Last reviewed: January 2026 · Prepared by: the iGaming editorial team, reviewed by Claire Bennett, senior casino content editor (see our online casino platforms coverage) · 18+ only
General advice warning. This guide is educational. It is not legal, financial or medical advice. Gambling involves the risk of financial loss, and no article, system or strategy can remove that risk. If gambling is affecting you or someone you know, contact Gambling Help Online on 1800 858 858 (24/7, free, confidential).
Roulette looks simple, and that is exactly why it deserves a careful read. Two minutes of rules, then a lifetime of people arguing about systems that cannot work. Below you will find the maths, the variants, the live dealer mechanics, the bonus traps and the Australian legal position, all in plain English.
This guide is written for two kinds of reader. The first is new to the game and wants the rules, the bet types and a clear answer on which wheel to sit at. The second already plays and wants the uncomfortable detail: bonus weighting, withdrawal caps, product-risk research and why the Martingale keeps failing.
Some readers should walk away instead, and we would rather say so early.
Fair enough? Then let us start with the machinery.
Roulette is the most recognisable wheel game in the casino world, and its appeal is structural. The rules take two minutes to learn, yet the betting layout offers more than a dozen distinct wagers at different risk levels. Your only task is to predict where the ball will come to rest. Everything else, from layout to payouts to dealer procedure, exists to price that prediction.
The wheel. A precision-balanced, slightly tilted wheel head with numbered pockets around the circumference. European and French wheels carry 37 pockets (0 and 1 to 36); American wheels carry 38 (0, 00 and 1 to 36). The croupier spins the roulette wheel in one direction and launches the ball along the upper ball track in the opposite direction.
The table (layout). The green baize is divided into two zones. The numbered grid is the inside area, used for bets on individual numbers and small groups. Everything outside the grid, meaning dozens, columns, red/black, odd/even and high/low, is the outside area of the roulette table.
The ball ("pill"). Historically turned from ivory; modern balls are manufactured from nylon, phenolic resin or moulded acetal, typically 18 to 21 mm in diameter. Material and size affect bounce and scatter, which is why regulated venues rotate balls.
The marker ("dolly"). A metal or acrylic marker, sometimes shaped like a small chess piece or a cylindrical glass block, placed on the winning number after the spin. Until the dolly is lifted, no player may touch chips on the layout.
Chips. Roulette uses table-specific chips so each player's wagers can be told apart. Smaller denominations are usually round, larger "plaques" rectangular. Online, chip denominations are selected from a toolbar, commonly $0.10, $1, $5, $25, $100 and $500.
The rake. The long-handled tool the croupier uses to sweep losing chips to the house and push winnings across the layout.
Design note: the arrangement of numbers around the wheel is deliberately non-sequential, alternating red and black and interleaving high and low numbers so that no quadrant clusters similar outcomes. This is a long-standing manufacturing convention documented in equipment specifications rather than in peer-reviewed literature. Independent published data on layout standards is limited, so treat the rationale as industry practice rather than a tested claim.
(Multimedia placeholder: flow diagram of a single roulette round, presented as a semantic ordered list in the DOM with a short caption for each step: choose chips → place bet → wheel spins → ball settles → bets are settled. Each step needs its own text explanation so the sequence is readable without the graphic.)
Pace note: in live dealer roulette a full cycle commonly runs somewhere in the region of 30 to 60 seconds, depending on dealer procedure and the number of seated players. RNG roulette can complete a round in a handful of seconds, because the player controls the "spin" button. Exact cycle times are set by each operator and studio and are not published consistently, so treat these ranges as indicative; verified game-round duration data from operators would be needed to state them precisely. The practical implication is the part that counts: faster cycles mean more spins per hour, and more spins per hour means more exposure to the house edge.
Playing roulette is a short loop, repeated. Learn the loop and the rest is detail.
Two beginner points worth stating plainly. First, roulette bets can be combined freely, so a single round might carry a straight-up bet, two splits and a dozen. Second, combining bets does not reduce the house edge; it only smooths the shape of your results.
Choosing the wheel is the single most consequential decision a roulette player makes, bigger than any betting pattern.
(Multimedia placeholder: comparison table with <caption>, <thead> and <tbody>, comparing European, French and American roulette by wheel structure, table features, available bets and odds, with a short text summary repeated underneath for readers who skip tables.)
| Parameter | European | American | French |
|---|---|---|---|
| Pockets on the wheel | 37 (0, 1 to 36) | 38 (0, 00, 1 to 36) | 37 (0, 1 to 36) |
| Zero structure | Single zero | Single zero plus double zero | Single zero |
| House edge | 2.70% | 5.26% | 1.35% on even-money bets (2.70% on other bets) |
| Special rules | None | Five-number bet (0-00-1-2-3) at 6:1, giving a 7.89% edge | La Partage and En Prison |
| Layout language | English | English | French (Passe/Manque, Pair/Impair, Rouge/Noir) |
| Outside bets placed | Below the number grid | Below the number grid | On both sides of the grid |
| Announced bets | Usually available | Rarely offered | Standard (Voisins du Zéro, Tiers, Orphelins) |
| Best for | Everyday play | Avoid where possible | Even-money strategists |
In one line: the French table is the best value if you bet even money, the European table is the sensible default, and the American wheel is the one to leave alone.
Both rules apply only when the ball lands in zero, and only to even-money bets (red/black, odd/even, high/low):
Either rule halves the effective edge on even-money bets from 2.70% to 1.35%, which makes French roulette mathematically the most player-friendly wheel game on a casino floor or in a live studio. Note that La Partage and En Prison are table rules, not wheel features. Some tables branded "French Roulette" quietly omit them, so always open the game's info panel.
Both wheels pay 35:1 on a single number. The European wheel has 37 outcomes; the American wheel has 38. The payout does not change to compensate for the extra pocket, so the gap between true odds and paid odds widens:
Over 100 spins at $5 a spin, that is a theoretical loss of about $13.50 on a single-zero wheel against $26.30 on a double-zero wheel. Same game, double the cost.
Payouts are quoted as "X to 1", meaning profit per unit staked. Total return equals net win plus returned stake. Bet $100 on a single number at 35:1 and a win returns $3,600 ($3,500 profit plus the $100 stake).
| Bet | Numbers covered | Chip placement | Payout | Win chance (European) | Win chance (American) |
|---|---|---|---|---|---|
| Straight up / Single | 1 | Centre of the number | 35:1 | 2.70% | 2.63% |
| Split | 2 | Line shared by two numbers (vertical or horizontal) | 17:1 | 5.41% | 5.26% |
| Street | 3 | Outer edge of a row of three | 11:1 | 8.11% | 7.89% |
| Corner / Square | 4 | Intersection where four numbers meet | 8:1 | 10.81% | 10.53% |
| Six Line / Double Street | 6 | Outer corner between two rows | 5:1 | 16.22% | 15.79% |
| First Four / Top Line | 4 (0-1-2-3) | Corner between 0 and 1, or 0 and 3 | 8:1 (European) | 10.81% | n/a |
| Five-number bet | 5 (0-00-1-2-3) | Corner between 0/00 and 1/3 | 6:1 (American only) | n/a | 13.16% (7.89% edge, avoid) |
Worked examples: a $100 split win at 17:1 returns $1,800; a $100 street win at 11:1 returns $1,200; a $100 corner win at 8:1 returns $900; a $100 six-line win at 5:1 returns $600.
| Bet | Numbers | Payout | Win chance (European) | Win chance (American) |
|---|---|---|---|---|
| Red / Black (Rouge / Noir) | 18 | 1:1 | 48.65% | 47.37% |
| Odd / Even (Impair / Pair) | 18 | 1:1 | 48.65% | 47.37% |
| Low 1 to 18 / High 19 to 36 (Manque / Passe) | 18 | 1:1 | 48.65% | 47.37% |
| Dozen (1st 12 / 2nd 12 / 3rd 12) | 12 | 2:1 | 32.43% | 31.58% |
| Column (2:1 boxes at the end of the grid) | 12 | 2:1 | 32.43% | 31.58% |
Zero, and 00 where it exists, belongs to no colour, no dozen and no column, which is precisely where the house edge lives. On a single-zero wheel every wager in both tables above carries the identical 2.70% edge; what changes is variance, not value. Straight-up bets deliver rare, large wins. Even-money bets deliver frequent small ones and a slower, steadier drawdown.
Fact-check note: you will sometimes read that a straight-up win on a European wheel is "37 to 1". That is imprecise. The probability is 1 in 37, so the true odds against are 36 to 1, while the casino pays 35 to 1, and that one-unit gap is the entire house margin.
Racetrack bets are simply pre-packaged combinations of splits, straights and corners. They do not alter the house edge. What they do is let you cover a wheel sector rather than a table area, which is why sector-based players prefer them.
| Aspect | Roulette demo / free play | Real-money play |
|---|---|---|
| Stakes | Virtual credits | Cash balance, with real losses |
| Withdrawals | None | Subject to KYC, limits and processing times |
| Account requirements | Often none | Registration, ID and age verification |
| Game engine | Frequently the same certified build | Same maths, though limits, autoplay and bonus features may differ |
| Live dealer tables | Almost never available (video streams require a funded seat) | Available |
Demo mode is genuinely useful for three things: learning chip placement (where exactly a corner or six-line chip sits), learning interface mechanics (rebet, double, undo, clear, statistics panel), and pressure-testing a betting pattern before it costs anything. It is also the cheapest way to compare games of roulette side by side, since a French table with La Partage feels different to a speed table with a five-second window.
What demo mode cannot simulate is the emotional weight of a real loss, the variable that breaks most staking plans.
There is a documented behavioural bridge between free-play products and cash play. Research on the convergence of gaming and gambling has repeatedly found that non-monetised play can increase subsequent monetary gambling.
"In a German study of 2,791 adults, participation in all forms of simulated gambling except lotteries predicted the frequency of real-money betting after controlling for demographic variables." Fiedler et al. (2024), cited in The Convergence of Simulated Gambling, Monetary Gambling and Gaming (2023). https://www.gamblingresearch.org.au/resources/convergence-simulated-gambling
Two practical consequences follow. First, free roulette demo play should be treated as a training tool with a time limit, not as entertainment in its own right. Second, demo returns are not representative: a demo bankroll that survives 300 spins tells you nothing about the next 300.
Legal and age disclaimer. Roulette for real money is restricted to adults, 18 or over in Australia and 18 or 21 elsewhere, and only in jurisdictions where the format is lawfully offered. Confirm the rules that apply where you are physically located before depositing. This is general information, not legal advice.
Live dealer roulette is a real-time broadcast of a real table. A human croupier operates a physical wheel in a studio or on a casino floor, multiple cameras cover the wheel head, the ball track and the dealer, and you place chips through a software layout overlaid on the video feed.
How the result reaches your balance. High-frame-rate cameras and optical character or pattern recognition (OCR) track the ball and the wheel and identify the winning pocket. That result is written to the game server and settled against every open bet at the table. Because the outcome comes from the bounce of a physical ball rather than an algorithm, players can see the mechanism that decides their wager, which is a meaningful transparency advantage.
Equipment. A professional-grade wheel, traditionally hardwood with a machined ball track, balls of phenolic resin or moulded acetal, a dolly for marking the winning number, and studio lighting engineered so the ball stays visible at speed.
Network requirements. Operators generally recommend around 12 Mbps for a smooth Full HD stream. Lower-bandwidth SD streams can function from about 4 Mbps, but latency is the real risk: a laggy connection means your chip does not register before "no more bets", and there is no appeal against a bet that never reached the server.
Table limits. Land-based minimums frequently sit at $10 to $25 per spin. Live online tables typically start at $0.10 to $1 and extend to $5,000 or $10,000 on VIP and salon privé tables, while mid-market live tables commonly cap out around $500 to $1,000.
Variety. Beyond classic European tables, live studios operate multiplier formats (lightning-style tables where random numbers receive boosted payouts at a reduced base return on straights), immersive multi-camera tables with slow-motion replays, dealer-free automated wheels running very fast cycles, dual-ball tables, and speed tables with shortened betting windows. Many tables offer side bets and racetrack layouts for neighbours and call bets. If you want an overview of formats and providers, our coverage of live dealer casino games goes deeper.
Warning signs. A "live" table that is really a looped recording, a studio that looks improvised, unpublished payout information, dealers who never respond in chat, and deposits declined by payment providers or banks are all reasons to close the tab.
"Live roulette showed a Spearman coefficient of 0.20 between participation and markers of harm, versus 0.10 for RNG roulette; slots recorded 0.34." Delfabbro et al. (2023), behavioural markers study, Kindred/Unibet UK dataset, 100,000 users. https://doi.org/10.1007/s10899-023-10200-7
RNG roulette resolves each spin with a certified random number generator. Advantages: instant rounds, tiny minimums, free demo access, no waiting for other players, and layouts that store favourite bet patterns. Disadvantages: no visible mechanism, no social element, and, critically, speed. A player who completes 200 RNG spins an hour is exposed to roughly three times more theoretical loss than a live player completing 60.
| Criterion | Live dealer roulette | RNG roulette |
|---|---|---|
| Outcome source | Physical wheel and ball, read by OCR | Certified RNG software |
| Transparency to the eye | High, you watch the spin | Requires trust in lab certification |
| Rounds per hour | About 40 to 70 | Potentially 150 to 300 plus |
| Minimum stake | Usually higher | Usually lower (from $0.10) |
| Maximum stake | Usually higher (VIP tables) | Usually lower |
| Demo mode | Rarely | Almost always |
| Bandwidth needed | About 12 Mbps recommended | Minimal |
| Social interaction | Dealer chat, table chat | None |
| Atmosphere | Casino-floor realism | Utilitarian |
| House edge | Identical for the same variant | Identical for the same variant |
The critical point: live streaming changes the experience, not the mathematics. A single-zero live wheel and a single-zero RNG wheel both carry a 2.70% edge.
Roulette is the classic "excluded or discounted" game in bonus terms, precisely because its even-money bets have low volatility and a low edge. Typical contribution rates look like this:
| Game category | Typical wagering contribution |
|---|---|
| Pokies and slots | 100% |
| Keno, scratch cards | 100% |
| Video poker | 10% to 20% |
| Blackjack (RNG) | 5% to 10% |
| Roulette (RNG) | 5% to 10%, sometimes 0% |
| Live dealer roulette | 0% to 10% |
| Baccarat, craps | 0% to 10% |
Worked example. A $100 bonus with 30 times wagering requires $3,000 of qualifying turnover. If roulette contributes 10%, every $1 wagered counts as $0.10, so you must actually stake $30,000 on roulette to clear it. At the theoretical 2.70% edge, the expected cost of that turnover is about $810, vastly more than the bonus is worth. If roulette contributes 5%, the required turnover doubles again.
Terms that void bonuses. Read these before you bet a cent:
The honest summary: if you like roulette, deposit without a bonus, or take a bonus intending to play the games that actually contribute. Trying to clear a wagering requirement on roulette is usually the worst mathematical option on the menu. That is not a popular sentence in this industry, and it is still true.
Disclaimer. This information is general in nature and does not replace professional advice. Gambling carries a risk of financial loss. If it is causing you problems, contact Gambling Help Online on 1800 858 858.
⚠ Alert box, read this before any system. No mathematical staking system changes the house edge, alters the probability of the next spin, or guarantees a win. Progressions redistribute when you win and lose; they do not change how much you lose on average. Every roulette spin is statistically independent. Table maximums and a finite bankroll make any "double until you recover" plan incomplete by design. Set a money limit for the session before you sit down.
Roulette spins are independent events with a fixed negative expected value. Multiply a negative expectation by any staking pattern and it stays negative. Systems such as the Martingale reshape the distribution of outcomes in the short term, producing many small wins punctuated by rare large losses, which is why they feel effective. Behavioural gambling research consistently describes this as a control-related cognitive bias: a structured plan produces a sense of agency over an outcome the player cannot influence. Independent psychological measurement of this specific mechanism in roulette players remains limited in the published literature, so treat it as a well-described cognitive pattern rather than a settled quantitative finding. What is not in dispute is the arithmetic. The edge does not move.
1. Martingale. Double the stake on an even-money bet after every loss, then revert to one unit after a win.
| Consecutive losses | Stake (unit equals $10) | Cumulative outlay |
|---|---|---|
| 1 | $10 | $10 |
| 2 | $20 | $30 |
| 3 | $40 | $70 |
| 4 | $80 | $150 |
| 5 | $160 | $310 |
| 6 | $320 | $630 |
| 7 | $640 | $1,270 |
| 8 | $1,280 | $2,550 |
| 9 | $2,560 | $5,110 |
| 10 | $5,120 | $10,230 |
Risk: severe. Ten consecutive losses on red is far from exotic. On a single-zero wheel it turns up roughly once in every 750 sequences of ten spins, and by spin 10 you are risking $5,120 to win $10. On a table with a $1,000 maximum, the progression is mathematically broken at step 7.
2. Fibonacci. Stake according to 1-1-2-3-5-8-13-21 on a loss, then step back two places after a win. Risk: moderate. Growth is gentler than doubling, but recovery is slower and long losing runs still collide with the table max.
3. D'Alembert. Add one unit after a loss, subtract one unit after a win. Risk: lower. The flattest of the three, best suited to short sessions with a fixed loss limit, and still expected to lose 2.70% of total turnover.
Disclaimer. This information is general in nature and does not replace legal advice. Interactive gambling legislation changes; confirm the current position with the relevant regulator before acting.
The Interactive Gambling Act 2001 (Cth), the IGA, regulates online gambling supplied to people physically present in Australia. Its central prohibition targets the supply of "interactive gambling services", a category that includes online casino games such as roulette, blackjack, baccarat and online pokies. Online wagering on sport, racing and lotteries is treated differently and is lawfully offered by licensed Australian operators, creating a clear distinction between legally accessible online wagering products and prohibited interactive casino games. The prohibition attaches to the provider, not the individual player, and each contravention exposes providers to civil penalties and enforcement action.
"Interactive gamblers were significantly more likely to report at least one gambling-related harm (34.0%) compared with non-interactive gamblers (15.6%)." Interactive Gambling Amendment (Credit and Other Measures) Bill 2023, Bills Digest, Parliamentary Library (2023). https://www.aph.gov.au/Parliamentary_Business/Bills_Legislation/Bills_Search_Results/Result?bId=r7060
The practical consequence for Australian consumers is that roulette played online is typically supplied from offshore, where Australian consumer protections, dispute resolution and self-exclusion mechanisms may not apply.
The Australian Communications and Media Authority administers IGA complaints, investigates services, issues formal warnings, refers matters for civil penalties and asks internet service providers to block access to offending websites. In one reporting period, ACMA received 220 complaints about interactive gambling services, 180 of which (82 per cent) were considered valid complaints that could be investigated under the IGA. Between January and March 2023, ACMA completed 15 investigations covering 21 websites, identified 20 breaches of the IGA and issued 14 formal warnings. That cadence has continued quarter after quarter, alongside hundreds of blocked domains.
Blocking is not a wall. Mirror sites and rebranded domains reappear quickly. Its practical effect for players is disruption at the worst possible moment, since a blocked domain can strand a balance mid-verification.
The Commonwealth has progressively tightened credit-related protections. Legislation banning the use of credit cards, credit-related products and digital currency for online wagering was introduced in line with an April 2023 Government announcement, extending to interactive wagering a prohibition that had long applied to land-based casinos. The Bills Digest supporting that reform documented harms concentrated among interactive gamblers, including increased credit-card debt, the sale of personal property and family violence. The National Self-Exclusion Register, BetStop, also became operational, allowing a person to exclude themselves from all licensed Australian online and telephone wagering providers for a period of three months up to a lifetime.
Casino roulette remains legal in licensed venues. Crown Melbourne, for instance, operates under Victorian Gambling and Casino Control Commission casino rules, has introduced carded play, a requirement to use a player card that supports time and loss limits, and offers cashless wallet functionality with account controls. Venues also maintain dress standards and, in Crown Melbourne's case, operate 24 hours a day except on Christmas Day, Good Friday and ANZAC Day, when entrances close from 4am to noon. Crown's public FAQ is blunt about the online question: Australian operators are prevented from offering online casino gambling.
If you play roulette on a casino floor, the same maths applies, with two differences worth noting. Minimums are usually higher, commonly $10 to $25, and single-zero wheels are not guaranteed. Check the wheel before you buy in.
Analyses of real player account data let researchers compare products by the behaviours they generate rather than by self-report. In a large operator dataset, indicators such as escalating deposit frequency, cancelled withdrawals, declined deposits and rapid session re-entry were examined across game categories, suggesting that roulette players showed more of these high-risk behaviours than players of other table games such as blackjack.
"Among table games, roulette was the most popular and the most closely associated with behavioural markers of harm in a sample of 100,000 Unibet users." Delfabbro et al. (2023), behavioural markers study, Kindred/Unibet UK dataset. https://doi.org/10.1007/s10899-023-10200-7
The Online Problem Gambling Behaviour Index (OPGBI) was developed to detect risk from account activity. Validation work compared the index against the Problem Gambling Severity Index with r = 0.68, indicating that behavioural measures captured by the OPGBI align strongly with established problem gambling severity metrics.
"Playing gaming machines, roulette and horse-race betting were positively correlated with problem gambling on the PGSI in a sample of 10,000 Croatian online gamblers." OPGBI validation study, Croatian online gamblers sample (n = 10,000), 2023 to 2024. https://doi.org/10.1007/s10899-023-10200-7
Between-session chasing, meaning how quickly a player returns after a result, is measurable from timestamps rather than inferred.
"Roulette was associated with returning 13.59% faster after a loss and 8.23% faster after a win compared with slots (p < 0.001)." Zhang et al. (2024), between-session chasing study, PlayNow.com eCasino, British Columbia; n = 15,544. https://doi.org/10.1007/s10899-024-10300-1
Notably, the same body of work found faster returns after wins as well as losses, giving limited evidence for loss chasing but clear evidence for win chasing. For roulette specifically, the practical takeaway is that both outcomes pull the player back to the table sooner, which is why a pre-set session limit is more protective than a post-hoc decision.
Message-based interventions have been tested experimentally on virtual roulette. In one randomised design, one group saw a responsible-gambling slogan displayed during play, a second saw it before play, and the third group played without any slogan present.
"Regardless of whether the message was present, participants placed a similar number of bets and spins and spent a comparable amount of time playing roulette." Randomised online experiment, "Take Time to Think", virtual roulette, n = 1,500 (2023). https://doi.org/10.1080/14459795.2023.2200000
The implication is not that responsible gambling is futile, but that passive slogans underperform structural tools such as deposit limits, loss limits, time-outs, mandatory breaks and self-exclusion.
Australia's most recent national pilot found that 65.1 per cent of Australian adults took part in at least one form of gambling during the reference period. Lotteries reached 52.7 per cent of adults, poker machines 19.8 per cent and sports betting 12.5 per cent, while 15 per cent of adults experienced at least one gambling-related risk of harm. Casino table games including roulette sit well below pokies and lotteries in participation, which is exactly why product-level risk data, rather than participation alone, matters for a game like roulette.
Source: National Gambling Prevalence Study Pilot 2024, AGRC / AIFS (2024). https://aifs.gov.au/agrc/research/national-gambling-prevalence-study-pilot-2024
Younger cohorts meet gambling mechanics through routes that did not exist a decade ago, including loot boxes, social casino apps and simulated wheels, which can introduce young people to gambling-like mechanics before they touch real-money products.
"Young adults are particularly active in sports betting, fantasy sports and esports wagering; the convergence of gaming and gambling creates risks of early involvement." Suomi et al. (2024), Young Adults Gambling Online in the ACT, ACT Commission for Gambling and Racing. https://www.gamblingandracing.act.gov.au/research
Work through this list in order. Any failed item is a reason to stop.
1. Licence. Locate the licence number and issuing authority in the site footer, then verify it in the regulator's public register. The Malta Gaming Authority, the UK Gambling Commission and Curaçao's licensing body all publish searchable registers. Confirm that the operating company name and the domain match the register entry, because brand and entity mismatches are a common red flag.
2. Independent testing. Look for a traceable RNG certificate or audit report from a testing laboratory such as eCOGRA or iTech Labs, and for live studios, evidence that physical wheels are inspected. A logo is not a certificate; the report or certificate number should be verifiable.
3. Game providers. Recognised roulette suppliers publish game rules, RTP and volatility inside the game's info panel. If the info panel is missing or the RTP is unstated, treat the table as unverifiable.
4. Published limits. Minimum and maximum bets per table, and maximum payout per game or per session, should be stated plainly. Hidden limits are a dispute waiting to happen.
5. Withdrawal terms. Confirm daily, weekly and monthly caps, pending periods, fees, verification holds and whether limits differ by payment method.
6. Table variety. The best roulette offering includes both single-zero RNG tables and multiple live tables at different limits, ideally with a French table applying La Partage, plus roulette demo access so you can test the interface before funding anything.
7. Responsible gambling tooling. Deposit limits, loss limits, session timers, reality checks, time-outs and self-exclusion should be reachable within two clicks of account settings.
8. Support. Test live chat with a specific question, for example "does this roulette table apply La Partage?", before depositing. Response quality pre-deposit predicts response quality post-dispute.
9. Payments. Deposits repeatedly declined by your bank or card issuer are a signal about the operator's status, not a technical glitch.
10. Terms readability. If the T&Cs cannot be located, downloaded or read in full, the offer is not assessable, and an unassessable offer is not worth taking.
Roulette is entertainment with a known, non-negotiable price. Treat that price as the cost of the entertainment, and use the structural controls that evidence supports:
Where to get help (free, confidential, 24/7):
Gambling is for adults only, 18 and over in Australia, with other jurisdictions applying 18 or 21 thresholds. Always confirm the law that applies where you are physically located before playing for money.
| Term | Meaning |
|---|---|
| Pill | The roulette ball (nylon, phenolic resin or moulded acetal) |
| Dolly | Marker placed on the winning number after the spin |
| Rake | Long-handled tool used to collect losing chips and pay winners |
| Inside bets | Wagers on the numbered grid (straight, split, street, corner, six line) |
| Outside bets | Wagers outside the grid (red/black, odd/even, high/low, dozens, columns) |
| Manque / Passe | French for low (1 to 18) and high (19 to 36) |
| Pair / Impair | French for even and odd |
| Rouge / Noir | French for red and black |
| La Partage | Half the even-money stake returned when zero wins |
| En Prison | Even-money stake held for one further spin when zero wins |
| Racetrack | Oval wheel-order layout used for announced or call bets |
| Voisins du Zéro | 17-number sector around zero |
| Tiers du Cylindre | 12-number sector opposite zero |
| Orphelins | The 8 numbers outside the two main sectors |
| House edge | Long-run casino advantage as a percentage of turnover |
| RTP | Return to player, 100% minus the house edge |
| OCR | Optical recognition used to read live wheel results |
| Table max | Maximum permitted stake per bet, the ceiling that breaks progressions |
| Game weighting | Percentage of a wager that counts towards bonus wagering |
| BetStop | Australia's National Self-Exclusion Register for licensed online wagering |
If you take one thing from this guide, take the wheel choice: single zero, French rules where you can find them, a stake size decided before you sit down. Compare tables and terms calmly, and use the limit tools if play stops feeling like entertainment. Help is available on 1800 858 858, any hour.