← The Lab
Deep Dive

We tested Fibonacci Dozens on 10,000 players — here's what happened

October 21, 2026 7 min read

Fibonacci Dozens shows up again and again in roulette videos: steady wins, a progression everyone recognises, and a promise that the dozens are the "safe" way to play. We built it level by level and ran it through our full simulation. Here is what 10,000 players — and an endurance test — tell us.

The system

Fibonacci Dozens bets on the dozens, a 2-to-1 bet, and raises the stake after losses following the Fibonacci sequence: each step is the sum of the two before it. In the version we tested, ten levels take the total stake from $5 to $275 a spin:

Level12345678910
Stake$5$5$10$15$25$40$65$105$170$275

Play every level once and you've staked $715 — 143 times the first bet. We simulated the system with a $1,000 bankroll. You can see every level on the table on the Fibonacci Dozens system page.

The results: 10,000 players, three goals

Each simulated player starts with $1,000 and plays until they either reach the profit goal or bust — can no longer cover the next bet.

GoalEuropean win rateBust riskAvg spinsAvg result
+$5093.2%6.8%15−$10.92
+$10088.0%12.0%27−$22.15
+$20078.9%21.1%48−$43.75

At first glance it's impressive: more than nine in ten players reach +$50, and they get there in about 15 spins. But look at the last column. Averaged over all 10,000 players, the result is negative at every goal.

How can 88% of players win and the average still lose? Because the winners each take $100, while the 12% who bust lose their $1,000. Many small wins, a few big losses — the house edge is still there, just packed into the rare bad session.

On an American wheel

With a double zero, every number moves the wrong way: 92.1% reach +$50 (bust risk 7.9%), 86.4% reach +$100 and 76.2% reach +$200, with average results of −$22.11, −$38.20 and −$76.49. At +$200, the average loss is almost twice as large as on a European wheel. More in European vs American roulette.

The endurance test: what happens if you keep playing

Goals end sessions early. To see what happens when they don't, we let 1,000 players run the system for a fixed number of spins with no profit target:

Spins playedStill in the gameMedian bankroll
5082%$1,180
10066.8%$1,265
20051.5%$935

After 200 spins, almost half the players have busted — they could no longer cover the next bet — and the typical player is below where they started. The survivors who did well did very well — a lucky tenth had $1,900 or more — but you can't choose which group you'll be in.

What the numbers mean for you

It is a short-session system. Fibonacci Dozens does what it's built for: it reaches a small goal quickly and very often. Its strength is entirely in the first few dozen spins.

Keep the goal small. Moving from +$50 to +$200 triples the bust risk, from 6.8% to 21.1%.

Respect the price tag. One full losing run costs $715. If that number makes you uncomfortable, this system isn't for your bankroll.

Stop when the session ends. The endurance curve is the clearest argument for stop rules we've seen: the longer you play, the more the house edge wins.

Our verdict

Fibonacci Dozens is one of the better-behaved systems in our catalog for short sessions on a European wheel. It is not a way to make money: averaged over everyone, the result is negative at every goal, and it gets worse the longer you play. Play it for a small, defined goal, with a bankroll you can afford to lose — or try it in Play it first and see how a losing run feels without risking a cent.

Next in The Lab

The only roulette strategy that's mathematically optimal — and why nobody plays it

Proven in 1965, faster than any system, and better at reaching a goal. Meet Bold Play — the benchmark every system on this site is now measured against.

Read next →