How Casino Odds Work
The two numbers behind every casino bet
Every bet on a casino floor has two numbers attached to it, even though only one of them is usually printed anywhere visible. The first is the true odds — the actual mathematical probability of an outcome, derived from counting cards, wheel pockets, or dice combinations. The second is the payout odds — what the casino actually pays when that outcome happens. Casino odds, as a general subject, is really the study of the relationship between those two numbers, and understanding that relationship is the single most useful piece of math a player can carry into any game.
This guide walks through that relationship on its own, as the foundational piece behind the broader casino odds & house edge guide, which builds it out into RTP, expected value, variance, and volatility. If you only read one guide in this category, the concept here is the one everything else is built on.
True odds: what the game actually produces
True odds describe the real, unadjusted probability of an outcome, expressed as a ratio of losing outcomes to winning outcomes. A single number on a European roulette wheel has 36 losing pockets and 1 winning pocket (ignoring the zero for a moment isn't quite right either — the zero is itself one of the 37 total pockets, and it loses for every standard inside or outside bet except a bet placed directly on it). So the true odds against any single number hitting are 36-to-1: for every 37 spins, on average, that number comes up once and misses 36 times.
True odds are fixed by the physical or mathematical structure of the game — the number of cards in a deck, the number of pockets on a wheel, the number of combinations two dice can produce — and they don't change based on anything a casino decides. What a casino does control is the second number.
Payout odds: what the casino agrees to pay
Payout odds are the terms printed on the felt or paytable: bet $1, win $35 (i.e., "35-to-1") on a straight-up roulette number. If payout odds matched true odds exactly, the game would have no house edge at all — over a large enough number of spins, the amount paid out on wins would exactly balance the amount collected on losses, and the game would be, in the statistical sense, "fair" to both sides.
That never happens on a standard casino bet. The straight-up roulette payout is 35-to-1 against true odds of 36-to-1 — a one-unit shortfall. That shortfall, and versions of it repeated across every bet on every game, is where the house edge lives. It's a small, deliberate gap between what the game's true probability would justify and what the casino actually agrees to pay, and it's present in some form on virtually every standard wager you can place.
Turning the gap into a percentage
The gap between true odds and payout odds converts into the familiar house-edge percentage through a simple calculation: weight the payout by the probability of winning, weight the loss by the probability of losing, and net the two figures against the amount wagered.
For the single-number roulette bet: the probability of losing is 36/37, losing $1 each time; the probability of winning is 1/37, winning $35. The math: (36/37 × $1) − (1/37 × $35) = $0.973 − $0.946 = $0.027. On a $1 bet, that $0.027 shortfall is a 2.70% house edge — arrived at through the payout math, and matching exactly what you'd expect from counting the extra pocket directly (1 losing pocket beyond the "fair" 36, out of 37 total, is also 1/37 ≈ 2.70%). Both routes to the number agree, because they're describing the same underlying gap from two different angles.
This same calculation applies to every casino bet, though the true-odds and payout-odds inputs look completely different from game to game. A blackjack hand's "true odds" are really a blend of dozens of possible two-card and multi-card outcomes weighted by probability, and a slot's "true odds" are the weighted probabilities across its entire reel and paytable structure — considerably more complex to calculate by hand than a single roulette number, which is why these figures are certified by independent testing labs rather than derived at the table. But the underlying principle — payout set below true odds, producing a fixed percentage the casino keeps on average — is identical across every one of them. The full walkthrough of that percentage, once calculated, and how it connects to RTP and expected value, is in the casino odds & house edge guide.
The one common exception: craps odds bets
There's exactly one widely available standard casino bet that pays true odds with no shortfall at all: the odds bet in craps, placed as a supplement to a pass line, don't pass, come, or don't come bet. If a point of 6 is established, true odds against making that point before a 7 rolls are 6-to-5, and the craps odds bet pays exactly 6-to-5 — no shortfall, no house edge, a mathematically neutral bet in isolation. The catch is that odds bets can't be placed on their own; they require an underlying pass line or don't pass bet, which does carry its own house edge (1.41% and 1.36% respectively), so the "blended" edge across the whole wager shrinks as more odds are added but never fully reaches zero unless the odds portion is infinite relative to the base bet.
This exception is worth knowing precisely because it's rare enough to prove the general rule: on essentially every other standard bet in a casino, payout odds sit at least slightly below true odds, and that gap is where the entire economics of casino gambling comes from.
Odds across different bet formats
Odds get expressed in a few different formats depending on the game and the region, and it's worth being able to translate between them quickly. "35-to-1" and "36:1" mean the same thing as a ratio. "800-for-1," common in video poker paytables for a royal flush, is subtly different — it describes the total payout including your original wager, rather than the profit on top of it, so an 800-for-1 payout on a $1 bet returns $800 total, equivalent to 799-to-1 in the more common ratio format. Percentages ("a 2.70% house edge") describe the same underlying gap as a long-run average cost rather than a per-bet ratio. None of these formats change the math — they're just different ways of presenting the identical true-odds-versus-payout-odds relationship this guide has been building from the start.
Why this matters more than "luck"
Framing casino odds this way — as a fixed, calculable gap between two numbers — is a more useful mental model than thinking about games purely in terms of "luck," because it explains both why the house edge is so durable (it's built into the payout structure itself, not into any pattern of outcomes) and why no amount of timing, betting pattern, or superstition can close that gap. A player who doubles their bet after a loss, or waits for a certain number to "hit," is doing nothing to the true-odds-versus-payout-odds ratio of the next bet — that ratio is fixed by the game's rules, identical on the next bet regardless of anything that happened on previous ones. The full explanation of why betting patterns specifically can't close this gap is in why casino strategies don't beat the house, and the related concept of expected value — turning this percentage gap into an actual dollar figure for a specific bet size — is covered in expected value in casino games explained.
Frequently asked questions
What are "true odds" in a casino game? The actual, unadjusted mathematical probability of an outcome, based on the physical or combinatorial structure of the game — pockets on a wheel, cards in a deck, or combinations two dice can produce — expressed as a ratio of losing to winning outcomes.
What are "payout odds"? The amount a casino actually pays when a bet wins, which is set at or below true odds on essentially every standard casino bet, with the shortfall (when present) producing the game's house edge.
Is there any casino bet that pays true odds exactly? Yes — the craps odds bet, placed alongside a pass line, don't pass, come, or don't come bet, pays exactly true odds with 0% house edge on that specific portion of the wager.
Why don't casinos just pay true odds on every bet? Because the gap between true odds and payout odds is the entire source of a casino's revenue on games of chance — a game paying true odds throughout would have no built-in house edge and no expected long-run profit for the operator.
Does understanding true odds help me win more often? It doesn't change your probability of winning any individual bet, but it does let you compare the actual cost of different bets accurately, which is the basis for choosing lower house-edge bets over higher-cost ones.
How is house edge calculated from true odds and payout odds? By weighting the payout by the probability of winning and the loss by the probability of losing, then netting the two against the amount wagered — see the worked roulette example above for the full calculation.
What does "800-for-1" mean versus "800-to-1"? "For-1" payouts include your original wager in the total returned; "to-1" payouts are profit paid on top of your original wager, which is still returned separately. The two formats describe different numbers even when the headline figure looks similar.
