Does a Martingale on dozens work?
No. On 20,000 seeded European-wheel sessions, Dozen Martingale walked away $200 up 77.8% of the time and still lost $28.98 per session on average: −$2.55 for every $100 wagered, the house edge within sampling noise.
How Dozen Martingale is played
- Bet on the third dozen (25–36), starting at one unit.
- After a loss, stake the smallest whole number of units whose 2:1 win covers all losses since the last win plus one unit.
- After a win, return to one unit.
What 20,000 sessions show
Sessions that finished ahead were up $201.90 on average; sessions that finished behind were down $837.90. 22.2% of sessions ended because the next stake no longer fit the bankroll or the table limit.
Because a dozen pays 2:1, the stake only needs to grow by about 1.5× per loss instead of 2×. The catch is that a dozen loses 25 of 37 spins, so long losing runs are far more common than on red or black. Ten or more misses in a row happen regularly.
The median session lasted 70 spins. The median longest losing run was 9 spins in a row (the worst of 20,000 was 12), and one session in ten put at least $350 on the table in a single spin.
European versus American wheel
The same 20,000 seeds, replayed on a double-zero wheel. The second zero roughly doubles the edge to $5.26 per $100 wagered. Measured return goes from −$2.55 to −$5.10 per $100, and the mean session from −$29 to −$61.
| Metric | European (one zero) | American (two zeros) |
|---|---|---|
| Walked away +$200 | 77.8% | 74.8% |
| Busted (stake no longer fit) | 22.2% | 25.3% |
| Hit the 500-spin cap | 0.0% | 0.0% |
| Finished in profit | 77.8% | 74.8% |
| Mean result per session | −$29 | −$61 |
| Median final bankroll | $1,200 | $1,200 |
| 10th / 90th percentile | $105 / $1,205 | $90 / $1,205 |
| Mean amount wagered | $1,135 | $1,195 |
| Return per $100 wagered | −$2.55 | −$5.10 |
The code that produced these numbers
This Lua is the strategy as Engineer mode runs it, inside the same sandbox and kernel. A test in the repository replays it against the bulk generator on fixed seeds and fails the build if the two ever disagree by a single dollar.
-- Dozen Martingale: bet one dozen and raise the stake after each loss so the next 2:1 win recovers everything.
-- Session rules shared by every 0xroulette.com/systems page:
-- $5 unit, $500 table max, walk away at +$200, and stop the moment
-- the next stake can't be covered by the bankroll or the table limit.
local UNIT, TABLE_MAX, TARGET = 5, 500, 200
local THIRD = {}
for n = 25, 36 do THIRD[#THIRD + 1] = n end
local down
local function reset() down = 0 end
local function place(ctx)
local units = math.ceil((down + UNIT) / (2 * UNIT))
return { { type = "dozen", pockets = THIRD, amount = UNIT * units } }
end
local function settle(result)
if result.net > 0 then down = 0 else down = down - result.net end
end
-- Shared wrapper: enforces the session rules around place/settle.
strategy = {}
local start, stopped
function strategy.setup(ctx)
start, stopped = ctx.bankroll, false
reset()
end
function strategy.bet(ctx)
if stopped or ctx.bankroll >= start + TARGET then
stopped = true
return {}
end
local bets, stake = place(ctx), 0
for _, b in ipairs(bets) do
if b.amount > TABLE_MAX then stopped = true end
stake = stake + b.amount
end
if stake > ctx.bankroll then stopped = true end
if stopped then return {} end
return bets
end
function strategy.on_spin(ctx, result)
if not stopped then settle(result) end
end
Session rules
- Bankroll / unit
- $1,000 / $5
- Exit rules
- walk away at +$200 ($1,200); 500 spins; stop when the next stake exceeds the bankroll or the table maximum
- Table maximum
- $500 per bet
- Seeds
- 20,000 per wheel from master seed
0x6d2e6c8b, identical for every system
Limitations
One bankroll, one unit size, one table limit, and one walk-away target. A bigger bankroll or a higher limit moves where a progression breaks; it does not change the return per dollar wagered. Spins are fair and independent, as on a correctly functioning wheel or a certified RNG. Physical wheel bias is a separate question, covered on Physics.