:
- Map Winner Markets ask who takes an individual map or game inside the series.
- Total Maps Markets ask about the number of maps played, for example whether a best-of-three goes the distance.
The same mutually exclusive team structure repeats for map winners. Total-maps markets work differently. Those are separate binary contracts on the length of the series rather than on a team. Between the match winner, two markets per map and the total, a best-of-three can carry seven or more separate contracts before a single round is played.
What Is Actually Listed: The Games On The Board
The exchange's esports section is broader than the six headline titles, and this is the part that surprises people who assume the board is a Counter-Strike novelty. Re-checked directly against the exchange's public data as of September 10, 2026, the board spans 108 esports-tagged market series, though roughly two dozen of those are duplicate tickers, dead shelves from the CS:GO era, or in one case a series titled TESTING. Every one of them sits on the same quadratic fee schedule, and the live ones cover, among others:
| Game | Market types listed (each row traced to its own series on the board) |
|---|---|
| Counter-Strike 2 | Match winner, map winner, total maps, tournament winner, qualifier |
| League Of Legends | Match winner, map winner, total maps, tournament winner |
| Dota 2 | Match winner, map winner, total maps, tournament winner, plus most-banned and most-picked hero |
| Valorant | Match winner, map winner, total maps, tournament winner |
| Call Of Duty | Match winner, map winner, total maps, tournament winner; nothing open on the September 10 check |
| Rocket League | Match winner, game winner (its "maps" are games), total games, tournament winner; no open match markets on the September 10 check |
| Rainbow Six Siege | Match winner, map winner (map book empty on the September 10 check), tournament winner (no total-maps series) |
| Overwatch | Match winner, tournament winner |
| PUBG, Brawl Stars, Crossfire | Tournament winner only |
| StarCraft II, Apex Legends, Honor Of Kings, Mobile Legends, Teamfight Tactics, Free Fire | Tournament markets only, listed as one-offs for the 2025 Esports World Cup and since settled |
Two rows there are worth reading slowly. The last one, because "listed" and "open right now" are not the same sentence: the exchange stood up two dozen one-off series for the 2025 Esports World Cup, and for several of the smaller titles that event is the only reason they ever appeared on the board at all. All of those series are done, one CS2 shelf under the event's name never listed a market at all, and on the September 10 check none of them carried an open market. Here is the trap I fell into myself on an earlier version of this page: the 2026 edition of the Esports World Cup never got its own dated series. Its Counter-Strike winner market lived inside the regular CS2 tournament-winner series, as one event with 28 team contracts. If you had searched the exchange by the event name, you would have found last year's dead shelf; if you had searched by the game's outright series, you would have found it, as event KXCS2-EWC26 under the CS2 tournament-winner series. That event has since settled, and the lesson holds for whenever the next edition lists. Search by game, not by event.
Then the Counter-Strike row, at the opposite extreme: its map market finishes more contracts than any other esports series I counted, and on that same September 10 check it was carrying 192 open match markets on its own, plus 272 open map markets, 68 total-maps lines and a 24-contract tournament-winner book split across two events, FISSURE Playground 3 and StarLadder StarSeries Fall. League of Legends was next, with 58 open match markets, 160 map markets and 51 totals. Rainbow Six, which showed nothing open when I checked on August 18, had 54 open match markets this time; Dota 2 carried 28 match and 56 map markets; Valorant 14 and 32; Overwatch eight. Rocket League's match market and Call of Duty's match and map markets had nothing open at all. Open counts reshuffle daily as brackets start and finish, so read them as a snapshot of the board's shape, not a standing total. The practical board a US trader can act on is narrower and far more concentrated than the full series list implies: Counter-Strike and League of Legends carry most of it, and everything else comes and goes with the calendar. There is also a small awards shelf most people never scroll to: HLTV Player and Team of the Year, and one-off MVP contracts on named players. The shelf even stretches to chess and GeoGuessr world championship winners. For the map-by-map layout of that board, the Counter-Strike board has its own full breakdown, and the Dota 2 board covers how the Dota version differs. When a major bracket is live, the tournament-winner futures work the way our coverage of the BLAST Bounty winner market walked through: one yes/no contract per team, the losers closing as they are eliminated and the champion's on official determination.
How Deep Are These Markets?
Depth is the real question behind "where to bet on esports," because a listed market you cannot get filled in is a screenshot, not a venue. There is a way to measure it, and there is a trap in how you read it. The exchange's public feed does not hand you an all-time settled count: on my September 10, 2026 pull, the oldest finished esports market it would return had closed on July 3, and results older than that were simply gone from the feed. Ask it again in a month and the back edge will have moved with you. So the number below is not a lifetime total and it does not only go up. Count the same series twice, weeks apart, and it can come back smaller. It is something more useful than a total anyway: a throughput reading, the rate at which these boards actually finish contracts.
Counted directly from that feed on September 10, 2026, covering the ten weeks of results still in the window:
| Series | Settled in the window (July 3 to September 10, 2026) |
|---|---|
| CS2 Map Winner | 8,074 |
| CS2 Match Winner | 5,049 |
| LoL Map Winner | 3,838 |
| Valorant Map Winner | 2,172 |
| LoL Match Winner | 2,135 |
| CS2 Total Maps | 2,017 |
| Dota 2 Map Winner | 1,582 |
| Valorant Match Winner | 1,032 |
| Dota 2 Match Winner | 965 |
| Call Of Duty Map Winner | 292 |
| Rainbow Six Match Winner | 280 |
Those 11 series finished roughly 27,400 yes/no results in ten weeks, which is the difference between a product and an experiment. The row that matters most is the first one: Counter-Strike's map market settles more contracts than any other series I counted, which tells you where the trading habit has actually formed. That is why a CS2 map market at a major is a very different proposition from a Rainbow Six market on a quiet Tuesday. The Call of Duty row is the other one to notice: its last settled map market closed on August 9, 2026, so that board has been dark for a month while the series still sits in the catalog. Listed is not live.
The feed has also started telling you something it would not tell me a month ago: per-market volume. Of the 5,049 CS2 match markets that settled in the window, 4,966 had traded at least once, about 98%; on the map side it was about 95%. That is the settled book, though, and the open book looks thinner. On the first page of open CS2 match markets that morning, 50 contracts across 25 matches, only 13 had traded at all, and the untraded ones were quoting spreads as wide as a 12-cent bid against an 88-cent ask, the most common shape on that page, and a few wider still: the Finishers and FlyQuest market I opened this page with sat at a 3-cent bid against a 97-cent ask. That is not a price. It is a placeholder waiting for the first trader to show up, and if you lift an 88 on one of those you have paid 88 cents for a team the same book bids 12 for. So treat depth as something you check on the specific market in front of you, not something you assume from a table. The order book itself is the truth: before you commit, look at how much is resting at the price you want, exactly as our guide to prediction market liquidity lays out. A headline match at a major will trade very differently from a qualifier between two teams you have never heard of.
What A Trade Costs, And How It Pays
All 108 esports series on the board use the exchange's quadratic fee formula, which means the trading fee is largest for contracts priced near 50 cents and shrinks toward the extremes. Buying a coin-flip match at 50 cents costs more in fees than buying a heavy favorite at 90 cents. The full math, with worked numbers, is in our guide to how the fees work.
Settlement is the clean part. When the winner is declared, the market closes rather than waiting out a scheduled clock, and settles from there: yes-holders on the winning team receive $1 per contract, and the losing side's contracts expire at zero. You are not waiting on a book's grading schedule, though determination is still a step, and it can occasionally lag or be contested. The process is the same one every exchange contract follows, described in what actually happens when a contract settles. The early-close flag from the top of this page is doing the work here: the declared winner is the settlement trigger, not a clock.
Now the story I promised. Our own Esports World Cup CS2 bracket page had Vitality as the favorite on August 18, bid 22 cents and offered at 23, an implied 22% to 23% chance, with Spirit third at 19 cents. Vitality's contract closed on August 21 and settled no. Spirit's closed on August 24 and settled yes, at $1, on about 128,000 contracts of volume, the busiest contract in that event by a wide margin. Nothing about that is a scandal. A 22-cent favorite is supposed to lose about 78 times in a hundred. Consider it the cleanest recent example of what "settles at zero" means with a real name on it, and of how fast it happens: the favorite's market closed on August 21 and was settled five minutes later.
One risk shape deserves its own sentence, because it is the mistake I care most about a new trader making. Selling an unlikely outcome collects a small premium while risking most of a dollar; roughly speaking, at equal size, one loss erases the premiums from about 11 wins. That arithmetic, not any hit rate, is what makes position sizing the whole game. When you sell a long shot, one loss costs many wins covers it in full.
A Worked Example: One Contract, Start To Finish
To make the mechanics concrete, walk a real contract through its whole life, using the tournament board from the story above. On August 18, 2026, the Vitality contract in the Esports World Cup CS2 winner market was bid at 22 cents and offered at 23. A trader who liked the favorite and lifted the offer paid $23 for 100 contracts, a position the market was pricing as a 23% chance, with the trading fee taken at the fill rather than skimmed off any payout later. Three days later Vitality was eliminated. The market closed at 4:32 PM Eastern on August 21 and settled five minutes later, and those 100 contracts expired at zero. The full $23 was gone, and nobody had to wait for a grading department to say so.
Now the other side of the same board. Spirit's last trade that same August 18 was 19 cents; call it about $19 for 100 contracts, with the real cost set by wherever the offer sat. The final was Sunday, August 23. Spirit's contract closed the next morning, 10:45 Eastern on August 24, once the exchange had its official determination, a reminder that "can close early" is permission tied to determination, not to the moment you see the trophy lift. It settled at $1 per contract: $100 back on roughly $19 out, on the busiest contract in the event at about 128,000 contracts traded. Same event, same rules, two prices, two endings. That is the arithmetic every market on this board runs, favorite or long shot.
The part a sportsbook never offers sits between those two endings. Had a holder of either contract wanted out while the bracket was still running, they could have sold into whatever bid was resting at that moment, banking a gain or cutting a loss before the result existed. On a busy outright like that one the bid is usually close to the last price; on a thin qualifier match it is the 12-against-88 placeholder from the depth section. Either way, the exit is a trade against another person, not a cash-out offer from the house, which is why the settlement process is worth understanding before your first order, not after.
Where To Bet On Esports, Legally
For a US reader, this is the practical difference. Traditional esports betting sits inside state sportsbook licensing, and whether a given state's books carry esports at all is a state-by-state, book-by-book question. Indiana, for one, has had licensed sportsbooks since 2019, and its sports wagering law names esports, alongside amateur youth events, as something those books cannot take a bet on. The exchange described on this page is Kalshi, and its esports contracts are CFTC-regulated event derivatives, available where the exchange operates, under federal oversight. You must be 18 or older to open an exchange account, a lower bar than the 21 sportsbooks require in most states, and the details are in our guide to the age requirement. Same event, two different regulatory doors.
If you got here by searching for esports betting sites, the honest answer is narrower than the results page suggests: a state-licensed sportsbook that happens to carry esports where you live, or the exchange. I am not going to pretend a third door exists.
Two things I will not do on this page. I will not give you an esports pick, and I will not quote anyone's record. We do not publish esports picks. The exchange boards where we do publish model verdicts are graded publicly against settlement on our scoreboard, every call and every miss, and that is the bar any esports tout should have to clear before you pay attention to a screenshot.
For transparency: we have no affiliate or commercial relationship with Kalshi, the exchange whose esports board this page describes, so nothing here is a paid placement. We do carry sign-up offers for some other prediction-market and betting platforms, and we say so wherever those platforms come up. Stokastic Inc., our parent company, trades on prediction markets and holds positions in them.
More on this: Esports World Cup CS2 Bracket: Kalshi Favors Vitality · Barcelona Vs Feyenoord Odds: Does The Favorite Survive The Fee? · Where Does Bitcoin End 2026? Kalshi's Price-Band Odds · The Longest Shot On This Board Might Be The Best-Kept Secret · 2026 Governor Races: Every Kalshi Board We Track
What To Check Before You Touch One Of These
Nothing above is a pick. I have not told you which team wins anything and I am not going to, because the point of this page is the machine, not a play. But if you came here asking how to bet on esports, the machine has a short list of things worth confirming before you click, and every one of them came up somewhere above.
Start with what you are actually looking at, because the board makes it easy to misread: a "CS2" ticker can be a match winner, a map winner, a total-maps line or an outright, and those four resolve on completely different events. The title should be genuinely open, not a leftover from a tournament that has already settled, and a live tournament's winner market probably lives under the game's outright series, not under the event's name. Resting size matters twice, at the price you want to enter and the price you would want to exit at, because your exit is whatever is actually bid when you want out; the Finishers and FlyQuest book sat at 3 against 97 the morning I pulled it. Where your price sits relative to 50 cents decides how hard the quadratic fee bites. And if you are selling a long shot rather than buying one, the sizing arithmetic comes before the thesis: 11 wins, one loss, flat.
That last one is the whole reason the mechanic at the top of this page matters in both directions. Ask anyone who paid 23 for Vitality.
If you also bet the traditional sportsbook side, that is the half of this we cover daily, with free expert picks today across the major sports. The deeper OddsShopper toolset runs on sportsbook markets, and you can try all of it with a free week of OddsShopper Pro.
Esports contracts, like every event contract, can lose their full value, and the early-close mechanic that opened this page cuts both ways: the market that lets you exit at any moment is also the market that will not wait for you once a winner is declared. Trade a size where that sentence is boring. For the same close-versus-settle mechanic in a very different category, start with how a weather market's two clocks work.
18+. Available where the exchange operates; eligibility restrictions apply. Event contracts are CFTC-regulated derivatives, not sportsbook wagers, and can lose their full value. Nothing on this page is trading advice.
Prices on this page are Kalshi's book. If you also trade on Polymarket, code OS4 gets new users a $50 trading bonus on a $10 deposit — affiliate link; terms as stated by Polymarket; 18+, availability varies by state.
