Keno

Reading the ten-spot paytable

Eighty numbers, twenty drawn

You select exactly ten distinct numbers from 1 through 80. The game then draws twenty numbers without replacement, and your result is how many of your ten appear among those twenty.

Because twenty of eighty numbers are drawn, a quarter of the board comes up every round. That means catching two or three of your ten is the ordinary outcome — the expected number of hits is two and a half — and it is precisely those ordinary outcomes that pay nothing.

Selection order does not matter, and quick pick uses the same unbiased shuffle the draw does. There is no difference in value between numbers you chose deliberately and numbers the machine chose for you.

There is no strategy, and the paytable explains why

Every number is equally likely and the draw has no memory, so no selection method changes your expected return. Birthdays, hot numbers, cold numbers, and quick pick all produce identical value. This is worth saying plainly because keno attracts more selection systems than almost any other game, and none of them do anything.

What is genuinely worth understanding is the shape of the paytable. Nearly all of the return is concentrated in outcomes you will almost never see — catching eight, nine, or ten. Meanwhile the outcome you will see most often, two or three hits, returns nothing. The zero-hit consolation exists to soften that, and catching zero is actually more likely than catching seven.

So the honest framing is that keno is a lottery-shaped bet: long odds, large top prizes, and a return that arrives in rare lumps rather than steadily. Knowing that is the only useful preparation there is.

The paytable, and where the return actually sits

Multipliers are on your stake and returns are gross. The theoretical return to player for this table is 95.1587 percent — a genuinely computed figure derived from the hypergeometric probabilities of the ten-spot game, not a target rounded to a convenient number.

Catching one, two, three, or four numbers pays nothing. That is four of the eleven possible outcomes and by far the most likely region of the distribution, which is why the game feels barren between wins.

Common mistakes

Keno’s errors are almost entirely about misreading randomness.

• Tracking hot and cold numbers. Draws are independent and the board has no memory, so a number that has not appeared recently is exactly as likely as any other.
• Believing quick pick is worse. It uses the same unbiased shuffle as the draw itself; a hand-chosen ten and a machine-chosen ten have identical expected value.
• Picking consecutive numbers or patterns. Any specific set of ten is as likely as any other — a pattern only feels less random, it is not.
• Expecting steady returns. Most of the paytable’s value sits in eight, nine, and ten hits, so long stretches with nothing back are the normal shape of the game, not bad luck.

What is different about this version

The theoretical return is published as an exact constant computed from the game’s own probabilities — 0.951587344114154 — rather than rounded to a marketing figure. Most keno games do not tell you their return at all, and among the practice tables here it is one of the few genuinely derived rather than set to a flat target.

The twenty numbers are drawn without replacement from a proper shuffle of all eighty, so no number can be drawn twice in a round and the hypergeometric probabilities the return figure depends on are actually the ones being played.

Written by the Silverweb Games team from this game’s own implementation. Last reviewed July 31, 2026.