๐Ÿ“Š
LiveCasinoData

Real-Time Statistics

StatsReportsRTPFairnessOpen Data
Live Data
Menu
๐ŸŽฎAll Games
Game Categories
Popular Guides
๐Ÿ“–How to Play๐ŸŽฏStrategy๐Ÿ“ŠRTP๐Ÿ’กTips
Popular Games
Crazy TimeLightning RouletteMonopoly LiveMega Ball
๐Ÿ“ˆStatisticsโš–๏ธCompare Games๐ŸŽฏRTP Monitorโš–๏ธFairness Index๐Ÿ“ฆOpen Dataset
Live Data Active
Auto Roulette
GuideHow to PlayStrategyRTP & OddsBonusesLive Stats
Home/Games/Auto Roulette/Bonuses

Auto Roulette Streaks, Sectors & Hot/Cold Numbers: What 103,221 Real Spins Show

By LiveCasinoData editorial team ยท Updated 2026-08-28 ยท 13 min read

Evolution's Auto Roulette has no bonus round and no multipliers; the wheel is a plain 37-pocket European layout paying 97.30% on every bet. Over 103,221 tracked spins the hottest number (10) hit 2.78% of the time and the coldest (21) hit 2.62%, against a theory of 2.70% for all of them. The longest red streak was 16 spins, the longest zero drought 396 spins, and none of it shifted the odds on the next spin.

Key takeaways

  • Auto Roulette has no bonus round, no Lightning-style multipliers and no side bets; the tracker's multiplier field is always 1.
  • Across 103,221 spins in 90 days, every one of the 37 numbers landed between 2.62% and 2.78% of the time, a spread of just 167 hits between the hottest and coldest number.
  • The longest colour streak we logged was 16 reds in a row; the maths expects a run that long roughly twice per 100,000 spins.
  • Zero went missing for 396 consecutive spins in a drought that ended on 2026-08-23, yet its average gap over the window was 36.6 spins, almost exactly the theoretical 37.
  • Voisins du Zero landed 45.62% of the time (theory 45.95%), Tiers 32.61% (theory 32.43%) and Orphelins 21.77% (theory 21.62%), so no wheel sector is favoured.
GuideHow to PlayStrategyRTP & OddsBonusesLive StatsLive tracker
On this page
  1. Why is there no bonus round in Auto Roulette?
  2. How evenly do the 37 numbers actually land?
  3. What is a hot number in Auto Roulette, and does it matter?
  4. How long do colour streaks get in Auto Roulette?
  5. How often does the same number repeat?
  6. What is the longest zero drought in Auto Roulette?
  7. Which number went missing the longest?
  8. Do Voisins, Tiers and Orphelins sectors run hot?
  9. Why hot and cold numbers cannot predict the next spin
  10. How to read the streak and sector widgets on the tracker
  11. Where the numbers on this page come from
  12. Frequently asked questions

Live data: last 90 days of Auto Roulette

Open the live tracker โ†’
Rounds (90d)
103,953
Rounds tracked in total
276,895
since 2025-12-28
Biggest multiplier
1x
2, 2026-08-27

37 possible results tracked. The five most and five least frequent in this window are shown; the full breakdown is on the live tracker.

ResultObservedTheoreticalDifferenceAvg multiplierMax
hot102.78%2.7%+0.081x1x
332.78%2.7%+0.081x1x
252.77%2.7%+0.071x1x
82.77%2.7%+0.071x1x
312.77%2.7%+0.071x1x
cold212.61%2.7%-0.091x1x
62.63%2.7%-0.071x1x
222.64%2.7%-0.061x1x
292.64%2.7%-0.061x1x
122.65%2.7%-0.051x1x

Source: LiveCasinoData tracker, refreshed every 10 minutes. Figures on the live page move as new rounds arrive.

Auto Roulette has no bonus round. There is no lightning strike, no multiplier board and no side bet; the compressed-air launcher fires the ball into a standard 37-pocket European wheel and the number that comes up pays at standard odds. What players search for on this game instead is streaks, wheel sectors and hot or cold numbers, so that is what this page covers. Our tracker has logged 275,852 Auto Roulette spins since 2025-12-28, and the last 90 days alone hold 103,221 of them. The live tracker shows the current streaks and hot numbers; the Auto Roulette guide covers the rules.

Why is there no bonus round in Auto Roulette?

Because the product is deliberately the plainest form of live roulette Evolution offers. A physical wheel in the Riga studio is spun automatically around the clock, with no dealer, no presenter and no chat, and the result is the winning number and nothing more. The tracker records that number as a string from "0" to "36" together with its colour, and the multiplier field is hard-coded to 1 on every single row.

That simplicity is the point. Every bet on the layout, from a straight-up number at 35:1 down to red or black at 1:1, carries the same 2.70% house edge and the same 97.30% theoretical return, according to Evolution's published rules. There is no bonus spot to chase and no multiplier to wait for, which is why the interesting data on this table is about sequences rather than payouts.

If you want a roulette table with a bonus mechanic, Evolution's Lightning Roulette and XXXtreme Lightning Roulette add random multipliers to straight-up bets in exchange for a reduced base payout. Those are different tables with different maths, tracked separately.

How evenly do the 37 numbers actually land?

Very evenly. Over the last 90 days, 103,221 spins, every one of the 37 pockets came up between 2.62% and 2.78% of the time. The theoretical share for each is 1/37, or 2.70%.

The table below shows the extremes. The difference between the hottest and coldest number is 167 hits over more than a hundred thousand spins, which is well within what random variation produces on a fair wheel.

Number Hits (90 days) Share Theory
10 (hottest) 2,870 2.78% 2.70%
25 - 2.77% 2.70%
33 - 2.77% 2.70%
1 - 2.77% 2.70%
31 - 2.76% 2.70%
8 - 2.76% 2.70%
22, 29, 7, 12 - 2.64% 2.70%
6 - 2.63% 2.70%
21 (coldest) 2,703 2.62% 2.70%

Colour and group bets tell the same story. Red landed 48.63% of the time, black 48.64% and green 2.73%, against a theory of 48.65 / 48.65 / 2.70. The three dozens came in at 32.49%, 32.29% and 32.49%, and the three columns at 32.55%, 32.41% and 32.31%, all against a theoretical 32.43%. The odd bet hit 48.59% and 1-18 hit 48.67%.

None of those gaps is large enough to mean anything. A 0.02 point difference on an even-money bet over 103,000 spins is noise, and the RTP page explains why we do not turn these shares into an "observed RTP" figure.

What is a hot number in Auto Roulette, and does it matter?

A hot number is simply the pocket that has come up most often over whatever window you are looking at, and it matters only as a description of the past. As of the 90-day snapshot the hottest number was 10, with 2,870 hits. The coldest was 21, with 2,703. That is the entire difference: 167 spins across a window of 103,221.

Hot and cold lists look meaningful because they are always non-empty. Some number must be top and some number must be bottom, and with 37 candidates the spread between them will never be zero. Shorten the window to the last 100 or 500 spins, which is what most in-game "hot numbers" panels use, and the spread gets proportionally wider because the sample is smaller. A number that has hit 6 times in 100 spins looks scorching at 6%, but the expected count is 2.7 and a count of 6 is unremarkable.

The test of whether hot numbers carry information is whether they stay hot. On a physical wheel with no memory, they do not. The number at the top of the list next week is drawn from the same 37 candidates as the number at the top this week, and our own data shows no pocket drifting away from its 2.70% share as the sample grows.

How long do colour streaks get in Auto Roulette?

The longest red run in our last 90 days was 16 consecutive spins, and the longest black run was 13. Zero came up three times in a row once. The average run of a single colour, counting from the spin that starts it, was about 1.95 spins.

The probability of a colour streak of length k, once the first spin of that colour has landed, is (18/37) raised to the power of k minus 1. A few reference points:

Streak length Probability from a given start Expected occurrences per 100,000 spins (approx.)
5 in a row 5.6% thousands
8 in a row 0.65% hundreds
10 in a row 0.15% dozens
13 in a row 0.017% a handful
16 in a row 0.0021% about 2

Our observed maximum of 16 lines up with the bottom row: a run that long should appear roughly twice in 100,000 spins, and we have 103,221 in the window. The 13-run on black is likewise comfortably inside expectation. Neither says anything about the wheel beyond "it is behaving like a wheel."

The trap with streaks is reading them as a cue. A red run of 10 is not a sign that black is "owed", nor is it a sign that red is "on". The 11th spin is 18/37 red, 18/37 black and 1/37 green, exactly as the first one was. The strategy page works through what that independence does to Martingale-style colour betting.

How often does the same number repeat?

On 2.6% of spins, in line with the 2.70% theory. The same number came up twice in a row 2,716 times across the 90-day window, three times in a row 85 times, and the longest identical run was four consecutive hits of one number.

The arithmetic is straightforward. Whatever number just landed, the chance it lands again on the next spin is 1 in 37, or 2.70%, because the ball has no idea what happened last time. Three in a row from a given start is 1/37 squared, which is 0.073%, and 85 such runs in 103,221 spins is what that rate produces. A four-in-a-row is rare enough that seeing exactly one is about right.

Repeats feel dramatic in the moment, and the in-game "last numbers" strip makes them easy to spot. They are not a pattern in any predictive sense; they are the tail of a distribution that must produce repeats at 1 in 37 whether or not anyone is watching.

What is the longest zero drought in Auto Roulette?

396 spins, in a drought that ended on 2026-08-23. Two other long dry spells of 277 and 270 spins both ended on the same day, 2026-07-25. Against those outliers the average gap between zeros over the 90 days was 36.6 spins, with a median of 24, against a theoretical mean gap of 37.

That pairing of a median of 24 and a mean of 36.6 is the shape you always get from a random gap distribution: most waits are short, a few are very long, and the long ones pull the average up. The probability that zero stays away for N spins is (36/37) to the power N. For N = 37 that is 36.3%, so more than a third of the time you wait longer than the "average" 37 spins. For N = 100 it is 6.5%. For N = 200 it is 0.42%. For 396 it is tiny per starting point, but there are 103,000 starting points in the window and 37 numbers, so a 400-spin drought for some pocket is close to inevitable.

Zero attracts special attention because it is green, it is the only pocket the even-money bets lose on, and its absence is easy to notice. The tracker's zero drought widget shows the current gap and the longest recorded one, and the number to keep in mind is that on the spin after a 396-spin drought, zero's chance was 1 in 37, the same as it was on spin 1 of that drought. Betting zero because it is "overdue" costs 1 unit per spin at a 2.70% hit rate regardless of history.

Which number went missing the longest?

Number 29, which failed to appear for 469 consecutive spins at one point in the last 90 days. That is the longest drought for any single number in the window, and it is longer than the zero drought simply because there are 36 other numbers competing for the record.

The average gap between hits of any given number was 37.0 spins across the period, which is as close to the theoretical 37 as a real sample gets. The per-number probability of a 469-spin absence is about 0.0003% from a given start, but with 37 pockets each having tens of thousands of opportunities to begin a drought, at least one drought past 400 is the expected outcome, not an anomaly. The live stats page explains how the tracker computes these gaps and what its drought widget does and does not show.

Do Voisins, Tiers and Orphelins sectors run hot?

No sector was favoured in our data. The three classic wheel sectors, which cover the physical wheel rather than the layout, landed almost exactly in proportion to the number of pockets they contain.

Sector Pockets Chips (racetrack) Theory Observed (90 days)
Voisins du Zero 17 9 45.95% 45.62%
Tiers du Cylindre 12 6 32.43% 32.61%
Orphelins 8 5 21.62% 21.77%

Voisins ran a third of a point below its pocket share, Tiers and Orphelins a fraction above. On a sample of 103,221 spins these are the ordinary wobbles of a fair wheel; no sector is warmer than its pocket count says it should be.

Sector bets deserve a specific warning because of how they are sold. The racetrack makes Voisins look like a bet "on the zero's neighbourhood", and on a wheel with a dealer that idea sometimes shades into talk of dealer signatures and wheel bias. Auto Roulette has no dealer, the launcher is mechanical, and the pocket order (0, 32, 15, 19, 4, 21, 2, 25 and so on around to 26) is the standard European sequence. A sector bet is just a bundle of straight-ups and splits placed with fewer clicks; each of its chips carries the same 2.70% edge as any other chip on the table. The how to play page shows where the racetrack sits in the client and how the chips are distributed.

Why hot and cold numbers cannot predict the next spin

Because the wheel has no memory and no mechanism that could give it one. Each spin is a fresh physical event: the launcher fires, the ball decelerates, it drops into one of 37 pockets, and the result is not written anywhere the next spin can read.

The evidence in our data points the same way from three directions. First, the per-number shares converge on 2.70% as the sample grows, which would not happen if some pockets were structurally favoured. Second, streak lengths and repeat counts match the (18/37) and (1/37) formulas rather than exceeding them, which would not happen if streaks fed on themselves. Third, droughts end at random rather than at a predictable point, and the spin that ends them is no different from the spins that did not.

There is a version of the hot-number idea that sounds more sophisticated: that the wheel might have a slight physical bias, and a big enough sample would reveal it. It is not impossible in principle, but our 90-day sample shows a top-to-bottom spread of 167 hits across 103,221 spins, which is what an unbiased wheel produces. Anyone claiming to have found a bias from the in-game panel of the last 500 spins is looking at noise with a sample roughly 200 times smaller than ours.

How to read the streak and sector widgets on the tracker

The Auto Roulette tracker shows the current colour streak, the current gap since the last zero, hot and cold numbers over the selected window, colour and dozen shares, the three sector shares, and the last spins in order. Everything on this page is a snapshot of those widgets taken as of 2026-08-27, so the live figures will have moved on.

Use the widgets to calibrate rather than to forecast. Knowing that the longest zero drought we have logged is 396 spins tells you that a 150-spin wait is nothing unusual; knowing that the longest red run is 16 tells you that 7 reds in a row is a routine Tuesday. The rounds-per-day figure (about 1,690, one spin every 51 seconds or so) is the number that actually matters for a session, because the pace of an automated wheel is what drives cost. Seventy spins an hour at any fixed stake is a lot of 2.70% edges.

One caveat on the data: our collector was offline from 2026-06-30 to 2026-07-09, a nine-day gap inside the 90-day window. Shares, streaks and droughts are unaffected because they are computed on the spins we have, but spins-per-day for those dates should be ignored.

Roulette is designed so that the house keeps 2.70% of every unit wagered over the long run, whatever you know about the last thousand spins. If tracking hot numbers is turning into chasing them, the observatory page has tools and links for keeping play within limits.

Where the numbers on this page come from

Wheel layout, pocket order, payouts and the 97.30% theoretical return are from Evolution's official game rules as shown in the Auto Roulette client help and on the Evolution roulette page. Every observed figure, from the 2,870 hits on number 10 to the 396-spin zero drought, comes from the LiveCasinoData Auto Roulette tracker: 275,852 spins logged since 2025-12-28, with the 90-day window of 103,221 spins and the 30-day window of 50,658 spins taken as of 2026-08-27. The table is a physical wheel and the data is a live feed, so the current values on the tracker will differ from the snapshot here. Fairness checks across all tracked games are summarised on the fairness index.

Frequently asked questions

Does Auto Roulette have a bonus round?

No. Evolution's Auto Roulette is a standard single-zero European roulette wheel spun by a compressed-air launcher, with no dealer, no multipliers, no bonus game and no side bets. Every bet pays at the standard European rate (35:1 straight up, 1:1 on red or black) with a house edge of 2.70%. Games with a bonus element, such as Lightning Roulette, are separate Evolution tables tracked on their own pages.

What is the longest red or black streak in Auto Roulette?

In the last 90 days of our tracking, 103,221 spins, the longest run of red was 16 consecutive spins and the longest run of black was 13. A red run of exactly 16 has a probability of about 0.0021% from any given starting spin, which works out to roughly two such runs per 100,000 spins, so seeing one in our window is expected rather than remarkable.

What are the hot numbers in Auto Roulette right now?

Over the last 90 days the most frequent numbers were 10 (2.78%, 2,870 hits), 25 and 33 (2.77%), 1 (2.77%), 31 and 8 (2.76%). The coldest were 21 (2.62%, 2,703 hits) and 6 (2.63%). Theory says 2.70% for every pocket. The hot and cold lists change constantly and are shown live on our tracker at /games/auto-roulette, but they carry no information about the next spin.

How long can Auto Roulette go without a zero?

Our tracker's longest zero drought in the last 90 days was 396 spins, ending on 2026-08-23. Two other droughts of 277 and 270 spins both ended on 2026-07-25. The average gap between zeros over the same period was 36.6 spins, with a median of 24, against a theoretical average of 37. The chance of no zero in 396 spins is tiny per starting point, but across 103,000 spins a drought that size is not surprising.

Is the Voisins du Zero sector hot in Auto Roulette?

Not in any meaningful sense. Voisins du Zero covers 17 of 37 pockets, so it should land on 45.95% of spins. In our last 90 days it landed on 45.62%, a third of a point below theory. Tiers du Cylindre (12 pockets) came in at 32.61% against a theoretical 32.43%, and Orphelins (8 pockets) at 21.77% against 21.62%. All three sectors sit within normal variance of their pocket counts.

Published 2026-08-28 ยท Last substantive update 2026-08-28. Figures marked as observed come from our own tracker and change as rounds arrive; theoretical figures come from the provider's published rules. 18+. Gambling involves risk; play only with licensed operators.

DISCLAIMER: LiveCasinoData.com provides statistical information for educational and entertainment purposes only. We do not guarantee the accuracy, completeness, or reliability of any data presented. Past results do not predict future outcomes. Gambling involves risk - never bet more than you can afford to lose. This website is not affiliated with, endorsed by, or connected to any casino operator or game provider. All game names and trademarks are property of their respective owners. You must be 18+ (or legal gambling age in your jurisdiction) to access this site.

About

  • About Us
  • Contact
  • FAQ
  • Blog
  • Analytics Reports
  • Editorial Team

Legal

  • Terms of Service
  • Privacy Policy
  • Cookie Policy
  • Disclaimer
  • DMCA Policy
  • Affiliate Disclosure

18+Responsible Gambling

  • Responsible Gambling
  • BeGambleAware.org
  • GamCare.org.uk
  • Gamblers Anonymous
  • GamStop

Tools

  • WordPress Widget
  • Telegram Bot
  • Telegram Features
  • Open Dataset
  • Research API
  • White Papers
  • RG Observatory
  • Fairness Index
  • RTP Monitor

Games

  • Crazy Time Stats
  • Monopoly Live
  • Lightning Roulette
  • All Games โ†’

Providers

  • Evolution Gaming
  • Pragmatic Play
  • All Providers โ†’
๐Ÿ“Š
LiveCasinoData

Real-Time Statistics Platform

18+Adults Only

ยฉ 2026 LiveCasinoData.com | All Rights Reserved

Game names and trademarks belong to their respective owners.