Nevada has never published a number for airport slot machines
Harry Reid International is the most-cited slot floor in America that isn't a casino. Every article about it quotes a payback figure. We went to the regulator's own filing to find the source of that figure — and there isn't one.
The Nevada Gaming Control Board publishes a monthly revenue report. It is the document every
serious claim about Nevada slot performance rests on. We downloaded the May 2026 edition —
48 pages, 139,588 characters — and ran a full-text search for airport.
The report is organised into reporting areas. There are 25 of them, and you can read the full list yourself in the table of contents. Statewide. Carson Valley. Churchill County. Clark County, then Clark County broken into Downtown Las Vegas, the Las Vegas Strip, North Las Vegas, Laughlin, Mesquite, Boulder, and Balance of County. Douglas County. Elko, with Wendover split out. Humboldt. Lyon. Nye. Washoe, split into Reno, Sparks, North Shore Lake Tahoe. White Pine. Balance of Counties.
There is no airport. There has never been an airport. The machines exist — more than a thousand of them — but as a nonrestricted licensee they fold geographically into Clark County Balance of County, a bucket that reported 57 filing entities and 17,831 slot units in that month. Whatever the airport's own numbers are, they are averaged into a group seventeen times its size.
So when you read that airport slots hold X percent, that number did not come from the regulator. It came from somewhere else. Two somewheres, actually, and both are worth looking at.
The $58.5 million that isn't gaming win
The most repeated figure about the airport is that its slot machines produce roughly $58.5 million a year. It has been in the Review-Journal, and from there into CDC Gaming, Fox5, News3LV, SimpleFlying, Airport World and others.
We opened the Clark County Department of Aviation's FY24 Annual Comprehensive Financial Report. The number is on page 28, and it is a revenue line for the airport — not a win figure for the machines:
Page 25 defines what that line is: the airport "receives a percentage of the net profits generated by gaming activity." A percentage of net profits is not the same thing as the amount players lost. The Review-Journal's own account says the county gets 86.5% of the win — which is a third, incompatible framing.
We can't reconcile those two definitions from the public record, and we're not going to guess which is right. What we can say is that $58.5 million is the airport's revenue from its gaming concession, and treating it as the machines' gaming win is a category error. If the 86.5% figure were accurate, the underlying win in FY2024 would be closer to $72 million — about 24% higher than the number in circulation.
The other number: a survey from 2001
The second figure everyone cites is that airport machines are about 8 percentage points tighter than casino machines. It traces to Michael Shackleford's slot survey on Wizard of Odds. That survey is real, careful, and still online. It is also from a specific moment and a specific sample:
- Fieldwork windowOctober 2001 – February 2002
- SampleFive-cent video reel slots only, 71 venues
- MethodPar sheet identification — 5–10 observed plays per machine
Here is what the table actually shows. Airport is the lowest return in the survey, which is the part everyone remembers correctly:
Airport 85.02% against a Strip average of 91.47% is a gap of 6.45 points, not 8. The 8.4-point figure appears only if you compare the airport to the single loosest venue in the survey, the Palms at 93.42%. Shackleford also notes that his data deliberately excludes video poker, keno and blackjack — so it is not comparable to NGCB statistics, which lump all electronic devices together.
A survey of nickel video slots conducted while the first iPod was on sale is not evidence about a floor that has since been rebuilt around penny multi-denomination cabinets. It was good work. It is now a historical document.
What the record does contain
None of this means there's nothing to know. The airport's own financial statements are public, audited, and specific — they just measure a different thing. Gaming fees revenue by fiscal year, in thousands:
Bars: gaming fees revenue, FY2015–FY2024. Bottom row: the same revenue divided by enplaned passengers.
Two things stand out. Revenue went from $27.7M to $62.6M in ten years — it more than doubled. And per enplaned passenger it went from $1.26 to $2.15, up 71%, while terminal concession revenue per passenger rose only 21% over the same period. Whatever is happening at that airport, gambling is growing three times faster than retail.
The most recent filing points the other way. For the first half of FY2026, gaming fees came in at $27,851k — down 10.4% year over year — which the airport attributes primarily to a 7.7% decline in enplanements. That is a leverage ratio of about 1.35: passenger traffic moves, gambling revenue moves harder. It makes the line a reasonable leading indicator for Las Vegas visitation generally.
Who actually runs them
The concession has been held for more than four decades by Michael Gaughan's Airport Slot Concession — Gaughan being the owner of the South Point. Reporting puts the count at more than 1,000 machines; one industry survey counts 1,416 across roughly 260 game titles, from penny up to dollar denominations. In November 2022 the concession announced cumulative gaming revenue had passed $1 billion over its 36-year run.
Airport gambling is rarer than its fame suggests. Reno-Tahoe has 243 machines, licensed and operated by IGT itself. Amsterdam Schiphol has 88 machines and three tables between the E and F piers, run by Holland Casino. Chicago authorised slots at O'Hare and Midway in its 2019 casino legislation — machines must sit past the TSA checkpoint — and seven years later not one has been installed; in June 2026 the licence holder was using it as leverage in a separate fight over city-wide video gaming terminals.
The scarcity isn't about demand. It's about the intersection of post-security floor space, exclusive concession rights, and state-level enabling law. All three have to line up.
What the airport machines are actually linked to
Here is the part nobody publishes, and it doesn't require a single estimate. Payback percentages at an airport are unknowable — but which wide-area progressive networks those machines sit on is a licensing fact, and IGT publishes the meters. Three airport concessions appear in that feed, carrying $20,088,159 between them.
The counter-intuitive one is Reno. A regional airport with 243 machines is carrying a bigger linked-progressive pool than the concession at the busiest gambling airport in the world — because Reno-Tahoe's machines are on Megabucks, the statewide network with the largest meter in Nevada, and the Las Vegas concession is not. Whatever else is true about airport slots, the ceiling in Reno is genuinely higher.
Meters as of Aug 18, 2026 · source: IGT public jackpot feed, refreshed every 30 minutes
What we're not going to tell you
We can't tell you what the machines at Harry Reid pay back, because nobody can. The regulator doesn't break the airport out. The concessionaire doesn't publish par sheets. The one survey that measured it directly did so 24 years ago on a machine type that barely exists anymore.
The honest position is that this is a documented gap, not a secret — and a widely-repeated number with no source behind it is worse than no number at all.
Sources
- Nevada Gaming Control Board — Monthly Revenue Report (May 2026 edition, 48pp, searched in full)
- NGCB Regulation 14 — the 75% statutory minimum and the mandatory-display list
- Clark County Department of Aviation — FY24 Annual Comprehensive Financial Report, pp. 25, 28, 143, 153
- Clark County Department of Aviation — FY26 Q2 Interim Financial Report, pp. 11–12
- Wizard of Odds — Slot Machine Odds in Las Vegas (survey conducted Oct 2001 – Feb 2002)
- Las Vegas Review-Journal — Why does the Las Vegas airport have slot machines
Every document above is public. We downloaded and read each one rather than citing coverage of them — which is how the $58.5 million error propagated in the first place.