Wild.io
Keno
KenoIntermediate

Keno House Edge: Why It Ranges From 1% to 40%

Every published keno house edge figure is true, just for a different game. From a claimed 1% online to over 40% at live lounges, and past 50% in the worst surveyed rooms, here's the full reconciliation, format by format.

MBMarcus Bell11 min readUpdated
Cover image for “Keno House Edge: Why It Ranges From 1% to 40%”
On this page

The short answer

Keno doesn't have one house edge. It has a range: from a claimed 1% at online crypto casinos to more than 40% at some live lounges, per Vegas Advantage's 2026 Las Vegas survey. Wizard of Odds' Laughlin survey found the worst lounges past 50%. Every published figure is accurate for a different format and paytable. This page reconciles all of them.

What is the house edge in keno?

The house edge is the average share of every dollar wagered that the game keeps over the long run. If a keno paytable returns 70 cents per $1 ticket on average, its house edge is 30%. It's the mirror image of RTP, return to player: the two always add up to 100%. If either term is new, start with what is RTP.

In keno, the edge lives entirely in the paytable. The draw itself, 20 numbers pulled from 80 (the rules are in our overview of how keno works), has fixed probabilities no venue can touch. What changes is what each catch pays, a "catch" being a drawn number that matches one of your spots; the keno terms glossary covers the vocabulary. Pay a little less for a catch, keep a little more of the money. That one lever explains everything below.

Why does the keno house edge range from 1% to 40%?

Because "keno" is one name stretched across at least five different products, and every published edge figure measures a different one. The 1% a crypto casino advertises and the 40%-plus a Las Vegas survey found are both real numbers. They just describe different games wearing the same name.

Here's every widely quoted keno house edge figure we could trace to a source, reconciled by format. Sources and years are in the table; each one is linked in its own section below.

Published house edge

Return to player

What it actually measures

Source (year)

1.00% (operator claim)

99% (claimed)

Online crypto keno, a 40-ball variant

Stake, published claim (retrieved 2026)

Under 4% to ~10%

~90% to 96%+

Best online RNG paytables, 80-ball game

Wizard of Odds (retrieved 2026)

5% to 16%

84% to 95%

Video keno machines, San Diego survey

Wizard of Odds (2008)

20% to 35%

65% to 80%

Live lounge keno, Las Vegas survey

Wizard of Odds (2001)

23% to 43%

57% to 77%

Live lounge keno, the 14 remaining Las Vegas lounges

Vegas Advantage (updated 2026)

About 30%

~70%

State lottery keno (Massachusetts)

GBH News (2023); Wild Academy calculation (2026)

26% to 50%

50% to 74%

Live lounge keno, Laughlin survey

Wizard of Odds (2012)

So when two articles tell you keno's edge is "about 5%" and "about 30%," neither is lying. One is describing a decent video keno paytable, the other a state lottery monitor game. The question is never "what is keno's house edge." It's "which keno, on which paytable."

Two caveats before you lean on any single row. First, the Wizard of Odds surveys are old: Las Vegas in 2001, video keno in 2008, Laughlin in 2012. That's 14 to 25 years ago, and no comprehensive public survey has replaced them since. Vegas Advantage's lounge count, updated in February 2026, is the only fresh live-keno datapoint we could find. Second, the 1% row is an operator's claim about a variant with different math entirely. Both get unpacked below.

One game, seven paytables: the proof

If you doubt the paytable is the whole story, Wizard of Odds settles it. Its Spot Keno page (retrieved 2026) documents seven real casino paytables for identical pick counts. Same draw, same probabilities, same catches. Yet pick-10 returns run from 86.43% on the stingiest table to 94.95% on the best. Pick-3 spans 83.25% to 94.35%. One degenerate pick-2 table on the page returns 0.00%: it pays on no outcome at all.

Here's the stingiest documented pick-10 paytable against one of the best, per unit wagered, from that same Wizard of Odds page:

Catch

Paytable 1 pays

Paytable 7 pays

0

1

0

4

1

1

5

5

5

6

25

24

7

120

146

8

500

1,000

9

2,000

4,500

10

10,000

10,000

Return

86.43%

94.95%

Look at the jackpot row first: both tables pay 10,000 for a solid 10. The nearly 7-point gap in return, which roughly doubles the house edge, hides in the middle rows. Paytable 7 pays 146 instead of 120 on a 7-catch, 1,000 instead of 500 on an 8, 4,500 instead of 2,000 on a 9. Players comparison-shop the jackpot; almost nobody comparison-shops catch 7. Venues know that.

To turn any paytable into a return yourself, you multiply each pay by the probability of that catch. Every one of those probabilities, for every ticket size, is in our keno odds and probability table.

How big is the edge at live keno lounges?

Live lounge keno is the expensive end of the game: surveys put it anywhere from 20% to 50%, depending on the city and the decade. Wizard of Odds' 2001 Las Vegas survey found live keno returning 65% to 80% of wagers, a house edge of 20% to 35%. Its 2012 Laughlin survey found worse: returns of 50% to 74%, meaning edges of 26% to 50%. That's past the 40% in our headline; the worst surveyed lounges break the typical range.

Fresher data is thin because the format itself is fading. Vegas Advantage's survey, updated February 4, 2026, counts just 14 live keno lounges left in Las Vegas, with paybacks ranging from 57% to 77%, edges of 23% to 43%. The same survey notes returns can differ by roughly 10 points for the same picks between casinos. In live keno, walking to a different lounge does more for your return than any number-picking method ever will.

Why does anyone accept a 30% edge? Pace. A live game runs every several minutes, so one $1 ticket buys a long stretch of anticipation, and the hourly cost stays small despite the huge per-dollar cut. That trade-off deserves its own numbers, and it gets them below.

What's the house edge on video keno?

Video keno sits in the middle of the range: Wizard of Odds' 2008 San Diego survey found machines returning 84% to 95%, a house edge of 5% to 16%. That's dramatically cheaper per dollar than the lounge. The same warning applies, though: the cabinet doesn't set the return, the paytable loaded onto it does. Two identical-looking machines can keep very different slices.

Players see this happen. On the Wizard of Vegas forum, a self-described 20-year video keno regular reports casinos adjusting a machine's payback simply by changing what a given catch pays. That's an anecdote, not a statistic, but it's precisely the mechanism those seven documented paytables prove.

The friendlier edge hides a trap: speed. A video keno machine runs as fast as you tap, easily hundreds of games an hour, so a 10% edge at 500 games an hour can cost far more per hour than a 30% edge at 15 games an hour. Pace multiplies edge. We run the hourly cost math, format by format, in video keno vs live vs online.

How much does state lottery keno really keep?

About 30 cents of every dollar, at least in Massachusetts, and that number isn't printed anywhere on the ticket. In 2023, GBH News reported that Massachusetts Keno is designed to return roughly 70%, quoting Albion College mathematics professor Mark Bollman: "The lottery can get 30% of your money in the long run by being honest." The money involved is serious. GBH put Massachusetts keno sales above $1 billion a year, second only to scratch tickets.

We didn't want to take one state's word for it, so Wild Academy computed the returns directly. State lotteries publish their full prize structures, and once you have a paytable, the return is arithmetic: multiply each prize by its exact probability and add them up.

Here is Wild Academy's 2026 calculation, per $1 base ticket with no multiplier add-ons, from the prize structures published by the Maryland, Ohio, and Massachusetts lotteries (retrieved July 2026):

Spots

Maryland RTP

Ohio RTP

Massachusetts RTP

1

50.00%

50.00%

62.50%

2

60.13%

66.14%

68.04%

3

62.44%

65.21%

69.38%

4

n/a*

64.94%

69.20%

5

54.27%

65.00%

69.99%

6

55.63%

64.79%

69.07%

7

59.54%

65.30%

69.96%

8

58.81%

64.75%

69.00%

9

56.37%

64.81%

69.78%

10

57.12%

63.67%

69.31%

* Our copy of Maryland's 4-spot paytable came from a mirror of the official prize page and produced an outlier we couldn't verify, so we're withholding that figure. The other Maryland figures come from the same mirror; treat their final decimals gently. Subtract any RTP from 100% for the house edge.

The Massachusetts column is the credibility check. It lands on 69% to 70% for nearly every spot count, independently matching the roughly-70% design target GBH reported and Bollman's 30% figure. Two routes, one from the state's own design and one from our hypergeometric math, arriving at the same number.

One worked example so the formula isn't abstract. Massachusetts' 3-spot pays $25 for catching all three and $2.50 for catching two. The catch-3 probability is 1 in 72 (0.013875) and catch-2 is 0.138754. So the expected return is 25 × 0.013875 ≈ 34.7 cents, plus 2.50 × 0.138754 ≈ 34.7 cents: about 69.4 cents back per $1, a 30.6% house edge.

Two findings from the table worth flagging. Ohio's 9-spot returns roughly 64.8%, in line with its 8-spot (64.75%) and 10-spot (63.67%) neighbors -- the state's paytables hold remarkably steady across spot counts. And a $1 10-spot returns roughly 57 to 69 cents depending on where you buy it, for the exact same draw. Why the bar-monitor game pays worse than even a Vegas lounge is a story of its own, told in lottery keno vs casino keno.

Is a 1% keno house edge real?

Treat it as an unverified operator claim about a different game. Stake, a crypto casino, publishes a 99% RTP and a 1.00% house edge for its keno on the game's own page (retrieved 2026). But Stake's game draws 10 numbers from a field of 40, not 20 from 80, so its probabilities come from entirely different math than everything else on this page. Its "risk level" settings reshuffle the paytable rather than the odds, and the multiplier tables are only visible in-game, so we can't independently compute the return. A 1% edge on a well-built RNG game is plausible. It's still a claim, not a measurement.

For the classic 80-ball game online, Wikipedia's keno entry (citing Wizard of Odds) puts the best online paytables at an edge of under 4%. The online spread runs to well over 35% on stingy paytables, against 20% to 40% for in-person casinos. In other words, online keno spans almost the entire reconciliation table by itself, depending on which paytable the operator loads. How online draws are generated, tested, and, on some platforms, verified by you directly, is covered in online keno RNG fairness.

The bottom line: the paytable is the price tag

The format tells you the neighborhood; the paytable tells you the price. Online RNG keno on a good paytable can cost under 4 cents per dollar. Video keno typically costs 5 to 16. Live lounges and state lottery monitors cost 20 to 50, per every survey and calculation above. Within each format, the spread between the best and worst paytable is often bigger than the gap between formats.

That points to the only keno "strategy" the math supports: read the paytable before you play, especially the unglamorous middle catches where the edge hides, and remember that speed multiplies whatever edge you've accepted. No number choice, pattern, or betting system moves any figure in this article by a single decimal. If you want to be able to run these expected-value calculations on any paytable you meet, our casino math course teaches the method step by step.

Run the numbers yourself

Our keno odds calculator shows the exact expected return and house edge for every spot count and wager — so you can see which tickets give you the best value.

Sources

Frequently asked questions

What is the house edge in keno?

The house edge is the average share of every dollar wagered that the game keeps over the long run; it's 100% minus the RTP. Keno has no single figure. Wizard of Odds surveys and Vegas Advantage's 2026 count put live lounge keno at roughly 20% to 43%, video keno at 5% to 16%, state lottery keno near 30%, and the best online paytables under 4% to about 10%.

Why does the keno house edge vary so much?

Because the paytable, not the draw, sets the edge. The 20-from-80 probabilities never change; what each catch pays does. Wizard of Odds documents seven real paytables for the same pick-10 ticket, returning anywhere from 86.43% to 94.95%, and one degenerate pick-2 table returning 0.00%. Formats then cluster: online paytables tend to pay best, video keno sits mid-range, and live lounges and lottery monitors pay worst.

What is keno RTP?

RTP (return to player) is the long-run average a keno paytable pays back per unit wagered; house edge is 100% minus RTP. Wild Academy computed state-lottery keno RTP from official 2026 prize structures: Massachusetts returns 69% to 70% on nearly every spot count, Ohio about 55% to 66%, and Maryland roughly 54% to 63%. Online operators claim far higher figures, up to a claimed 99%, which we can't independently verify.

Is keno the worst bet in the casino?

Per dollar wagered, live keno is among the most expensive: surveyed edges of 20% to 50% versus under 1% for blackjack played well. Per hour, it's less brutal, because live keno runs only a handful of games an hour. The reverse trap is video keno: a friendlier 5% to 16% edge played at hundreds of games an hour can cost more per hour than the lounge. Pace multiplies edge.

Which keno format has the lowest house edge?

Online RNG keno, on its best paytables: under 4% per Wikipedia, citing Wizard of Odds, though stingy online tables run far higher. One crypto operator, Stake, claims a 1% edge, but that's an unverified operator claim for a 40-ball variant with different math. Video keno comes next at 5% to 16% (Wizard of Odds, 2008), with live lounges and state lottery keno at the expensive end, 20% to 50%.

Go deeper with a course

About the author

Marcus Bell

Trust & Responsible Play Writer

Marcus focuses on judging casinos before you deposit — licensing checks, red flags, scam patterns — and the bankroll habits and limit tools that keep play under control.

One clear lesson, every week

Join thousands learning crypto the right way. New guides and plain-English breakdowns — never spam, unsubscribe anytime.