| Spec | Value |
|---|---|
| RTP | 97% — fixed by Spribe, operators cannot lower it (iTech Labs cert IL-A2-180417-1) |
| Game type | Crash / multiplier curve — not a pokie; every target pays the same −3% EV |
| The distribution | Mean crash 4.57x (250,000-round community sample), modal crash ~1.89x — most rounds die under 2.00x while the average says otherwise; survival at 2.00x = 48.5%, at 1.50x = 64.67%, at 5.00x = 19.4%, at 10.00x = 9.7%; ~3% of rounds crash at exactly 1.00x, an instant loss no target avoids |
| The formula | P(reaching target T) = 0.97 / T; EV per round = −A$0.03 per A$1 staked at every T — only σ moves with the target |
| The structure | Two bet slots per round, each with its own stake and auto-cashout — the playbook’s core-plus-sleeve shape lives there |
| Bet | A$0.10-A$100 per slot; casino-imposed max win typically A$10,000 per round |
The winning line
Aviator comes down to one curve and two numbers. P(T) = 0.97/T charges the same 3-cent toll at every cash-out target; only the variance shifts. The modal crash lands near 1.89x, yet the mean boasts 4.57x, which means most rounds still die below 2.00x regardless of what that average suggests. Lock auto-cashout at 2.00x on slot 1, probe 20 rounds, arm a small 5x sleeve on slot 2, then walk at +50% or −60%.
Step 1 — Lock the cash-out before the plane moves
Pick the target at the table, not halfway up the climb. Slot 1 does the session’s heavy lifting: auto-cashout at 1.80x-2.20x (default 2.00x), staking 0.5-1.0% of the session roll, or A$1.00-2.00 from a A$200 stash. Keep the auto toggle locked ON. Slot 2 stays dark until Step 3.
Action: Before the first round, confirm the stake has 50-100 units of bankroll behind it (community-derived 95%-survival figure across 100 rounds).
Watch for: The bankroll-to-stake ratio, not whatever crashed last. The curve’s history is decoration. That ratio is load-bearing.
Decision branches:
- If the roll is under 50x the planned stake → target down to 1.50x (64.67% survival) and stake re-sized to 1%.
- If the roll clears 100x AND a stop-win already banked today → hold 2.00x, cut stake to 0.5%; more rounds, same edge, lower per-round σ.
- If a deposit bonus caps the bet at A$5 → A$5 max on slot 1, slot 2 stays dark until the bonus clears; voided punts have cooked more sessions than any crash streak.
Step 2 — Probe 20 rounds against the theory table
The distribution is memoryless, meaning each round stands by itself (Tversky & Kahneman 1971), so the probe predicts nothing. Its job is simpler: check whether today’s run resembles the baseline. At 97%, theory puts roughly 48% of rounds above 2.00x, around 11% above 5.00x, and about 3% instant-bust at 1.00x.
Action: Play rounds 1-20 at the locked baseline on one slot, with no tinkering. Log 2.00x clears, crashes below 1.50x and 5.00x clears. The history scroll keeps the last 20.
Watch for: Results against those three theory numbers. Ignore everything else in the scroll.
Decision branches:
- If ≥10 of 20 cleared 2.00x → baseline holds into Step 3; no pushing higher on “form” — the next round still pays −3%.
- If ≤6 cleared → cold variance; baseline drops to 1.50x for rounds 21-60 to keep round-count up through the dry stretch.
- If ≥4 cleared 5.00x → unusual top-end, NOT a signal; slot 1 holds at 2.00x, slot 2 arms next step.
- If 2+ instant-busts inside 20 → double the expected rate; stake halves (A$0.50-1.00) for the next 20 as short-term variance tax.
Step 3 — Arm the sleeve as a different bet, not a bigger one
From round 21, slot 1 remains the core: 1.50x or 2.00x based on the probe, staking 0.5-1.0% of the remaining roll. Slot 2 becomes the sleeve, with auto-cashout at 5.00x-10.00x and a 0.1-0.25% stake, or A$0.20-0.50 on A$200. Same −3% EV, different σ. Both slots also die on the same bust, so the combined amount at risk each round is their sum.
Action: Run both slots within the variance budget confirmed by the probe. The sleeve is a deliberately priced boom-or-bust punt with a 19.4% (5x) or 9.7% (10x) win rate.
Watch for: Slot 2’s stake starting to resemble slot 1’s. Two equal core bets double the amount at risk without shifting EV. The sleeve only does its job when it stays small and aims high.
Decision branches:
- If rounds 21-60 put the session +25% → slot 1 lifts to 2.50x, sleeve holds at 5.00x; house money absorbs the structure.
- If −25% → sleeve dies, slot 1 reverts to 1.50x; the aggressive side compounds drawdown faster than the core recovers it.
- If the sleeve lands its first 5.00x → banked, and sleeve stake drops to A$0.10 for 10 rounds; the hit changed the bankroll, not the next round’s EV.
- If 7+ consecutive crashes under 2.00x → a ~0.95% per-streak event (0.515^7), uncomfortable and expected; baseline holds, no martingale — doubling after losses just makes the next −3% bet bigger (PrimeDope 2024).
Step 4 — Exit on the written numbers, not the climbing curve
The curve is the most addictive reward loop in online play (Schull 2012). Every cash-out delivers a micro-reward; every near-miss flashes like a hot signal. The exits are there because the screen is designed to fight them, so hard-code those limits before round one, on paper when discipline is shaky.
Action: Set the stop-loss at −60% of the session roll, so A$200 exits at A$80 with no re-deposit. Set the stop-win at +50% on the 2.00x core OR +200% on any sleeve hit of 5x-plus. Whichever arrives first ends the session.
Watch for: Your hand reaching to disable auto-cashout “to ride one”. That switch is the giveaway. Tilt has taken the controls from variance.
Decision branches:
- If either exit number lands → tab closed the same minute; the next round’s EV is −3% forever.
- If 60 rounds pass with no sleeve hit AND the roll sits −20% → sleeve dies, core finishes at 1.50x, exit at the first +10% recovery from the low.
- If auto-cashout gets manually overridden mid-flight → session over immediately; the playbook just left the chair.
The maths that backs the playbook
It all comes from P(T) = 0.97/T and the fixed −3% EV attached to every target (Spribe Provably Fair Documentation). The game is selling different variance shapes, nothing more. The σ ladder puts numbers on them: ≈0.72 per unit at 1.50x, 1.00 at 2.00x, 1.98 at 5.00x, 2.96 at 10.00x and 6.90 at 50x, with the last figure matching high-vol pokie territory (derived from σ² = p(T−1)² + (1−p) − EV²). Drawdown scales roughly with (stake/bankroll) × σ × √N. At 2.00x, using 1% stakes across 100 rounds, a one-σ drawdown comes to ~10% of bankroll on top of 3% drift; double the stake and both figures double. Those are normal-approximation estimates, while the real distribution has fatter tails. The core-plus-sleeve split is the playbook’s only structural edge: 0.7% at 2.00x paired with 0.2% at 5.00x blends to σ ≈ 1.22 for the money at risk, compared with 1.98 when the same 0.9% sits entirely on the 5x target. Same amount at risk, ~38% less volatility, sleeve upside preserved. The behavioural trap has its own number as well. The 4.57x mean towers over the ~1.89x mode, so anyone planning around the average keeps busting on the typical result. Plan around the mode. Let the mean look after itself. The forced ~3% instant-bust at 1.00x, roughly once every ~33 rounds, does not care either way.
What breaks the playbook
- Skipping the probe. Staking 2% at 2.00x from round one chases variance into bust streaks — 30-50% of bankroll inside 40 rounds is the going rate.
- Running slot 2 as a doubled core. Equal stakes at equal targets doubles at-risk at unchanged EV; the sleeve works only smaller and higher.
- Pattern-reading the scroll. “Five reds, a big one’s due” is the gambler’s fallacy on a memoryless curve (Croson & Sundali 2005) — P(2x) is 48.5% after any history whatsoever.
- Disabling auto-cashout mid-flight. The whole structure hangs on the lock; manual “riding” costs full sessions inside ten rounds.
- Downloading a “predictor” APK. Fair dinkum malware (BOOM investigative reporting 2025) — provably fair outcomes are mathematically unpredictable; the price is credentials, KYC documents and the deposit balance.
- Ignoring bonus max-bet rules. AU-facing brands cap bonus play at A$5-10 per round, and a dual-slot bet can read as one combined punt — one oversized round voids the lot.
The cut-off
Stop-loss: −60% of the session roll. A$200 exits at A$80, with no re-deposit and no top-up.
Stop-win: +50% on the core or +200% from a 5x-plus sleeve hit. The first one reached ends the session.
Sleeve law: kill it after 60 hitless rounds at −20%. It returns next session only.
Override rule: one manual cash-out override makes that round the session’s last, whether it wins or loses.
Strategy by Dave Holloway — Floor Veteran. Last updated: 16 Jul 2026. 18+. Gamble responsibly. Gambling Help Online — 1800 858 858 | Lifeline 13 11 14.
Aviator
Specs
| Spec | Value |
|---|---|
| Provider | Spribe |
| Released | 2019-02-15 |
| Mechanic | Crash game (single rising multiplier) |
| RTP | 97% (fixed across all operators) |
| Min bet | A$0.10 |
| Max bet | A$100 to A$250 (operator-dependent) |
| Multiplier cap | Theoretically unbounded; practical ceiling = casino max-win-per-round (~A$10,000) |
| Crash distribution | Geometric / exponential; mean ~4.57× (community sample, 250k rounds) |
| Volatility | Low at 1.5× target (σ ≈ 0.72), extreme at 100× target (σ ≈ 9.80) |
| Bet slots | 2 simultaneous bets per round |
| Certifications | iTech Labs (IL-A2-180417-1), GLI, BMM Testlabs |
| Theme | Aviation / minimalist HUD |
Mechanic
There are no reels or paylines here. No symbols either. A red prop plane takes off while a multiplier counter climbs from 1.00×. Punters cash out before the plane flies away, banking stake × multiplier the moment they tap. Miss the cashout and, well, the round is lost.
- One round runs for roughly 8 to 30 seconds of active flight, followed by a ~5 second betting window.
- The multiplier rises continuously. Its movement is non-linearly shaped by a geometric distribution.
- The crash point is locked cryptographically before betting opens. Neither Spribe nor the casino can, honestly, nudge a round once it is running.
- Players can place up to two bets in one round. Each bet can have its own auto-cashout target.
- By design, roughly 3% of rounds crash at exactly 1.00×. The house edge is built into that instant-crash portion.
Game rules
How a round works
A round has a betting phase and a flight phase, then the crash closes it. The player only has agency during the first two. Once the plane flies away, the round is over. Any bet still live is forfeit.
- The betting phase opens for ~5 seconds. Players place 1 or 2 bets and can, of course, set an auto-cashout target on either one.
- Then the flight begins. The multiplier starts at 1.00× and rises continuously along the visible curve.
- The crash event fires at the pre-determined point. The plane animates off-screen. “FLEW AWAY!” then overlays the final multiplier.
- Any bet that has not cashed out before the crash loses its full stake.
- Up top, there’s a scrollable history bar showing the last ~15 round results. Red marks <1.5× and blue covers 1.5 to 2×, while green shows >2×.
Cash-out mechanics
Cashing out is the only lever the player controls. Aviator gives you manual taps or pre-set auto-cashout targets. The dual-bet feature, mind you, lets a punter run two strategies on the same round.
- Manual cashout: tap the green Cash Out button at any point during flight. The win equals stake × multiplier at that exact moment.
- Auto-cashout: pre-set a multiplier target, with 1.01× as the minimum and no fixed maximum. The bet cashes out automatically once the live multiplier reaches that target.
- Dual bet: two separate betting panels sit side by side. Each has its own stake and its own auto-cashout target.
- Most community presets tend to sit around 1.50× or 2.00×, with 3.00× another common pick. Around 60% of tracked bets use auto-cashout (AviatorSmart, 250k-round sample).
- Both bets share the same round. If the plane crashes before either one cashes out, both lose.
Crash distribution
The crash point follows a geometric distribution shaped by the 97% RTP. The maths is straightforward, honestly. The chance of the plane reaching multiplier T is 0.97/T. Expected value at every target is exactly −3% of stake.
- P(reach 1.50×) ≈ 64.7% and P(reach 2×) ≈ 48.5%. P(reach 5×) ≈ 19.4% and P(reach 10×) ≈ 9.7%, while P(reach 100×) ≈ 0.97%.
- In a 250,000-round community sample, the mean crash multiplier was 4.57×. The modal cluster sat at ~1.89×.
- Multipliers above 5× show up in roughly 11% of rounds. Those above 10× appear in roughly 8.5%.
- Each round is independent. There is no “due” multiplier and no streak memory. The distribution is memoryless.
- The widely shared crash-point formula derived from Spribe’s algorithm is
crash = max(1.00, (100·e − h) / (e − h) × 0.97). Here,e = 2^52, whilehis parsed from the SHA-512 of the seed string.
Provably fair verification
Aviator’s provably fair system lets anyone confirm that a round’s outcome was locked before bets opened. Before each round, a server seed is hashed with SHA-256. That hash is then published. Three client seeds drawn from players in the round are folded in. After the crash, the server seed is revealed for verification.
- Pre-round: server seed → SHA-256 hash → published to all players through the green shield UI element.
- During the round, three client seeds are combined with the server seed.
- Post-round: the server seed is revealed. Anyone can compute SHA-512(server_seed + client_seed_1 + client_seed_2 + client_seed_3), parse the first 13 hex chars as integer h, then run the crash-point formula to confirm the multiplier.
- The shield icon in-game links to Spribe’s hosted verifier. Third-party tools such as AviatorStats and gamblingcalc mirror the same algorithm.
- Since the server seed is committed before betting opens, neither the operator nor Spribe can change the result later. Simple as that.
Rain feature
Rain is an operator-configurable promo system. The casino, or Spribe, drops free bets into the in-game chat at random intervals. Not every AU-facing operator switches it on, mind you.
- Free bets appear as a claim button inside the chat panel.
- Claims are first come, first served. A drop usually hangs around for only a handful of seconds before it is gone.
- Drop values are small, commonly A$0.10 to A$1.00 from what has been observed. Frequency is controlled by the operator.
- There is no fixed schedule. The feature is designed to keep the chat active and, of course, drive engagement.
- Most casinos exclude Rain from welcome-bonus wagering. Depending on operator terms, Rain bets pay out as cashable balance rather than bonus balance.
Tournaments (Aviarace)
Aviarace is Spribe’s branded leaderboard layer running on top of the base game. Operators decide whether to opt in. Prize pools are casino-funded, so availability can vary widely across AU-facing brands. Pool size varies too.
- The leaderboard can rank players by total wagered or highest multiplier cashed. Biggest single win is another metric used.
- Tournament windows can run for a few hours. Some stretch to a full week.
- Prize pools are operator-funded. Stake and BC.Game have run Aviator-specific events in 2025, as has 7Bit.
- Entry is automatic for anyone playing Aviator at a participating casino during the tournament window.
- There is no additional house edge on tournament play. The standard 3% cut applies whether the leaderboard is active or not.
Screenshots
Idle screen between rounds. The history bar across the top shows recent multipliers in red and blue bubbles, plus green ones. Both bet panels are primed at A$1 each.
Mid-flight at 6.10×. The red curve filling the lower frame tracks the live multiplier. Tap Cash Out now and, sure, the round closes at this exact figure.
The “FLEW AWAY!” state at 1.30×. Any bet still live at this point loses its full stake. About half of all rounds, honestly, end below 2×.
RTP context
Aviator ships at 97% RTP, and that’s the only tier in the wild. Spribe doesn’t expose lower-RTP variants to operators. There’s no per-casino dial to drop the cut either. For AU punters, this removes a class of risk that comes built into most Pragmatic pokies and plenty of NetEnt titles.
| Tier | Status | Where it runs |
|---|---|---|
| 97% | Default and only | All 5,000+ operators carrying Aviator, including every AU-facing offshore casino |
| <97% variants | None documented | No certificate, no operator implementation found |
Certification comes from iTech Labs (cert IL-A2-180417-1) and GLI, plus BMM Testlabs. The 3% house edge is hardcoded directly into the crash-point formula rather than a per-casino RTP file. That means the figure can’t drift between operators.
Spribe portfolio
Spribe builds a tight catalogue of instant titles and crash games rather than a giant slot library. Aviator is the flagship. The rest of the portfolio, to be fair, uses the same provably fair backbone.
| Title | Type | Notable trait |
|---|---|---|
| Aviator | Crash | 97% fixed RTP, dual bets, the genre’s standard |
| Mines | Instant grid | Player-set mine count; provably fair tile reveal |
| Plinko | Drop ball | Adjustable risk levels and row counts |
| HiLo | Card | Cumulative multiplier from sequential card calls |
| Dice | Instant | Roll-under target with player-set odds |
| Goal | Sports-themed | Path-pick with provably fair payouts |
| Hotline | Crash-adjacent | Multi-track multiplier with checkpoint cashouts |
There are no other Spribe-built crash competitors from the same studio. The head-to-head field includes JetX (SmartSoft) and Lucky Jet (1Win). Spaceman (Pragmatic Play) and BGaming Crash are in the mix too, each with different RTP rules and bet-slot counts.
AU casino availability
Spribe doesn’t hold an IGA 2001 licence, so Aviator reaches AU punters through offshore-licensed casinos in Curacao and Malta, plus Kahnawake. The list below covers AU-facing brands where Aviator was directly confirmed at the time of writing. None are domestically licensed in AU.
| Casino | Currency | Notes |
|---|---|---|
| Stake | Crypto only (no AUD fiat) | Listed at stake.com/casino/games/spribe-aviator; high limits; 0 to 24h crypto withdrawals |
| BC.Game | Crypto (150+ altcoins) | Demo available pre-deposit; fast crypto withdrawals |
| Rainbet | Crypto primary | Listed at rainbet.com; standard offshore bonus terms |
| 7Bit Casino | AUD + crypto | AUD fiat withdrawals 2 to 5 business days; max-bet rule during bonus play |
| KatsuBet | AUD + crypto | KYC before first withdrawal |
| Casinonic | AUD | KYC before first withdrawal; AU-targeting brand |
| Roobet | Crypto primary | Also runs BGaming AviaMasters, a different crash game; don’t mix them up |
| Crown Play | AUD + crypto | A$750 welcome bonus on file; AU-targeting brand |
| Richard Casino | AUD + crypto | A$5,000 package; AU-targeting brand |
| Golden Crown | AUD + crypto | A$1,000 bonus; AU-targeting brand |
The UKGC suspension of Spribe from Oct 2025 to March 2026 came down to a hosting licence gap, not a game-integrity issue. It never affected AU access because, of course, the AU market runs through offshore operators.
FAQ
Q: What’s Aviator’s actual RTP, and why is it locked at 97%?
A: Straight up, it is 97% across every casino carrying the game. Spribe builds the 3% cut directly into the crash-point formula. Operators therefore can’t ship a low-RTP version like they can with most Pragmatic pokies.
Q: How is auto-cashout different from manual?
A: Auto-cashout fires the bet at a pre-set multiplier even when the player looks away. Manual needs a live tap. EV is identical at −3% per round. Auto does, however, remove reaction-time risk and those tilt-driven “just one more second” holds.
Q: Can the maximum multiplier really hit 1,000,000×?
A: Technically, yes. The algorithm has no programmatic ceiling. The highest recorded crash sits at 2,586,812× (SlotCatalog). The practical cap is the casino’s max-win-per-round rule, usually A$10,000 per bet.
Q: Is the “Aviator Predictor” app legit?
A: Not even close. Provably fair crashes are mathematically unpredictable because each round’s seed is locked before bets open. The APKs floating around Telegram and YouTube are malware. You are looking at credential stealers or miners, and some run up-front “deposit to unlock” scams.
Q: Which AU-facing casinos restrict auto-cashout?
A: None block auto-cashout directly. Most do apply max-bet-per-round caps of ~A$5 during active deposit-bonus wagering. The dual-bet feature creates an edge case, mind you. Some operators count two A$5 bets as one A$10 bet against the cap. Others don’t. Check support before high-stakes bonus play.
Q: What’s provably fair and how do I check?
A: Well, each round’s outcome is locked through a SHA-256 hash before betting opens. After the round, it can be verified through SHA-512 using the revealed server seed and three client seeds. Tap the green shield icon in-game for Spribe’s hosted verifier.
Q: Is Aviator a slot? Why is it on a slot site?
A: Not really. There are no reels or paylines, and no symbols either. It’s a crash game. Still, it sits in the same offshore-casino lobbies as pokies, and the bankroll maths punters need is close enough that the strategy desk treats it alongside high-vol slots.
Q: Does dual-bet improve the EV?
A: No. Each bet carries its own independent −3% EV. Running two A$1 bets at different auto-cashout targets is a variance-management play, not an edge. Combined EV equals −3% of total staked. That is the same as a single A$2 bet.
Last updated: 03 Jun 2026
Reviewed by Mark Henderson, Senior Slots Analyst
Demo embed · played in fun-money mode. No real-money play here.