Every catalogue has a section for lottery-style games — keno, scratch cards, instant wins, bingo variants. They are usually filed under “Other”, they carry low minimum stakes, and they are the least examined games in the building.
They are also, with rare exceptions, the most expensive. House edges in this category commonly run three to ten times what the slots beside them charge, and unlike slots, the reason is visible in arithmetic you can do yourself. Keno in particular is the clearest worked example of a house edge available anywhere, because both the odds and the payouts are fully published.
Why these formats sit where they do
Lottery-style games descend from paper products — state lotteries, physical scratch tickets, hall bingo — where distribution costs, printing, retail margin and prize funds all came out of the same ticket price. Return rates of 50–65% were normal, because the alternative was no product at all.
Online versions inherited the format and, to a substantial degree, the pricing expectation. There is no printing cost and no retail margin, yet published returns in this category remain far below what the same operator offers on slots or tables. Comparing the RTP figures across a single lobby makes the gap immediately visible, and platforms that publish return percentages for every category rather than only for slots, Caswino among them, make the comparison something you can run in a minute rather than something you have to infer.
The gap is not hidden. It is simply in a section people do not check, attached to games that cost twenty cents to play.
How keno probabilities actually work
Keno draws 20 numbers from a pool of 80. You select between one and ten (sometimes more), and are paid according to how many of your selections appear.
The probabilities are pure combinatorics and identical everywhere. The table below shows the chance of catching all of your selections.
| Spots picked | Chance of catching all | Approximate odds |
|---|---|---|
| 1 | 25% | 1 in 4 |
| 2 | 6.01% | 1 in 16.6 |
| 3 | 1.39% | 1 in 72 |
| 4 | 0.306% | 1 in 326 |
| 5 | 0.0645% | 1 in 1,551 |
| 6 | 0.0129% | 1 in 7,753 |
| 8 | 0.000435% | 1 in 230,115 |
| 10 | 0.0000112% | 1 in 8,911,711 |
The ten-spot line is the one worth sitting with. Catching all ten is roughly a one-in-nine-million event — comparable to a mid-sized lottery jackpot — and the payout offered for it is typically 10,000 to 25,000 times the stake. A one-in-nine-million chance paying 20,000 to one is the entire house edge in a single line.
This is why keno edges land where they do. Typical online keno returns 88–95%; land-based and some legacy online versions run as low as 70–75%. The payout table is simply set far below the true odds at every level, and because the true odds are computable, the gap is verifiable rather than a matter of trust.
Why the RTP range is so wide
Slots cluster between 94% and 97%. Lottery-style games in the same lobby can range from 85% to 95%, and occasionally lower, with no visual indication of which is which.
Three reasons. First, there is no competitive pressure on these figures because almost nobody compares them. Second, the format supports very low stakes, which makes a large percentage of a small number feel inconsequential. Third, many of these games are supplied by smaller studios whose commercial model depends on higher margins per title.
The practical consequence is that two scratch card games sitting next to each other, with identical stakes and near-identical presentation, can differ by eight percentage points in return. On slots that spread would be extraordinary; here it is routine, which makes checking the info panel unusually valuable in this section specifically.
Scratch cards and instant wins
Online scratch cards work exactly like slots with a different presentation layer. An RNG determines the outcome at purchase, and the scratching animation reveals a result that was already fixed. The “choice” of which panel to scratch first changes nothing.
Published returns typically sit between 85% and 95%, and the variance is unusual: most tickets return nothing, and the distribution is heavily concentrated in a small number of large prizes. This produces long losing sequences with occasional sharp reversals, similar to a high-volatility slot but usually at a worse return.
Instant-win formats — crash-style games, mining grids, plinko variants — vary more widely. Some carry returns competitive with slots at 96–97%; others sit well below. The category is too heterogeneous to generalise about, which means the info panel matters more here than anywhere else in the lobby.
What makes them feel cheaper than they are
Several design properties consistently cause these games to be underestimated:
- Very low stakes. A 12% edge on a $0.20 ticket is $0.024 per play, which registers as nothing — until you notice the format supports 400 plays an hour.
- Fast resolution. Scratch cards and instant wins resolve in two to four seconds, which makes the effective hourly cost far higher than the stake suggests.
- The near-miss structure. Matching two of three required symbols is engineered to occur frequently, and it reads as almost winning rather than as losing.
- Familiarity from paper products. People carry expectations from physical tickets, where a 50% return was normal and nobody expected otherwise.
- No comparison habit. Nobody researches scratch card RTP the way they research slot RTP, so the figures go unexamined.
- Selection rituals. Picking your own keno numbers, choosing which panel to reveal — these create an impression of participation in an outcome that is already determined.
The hourly arithmetic is the useful corrective. A $0.50 scratch card at 88% return, played 300 times an hour, has an expected cost of $18 an hour. A $0.50 slot spin at 96%, played 600 times an hour, costs $12. The scratch card costs more per hour at half the play rate and a quarter of the visible stake.
Practical points
Check the RTP in the info panel before playing anything in this category — the spread is wider here than anywhere else, and it is published. Compare against the slots in the same lobby, which is a fair comparison at similar volatility. Note the resolution speed and calculate cost per hour rather than per ticket. On keno, pick fewer spots: the edge is generally lower on two to four spots than on eight to ten, where the large payouts sit. And treat the selection rituals as presentation, because the outcome is generated before any choice you make.
A closing note that belongs in any honest piece on this subject: this category carries the highest house edges in most lobbies, and low stakes disguise rather than reduce that cost. Set a budget you are entirely comfortable losing, decide your stop point in advance, and step away when you reach it. Gambling is for adults 18+ only, and free confidential support services are available in most countries.
About the Author
