House Edge and RTP Explained: Is an Online Casino Rigged?

By The editorial team · Updated 7 October 2026

The short answer

House edge is the share of each bet the casino keeps on average over the long run. Return to player (RTP) is the same fact seen from the other side: an RTP of 96% means a 4% house edge. Volatility is a different thing: it describes how bumpy the ride is, not how much you lose on average.

A game can be perfectly fair and you can still lose, because the edge is built into the payouts, and short-term results swing far above and below the average. No one can tell you which online casino is “not rigged” from a distance, but you can check the licence, the published rules and, for in-house games, provably fair verification. None of that removes the edge. This is general information, not a promise about any operator.

House edge, in plain numbers

The Wizard of Odds defines the house edge as “the ratio of the average loss to the initial bet”. A 1% edge means that, on average, you lose 1 unit for every 100 units you bet. It is an average over many bets, not a forecast for tonight.

Here is a real, checkable example. In single zero (European) roulette there are 37 pockets, and a straight bet on one number pays 35 to 1. Wikipedia gives the calculation: you lose your unit 36 times in 37 and win 35 units once in 37, so the expected result is a loss of 0.0270 units per unit bet, a 2.70% house edge. With a second zero (American roulette, 38 pockets) it becomes 5.26%. The payout stays the same, and the extra zero makes the edge larger.

RTP is the mirror image

RTP is quoted mostly for slots and similar games. It is the theoretical share of money wagered that the game returns to players over a very long run:

House edge = 100% minus RTP.

Wikipedia’s slot machine article gives the textbook case: a 1 unit per spin slot with a 95% RTP returns on average 950,000 units out of 1,000,000 wagered “over a sufficiently long period such as 1,000,000 spins”.

Hypothetical example, RTP

All figures below are made up to keep the arithmetic simple. They are not any real game’s numbers.

Note that “total wagered” counts every bet, including money you won earlier and bet again. Bet 1,000 times with a 100 unit bankroll and you are wagering 1,000 units.

Volatility: same RTP, very different ride

Wikipedia describes volatility as “the measure of risk”: a low volatility slot has regular but smaller wins, and a high variance slot has fewer but bigger wins.

Hypothetical example, two games with the same 96% RTP

• Game A (low volatility): wins 1.92 units on half of all spins (50% x 1.92 = 0.96 units per spin returned).
• Game B (high volatility): wins 96 units on 1 spin in 100 (1% x 96 = 0.96 units per spin returned).

Both return 96% on average. But on Game B, the chance of getting no win at all in 100 spins is 0.99 to the power of 100, which is about 37%. On Game A, a dry spell that long is practically impossible, while Game B regularly produces both big wins and long losing runs. Neither game is “rigged” or “loose”. They are shaped differently.

Why a fair game can still lose

Take the roulette example above. Our own arithmetic from the same payout table, for 100 straight number bets of 1 unit on a single zero wheel:

• The average result is a loss of 2.7 units (100 x 2.70%).
• The standard deviation of a single bet is about 5.8 units, so after 100 bets the typical swing is about 58 units, many times the average loss.
• You finish ahead if you hit the number at least 3 times (3 x 35 = 105 units against 97 lost bets). The chance is about 51%.

So after 100 bets, roughly half of players would be ahead even though the game has a 2.70% edge. The edge has not gone anywhere. A few lucky hits mask it, and the average is pulled by the rare large win. Over 1,000 bets the average loss grows to about 27 units, and the chance of being ahead (28 or more hits) falls to about 45%. Keep playing and the share of winners keeps falling, which is why the house is not worried by individual winners.

The 100 and 1,000 bet probabilities above are our own arithmetic from the roulette payout table, not published figures. Wizard of Odds makes the same point: results can deviate a lot in the short term, but over extended play they converge toward the expectation defined by the house edge.

RTP is not a promise for your session

RTP describes a theoretical long run. Wikipedia notes that the Federal Court of Australia found that information about RTP can mislead players because it describes the percentage of winnings for a long-term game, “which is not true for a short-term one”. Practical consequences:

• A slot with a 96% RTP can pay nothing for hundreds of spins and still be working as designed.
• A “hot” or “cold” machine is a story, not a signal. Each spin of a properly working random game is independent.
• A bonus changes your expected cost. If you take a 100 unit bonus with 30x wagering, you must bet 3,000 units. At a hypothetical 4% edge, that is an expected cost of 120 units, more than the bonus. See wagering requirements explained for the full method.

How big are house edges?

The Wizard of Odds table shows how much they vary by game: roulette at 2.70% (single zero) and 5.26% (double zero), baccarat banker bet 1.06%, and craps pass line 1.41%, all assuming correct play. For slots it says the edge varies considerably, “2%-15%”, depending on the machine. We do not print the RTP of any named slot here, because it can change between versions and must be read where the provider or operator publishes it. Look for the RTP in the game’s own information or paytable screen, and note the date. Some commentary says the same game can be offered in more than one RTP version (UNVERIFIED, industry commentary only, 7 October 2026), so read the figure inside the game you are actually playing rather than a number from a review.

What provably fair does and does not change

Provably fair lets you verify that a result was generated from a committed seed and was not altered after your bet. It is a check on honesty, not on generosity. It does not change the house edge: a dice game with a 1% edge is still a 1% edge game after you verify it. It also only applies to the casino’s own games, not to third party slots. The step by step method is in provably fair explained.

So which casino is not rigged?

You cannot prove a negative, but you can reduce the risk:

1. Check the licence on the regulator’s own register, and that the domain matches. See how to check a casino licence.
2. Read the published rules, the paytable, and where available the RTP shown in the game.
3. Use provably fair verification on in-house games, as a sample check.
4. Watch for payment behaviour. Fair games mean little if the operator will not pay. See casino withdrawal pending or not paid and crypto casino scam red flags.
5. Remember the edge exists everywhere. The honest question is not “is it rigged” but “what is the edge, and can I afford the swings”.

For a broader pre-deposit checklist, see how to choose a crypto casino. When you are ready to compare operators, see the casinos we compare on our best crypto casinos page.

A note on responsible play

Because the house edge is built in, gambling is a cost, not an income. Only stake money you can afford to lose and set limits before you play. The legal age is 19+ in most provinces (18+ in Quebec, Manitoba and Alberta). If it stops being fun, see our responsible gambling page.

Sources

• Wizard of Odds, house edge (definition, game figures, short term variance): https://wizardofodds.com/gambling/house-edge/
• Wikipedia, Roulette (house edge calculation for single and double zero): https://en.wikipedia.org/wiki/Roulette
• Wikipedia, Slot machine (RTP example, volatility definition): https://en.wikipedia.org/wiki/Slot_machine
• Wikipedia, Return to player (RTP and house edge, Federal Court of Australia note): https://en.wikipedia.org/wiki/Return_to_player