U.S. Sports Betting Data

I've been tracking the regulated U.S. sports betting market since New Jersey took the first post-PASPA wager in June 2018. Eight years later, the market has processed more than $600 billion in cumulative legal wagers and shows no signs of slowing. The data below tells the story of who's winning, who's losing, and where the money actually goes.

Our team of journalists and analysts maintains this live database, built on state agency filings, in collaboration with independent analyst Alfonso Straffon, a longtime industry observer, former sports trader, and equities analyst at Deutsche Bank. We update on or about the third Thursday each month. You're seeing operator-level figures and state-level data through September 2026, which is the latest data available.

Citation request. If you reference any data from this page, please support our work and cite this page with a link. Spotted an error? Reach out to brett@thirdplanet.us.

1. Handle market share by operator (last four years)

Monthly mobile handle share outside Nevada, August 2022 through August 2026.

Key Insights:

• DraftKings reclaimed the handle crown from FanDuel in May 2025 and still holds it. DraftKings sits at 36.7% to FanDuel's 33.0% as of August. FanDuel's late-summer jump, its strongest reading since late 2025, cut the spread to 3.7 percentage points from nearly 5 points in our last update.
• The combined duopoly share sits at 69.7% of all dollars wagered in the regulated market, still below historical highs near 75%.
• Fanatics is now the clear No. 3 at 9.0% of handle. Its August 2025 spike to roughly 13% was promo-fueled and short-lived. Fanatics settled back near 7.5% that fall, then climbed past BetMGM in early 2026 and has held in the 9% to 10% range since.
• BetMGM has slid to 6.3%, the lowest reading in our four-year window and a long way from the 12% it held in mid-2022.
• Caesars (5.5%) moved back ahead of bet365 (3.8%) in August. bet365 had been running above Caesars for much of 2026 and dropped sharply from roughly 5.7% in July. It's one month of data, so treat it as a development to watch rather than a reversal.
• theScore Bet (formerly ESPN Bet) no longer appears in the tracked operator set. PENN Entertainment exercised its opt-out clause in November 2025, shuttering ESPN Bet and relaunching as theScore Bet on December 1, coinciding with Missouri's market opening. ESPN subsequently partnered with DraftKings for odds integration across its media properties. Rush Street Interactive (BetRivers) remains the smallest tracked operator at 1.6% handle share.

Note: Caesars' early-2022 surge, timed to New York's launch and the NFL playoffs, predates this chart's four-year window. It was a massive ad-and-bonus blitz, and the momentum didn't stick. Caesars has since held in the 5% to 8% range for sports betting and remains a significant player in the more lucrative iCasino sector. Its tier of the market is now a three-way scramble with BetMGM and bet365, all well behind Fanatics.

2. GGR market share by operator (last four years)

Monthly mobile gross gaming revenue share outside Nevada, August 2022 through August 2026.

Key Insights:

• FanDuel still leads on revenue, but by the thinnest margin in some time: 35.9% to DraftKings' 34.6%, a spread of just 1.3 percentage points, down from 2.9 points in our last update. DraftKings is winning on volume and has nearly closed the margin gap. It has edged past FanDuel on GGR in individual months before, including March and August 2025, and another flip wouldn't be surprising.
• Fanatics ranks third on revenue at 9.3%, slightly above its 9.0% handle share. Fanatics is now converting its volume into revenue at roughly a market-average rate, and it has outearned BetMGM in most months since late 2025.
• BetMGM dipped to 6.9% of GGR from about 7.7% in July. It is fourth on both handle and revenue.
• Caesars sits at 4.2% of GGR and bet365 at 3.4%, down from roughly 5% in July. Rush Street rounds out the tracked set at 1.8%.

Also note: Hard Rock Bet, backed by the Seminole Tribe, will use its Florida monopoly position to drive national expansion, but Florida figures aren't publicly reported. One more thing: the impact of prediction markets (Kalshi, Polymarket, plus DraftKings and FanDuel's own vehicles) is not visible in these figures. Analyst estimates from early 2026 peg sports betting cannibalization at 0-5%.

3. Handle by state, last twelve months

Trailing 12 months: September 2025 through August 2026.

Key Insights:

• $26.9 billion was wagered in New York over the trailing 12 months. The state's nine regulated sportsbooks have averaged about $2.24 billion per month, more than the entire LTM total in most other states.
• That level isn't surprising given New York's population and per capita income, concentrated around New York City, Long Island, and Westchester County.
• Illinois at $15.5 billion holds its lead over New Jersey ($11.9 billion) for the No. 2 spot.
• Ohio ($10.4 billion) and Arizona ($9.4 billion) round out the top five. Both states have benefited from broad mobile access and large, sports-hungry populations.
• Massachusetts ($8.65 billion) still edges Pennsylvania ($8.53 billion) for sixth, despite being a much younger market. Just behind them, Nevada ($7.97 billion), Virginia ($7.92 billion), and North Carolina ($7.81 billion) are separated by less than $160 million.
• Missouri has climbed to $2.6 billion, already ahead of Connecticut, in a market that only launched in December 2025.

Mississippi regulators and lawmakers have flirted with authorizing online sports betting repeatedly, most recently in January 2026, but it remains a casino-only state. If and when Texas and California join the regulated ranks (Las Vegas Sands is spending heavily to make it happen in Texas), the top of this chart will get interesting.

4. Gross revenue by state, last twelve months

Trailing 12 months: September 2025 through August 2026.

Key Insights:

• Operators collectively grossed $2.57 billion in New York over the LTM, but they also paid a $25 million up-front licensing fee and face a nation-leading 51% tax on GGR. Only Rhode Island, where the state lottery runs sports betting, takes a comparable cut.
• Do not conflate gross revenue with net revenue or net profit. These are top-line figures before taxes, operating costs, and interest payments.
• Illinois ($1.48 billion), New Jersey ($1.14 billion), and Ohio ($1.09 billion) join New York in the billion-dollar GGR club.
• The next tier is a pileup. Pennsylvania ($918.7 million), Massachusetts ($892.3 million), Arizona ($867.8 million), Virginia ($852.8 million), and North Carolina ($833.9 million) are separated by less than $85 million. Massachusetts' rapid ascent is notable for a market that only launched in early 2023.
• Missouri has reached $344.8 million in GGR, effectively tied with Kentucky ($345.1 million) and ahead of Iowa and Kansas, in less than a year of operation.

Illinois taxation. In June 2024, Gov. JB Pritzker signed a progressive tax scheme ranging from 20% to 40% of adjusted gross sports wagering receipts.

The 40% rate hits only DraftKings and FanDuel, which was the intent. Other operators, barring a revenue surge, won't reach the thresholds for higher brackets.

It got worse. In June 2025, the legislature added a per-wager tax: 25 cents for an operator's first 20 million wagers, 50 cents for wagers above that. On June 10, FanDuel announced a 50-cent transaction fee on all Illinois wagers in response, effective September 1, saying it would drop the fee if lawmakers killed the per-bet tax. Most other Illinois sportsbooks followed with their own per-bet fees or minimum bet amounts of $1 to $2. In the first three months, the per-bet tax generated over $21 million in additional revenue, nearly $16.5 million of it from FanDuel and DraftKings alone. But total bets placed dropped by more than five million year-over-year in September 2025.

5. Tax revenue by state, last twelve months

Trailing 12 months: September 2025 through August 2026.

Key Insights:

• New York collected $1.31 billion in sports betting tax revenue over the LTM, a staggering haul that exceeds what most states generate from their entire gaming sectors.
• New Jersey has pulled clear of Illinois for the No. 2 spot, at $729.8 million to Illinois' $619.1 million. The two were in a virtual tie in our last update. New Jersey raised its rate from 13% to 19.75% in 2025, and the current LTM window runs entirely at the new rate.
• Pennsylvania sits fourth at $239.2 million, followed by Ohio at $219.3 million. Ohio doubled its rate from 10% to 20% in June 2023; lawmakers contemplated raising it again but backed off.
• Massachusetts ($173.4 million), North Carolina ($153.3 million), and Maryland ($131.5 million) round out the top eight. North Carolina's position is impressive given how recently the online market launched.
• Pennsylvania's 36% rate, established at legalization, is among the steepest in the country. Its LTM tax take ticked down slightly from $242.7 million in our last update as gross revenue softened.
• Virginia ($116.6 million), Tennessee ($115.2 million), and Louisiana ($112.3 million) form a tight cluster below the top eight. Louisiana's LTM now reflects a full year at its higher 21.5% rate.

Tax rate changes in 2025. Illinois' shift to a progressive structure inspired a wave of increases across the country. Here's where things landed:

• New Jersey: Rate increased from 13% (sports) and 15% (iGaming) to 19.75% for both. Gov. Murphy had pushed for 25%. The current LTM window is the first to run entirely at the higher rate.
• Maryland: Rate rose from 15% to 20% under Gov. Moore's budget. He'd initially sought 32%.
• Louisiana: Gov. Landry signed a hike from 15% to 21.5%, effective August 1, 2025. Revenue funds a new college athletics (SPORT) fund.
• North Carolina: The Senate proposed doubling the rate from 18% to 36%, but the final budget passed without the increase. Rate remains 18%.
• Indiana and Ohio considered similar proposals. Ohio's was dropped from budget negotiations.
• Pennsylvania explored changes in November 2025 but backed off.

2024 background for posterity: DraftKings, during its Q3 2024 earnings, floated a plan to pass Illinois' elevated tax back to bettors as a surcharge on winning bets. The reaction was brutal. FanDuel (Flutter) said during its own earnings call two weeks later that it absolutely would not follow DraftKings' lead. DraftKings abandoned the idea, though CEO Jason Robins has said the company will consider any and all ideas to protect margins.

6. Parlay hold % versus non-parlay hold (last four years)

Monthly, August 2022 through August 2026. Four-state sample: NJ, CO, IL, MD.

Key Insights:

• Month-to-month volatility is significant, but the long-term trend for hold percentage at U.S. sportsbooks points upward. Multiple forces are at work: growing customer adoption of Same Game Parlays (SGPs), improvements in operator pricing algorithms, and aggressive product emphasis on higher-margin offerings.
• This chart reflects betting activity in New Jersey, Colorado, Illinois, and Maryland only, though it's a strong and representative sample given the overall volume in those states. Granular bet-type data reporting is better in some states than others. Illinois data runs through June and Colorado through July, so the most recent months reflect a partial sample.
• Summer 2026 produced mirror-image months. In June, non-parlay hold sank to 2.7%, the lowest in our dataset, while parlays held 14.4% and total hold came in at 7.4%. In July, the script flipped: non-parlay hold spiked to 8.2%, the highest in our dataset, pushing total hold to 11.3%. August settled in between, with parlays at 15.8%, total hold at 8.3%, and non-parlay hold at 4.6%.
• Don't read too much into a single month. Parlay hold has peaked near 24% in prior cycles (mid-2023, fall 2024, mid-2025) and has dipped to around 10% during stretches of customer-friendly football results, most notably in late 2023 and the 2024-25 season. It tends to recover heading into the heart of football season.
• Parlay and non-parlay hold percentages are generally correlated. When one rises or falls, the other tends to follow. June and July 2026 were notable exceptions.

7. Parlay revenue as a % of total revenue (last four years)

Monthly, August 2022 through August 2026. Four-state sample: NJ, CO, IL, MD.

Key Insights:

• Parlays as a share of operator revenue hit a new all-time high of approximately 78% in June 2026, revised up from the initial reading as late-reporting states filed. That shattered the prior peak of about 69%, reached in September 2024 and again in November 2025.
• The swing back was just as dramatic. With non-parlay hold spiking in July, parlays' share of revenue fell to roughly 52% before rebounding to about 63% in August, right around the long-term trendline.
• The chalk-heavy NCAA tournament in March 2025 knocked parlay revenue below 50% of total operator revenue (47%), something that had only happened three other times since January 2023. It bounced back quickly, and then some.
• The trendline across the full four-year period is pretty apparent. Parlays have gone from generating roughly half of operator revenue to a trend level near 65%, with surges now pushing past 75%.
• How high can this go? Probably limited only by product ingenuity on the operator side, tempered by the risk of burning out bankrolls too fast. The June reading suggests we may not be near the ceiling yet.

8. Parlay handle as % of total handle (last four years)

Monthly, August 2022 through August 2026. Four-state sample: NJ, CO, IL, MD.

Key Insights:

• Parlays set a new handle share record of 40.0% in June 2026, eclipsing the prior high of roughly 35%, reached in fall 2025 and again in May 2026. For that month, four of every ten dollars wagered went on parlays.
• July and August came back to earth at about 33%, right on the long-term trendline.
• Over four years, this figure has grown from the low 20s to a trend level in the low-to-mid 30s. The trendline points clearly upward and shows no sign of plateauing.
• The ceiling isn't limitless. As long as VIPs continue placing large straight wagers on football and basketball spreads and totals, non-parlay volume will remain significant. But recreational bettors, who are the market's growth engine, skew heavily toward parlays, and their preferences are increasingly defining the market's shape.

9. National quarterly hold percentage (Q1 2021 to present)

Quarterly, Q1 2021 through Q3 2026 (quarter-to-date through August).

Key Insights:

• Q3 2026 (quarter-to-date) is tracking at 10.6%, cooling from the 11.6% pace we flagged in our last update. That's still a strong number, but it's no longer on track for a record. The all-time high remains 10.9%, set in Q2 2025. September brings the start of the NFL season, and early results will decide where the quarter lands.
• Q2 2026 finalized at 10.1% (revised from 10.2%), following 9.8% in Q1 2026.
• Q4 2025 hit 10.8%, a big finish. Compare that to Q4 2024's 7.9%, which was dragged down by "customer-friendly" NFL results, meaning lots of favorites covering. Booking sports can be volatile, even in the era of parlays.
• The quarterly hold has now hit or exceeded 10% six times on a finalized basis: Q3 2022 (10.6%), Q2 2024 (10.0%), Q3 2024 (10.5%), Q2 2025 (10.9%), Q4 2025 (10.8%), and Q2 2026 (10.1%). Q3 2026, at 10.6% to date, is positioned to make it seven. If you're betting on where this number trends, the over on 10% for future quarters looks right.
• For perspective: pre-PASPA, when Nevada was the primary regulated market and parlays got far less emphasis, the national hold lived closer to 5%. The structural shift is real and probably permanent.

10. National annual hold percentage (2018 to present)

Annual, 2018 through 2026 year-to-date (through August).

Key Insights:

• 2025 was the first calendar year in which the national hold exceeded 10%. That's a milestone worth pausing on. The full-year print of 10.1% means operators kept more than a dime of every dollar wagered across the entire regulated market, on average, for 12 straight months.
• 2026 year-to-date is tracking at 10.1% through August, suggesting this isn't a one-year anomaly. The floor has moved.
• The annual view smooths out the quarterly noise and makes the long-term arc unmistakable: from 7.2% in 2020 (the trough, driven partly by COVID-era dynamics and a thinner parlay mix) to 10.1% today, a nearly 3-percentage-point climb in five years.
• The 2018 reading of 8.8% is an outlier in the other direction. With only a handful of states live and Nevada dominating the sample, that number reflects a different market entirely. The post-2020 trajectory is the one that matters for projecting forward.
• For bettors, the implication is stark: the house edge is structurally higher than it was when legal sports betting launched. Parlays, SGPs, and improved pricing algorithms have made the recreational bettor's dollar go shorter. We couldn't confirm whether any operator has publicly acknowledged this trend as a long-term strategic objective, but the data doesn't leave much room for alternative explanations.

What's next

Visit our companion database for U.S. Online Casino Data: Market Share By Brand, Gross Gaming Revenue Stats, and Taxation.

Selected data sources

• Illinois Gaming Board
• Michigan Gaming Control Board
• New Jersey Division of Gaming Enforcement
• New York State Gaming Commission
• Pennsylvania Gaming Control Board

Source: State filings. Charts and estimates by @astraffon. Monthly shares subject to revision as states report.