Walkthroughs · Walkthrough · 8 min

Walkthrough: A Labouchère Line

Write 1-2-3-4, bet the ends, cross them off on a win, add the loss to the end. The cancellation system through one line that clears and one that runs away.

Short answer

A simulated Labouchère session at $5 a unit, target $50 per line. The first line cleared in 8 spins for exactly $50. The second ran to a 16-number line and a $200 stake before the table limittableTable limitMinimum and maximum bet allowed. The maximum is what stops doubling progressions.Read full → made us abandon it $400 down. Usually works, and then does not.

unit
$5
line
1 2 3 4 (target $50)
stake
first + last number
result
−$350 over 39 spins

Walk the system

The stake is the first number plus the last. A win crosses both off. A loss writes the stake at the end.

1234
next stake5result so far0lines cleared0spins0

Stakes and results are in units of the base bet.

The plan

Write 1 2 3 4 on a card. Each spin, stake the sum of the two end numbers, on redbetRedEven-money bet on the 18 red numbers. French: rouge.Read full →. A win crosses both numbers off; a loss adds the stake to the right-hand end. The line clears when nothing is left, which nets the original total, 10 unitsmoneyUnitThe base stake a system or a player works in. Progressions are described in units so they apply at any table.Read full →, $50. A line needs only one win for every two losses to clear, because each win removes two numbers and each loss adds one. Red wins 48.65% of spins, which is more than the one-in-three required, so most lines clear.

Line 1, spins 1–8

1 2 3 4 → stake 5 (1+4): L, line 1 2 3 4 5. Stake 6: W, line 2 3 4. Stake 6: L, line 2 3 4 6. Stake 8: W, line 3 4. Stake 7: L, line 3 4 7. Stake 10: W, line 4. Stake 4: L, line 4 4. Stake 8: W, cleared. Eight spins, four wins and four losses. Net +$50, exactly the target, with the stake never above $50. It works like a small machine.

spinline beforestakeresultline after
11 2 3 4$25L1 2 3 4 5
21 2 3 4 5$30W2 3 4
32 3 4$30L2 3 4 6
42 3 4 6$40W3 4
53 4$35L3 4 7
63 4 7$50W4
74$20L4 4
84 4$40Wcleared · +$50
Stakes in dollars at a $5 unit. The line clears whenever wins are at least half as frequent as losses.
The first line clears in eight spins: four wins, four losses, and the whole target won.

Line 2, spins 9–39

1 2 3 4 again. Eleven losses in the first twelve spins. Each loss appends a number that is the sum of two others, so the line grows at both length and size: after twelve spins it read 2 3 4 5 6 7 8 9 10 12 14 16 18 and the stake was $100. By spin 23 it had 16 numbers and the cycle stood at −$850. Wins cross off the big end numbers, which helps: nine wins in the next fifteen spins cut the line to 12 14 16 18 and the cycle to −$250 by spin 38, with the stake touching the $200 limit on the way. Then zerowheelZeroThe single green pocket that no even-money, dozen or column bet covers; the source of the house edge. Also: single zero, 0.Read full →: the $150 stake lost, the line read 12 14 16 18 30, and the next stake, $210, was over the limit. The cycle stood at −$400. We stopped. The line would have cleared for +$50 eventually, if “eventually” arrived before the limit and the bankrollmoneyBankrollThe total amount set aside for play. Bankroll management is deciding how much of it to risk per bet and per session.Read full → did not.

This session and nine like itresult vs start
+$1,743$0−$500+$1,360−$2500125250375500spinsresult
1,000 sessions × 500 spins · violet p10/p25/p75/p90, light blue median, 83 faint samples · move over the chart for quantiles at a spin
  • p10 −$500
  • p25 −$500
  • median −$500
  • p75 −$500
  • p90 +$1,360
1,000 sessions × 500 spins · y-axis clipped to the 1st–99th percentile (extremes −$500 to +$1,933) · 500 spins · $5 base · even money
Most lines clear in small steps. The ones that do not, run away faster than Martingale because the stake is a sum, not a single number.

The accounts

Total wagered $3,770 over 39 spins against $195 flat. Expected lossmathExpected lossExpected value times total amount wagered; the average amount lost over a session or a period.Read full → on $3,770 is $102; actual −$350, about a third of a standard deviationmathStandard deviationSquare root of variance; the typical distance of a result from the expected value.Read full → below, because at these stakes one standard deviation is about $700. A runaway LabouchèrestrategyLabouchèreCancellation system: write a line of numbers, bet the sum of the two ends, cross them off after a win, add the lost stake after a loss.Read full → line is the heaviest turnovermoneyTurnoverThe total amount wagered over a session or period, counting re-bet winnings. Also: handle, action.Read full → of any common system, because each loss raises the stake by the amount of the loss and the line has no ceiling but the table’s.

What the session shows

Labouchère has the rare property that it usually works, which is why it was a conversation piece at Victorian dinners and why it still sells. The cleared lines are small and frequent; the runaway line is large and rare, and its size is open-ended. The expected loss per dollar is −2.70% in both halves, and the system’s only effect is to decide which half you are in on a given night.

Questions

Can I split a long line?

Yes: a line of 1 2 3 4 5 6 8 9 can be split into two lines with the same total, which lowers the stake and doubles the number of lines to clear. It changes the shape, not the price.

Why not use a shorter line?

A line of 1 1 is a two-unit target with a $10 stake; it clears quickly and runs away just as readily. The length only sets the target.

Is this the “cancellation system” sold online?

Yes, under a dozen names. Henry Labouchère popularized it in the 1870s; he did not invent it.

Comic relief

Usually clears. The exceptions need more paper.

Usually clears. The exceptions need more paper.

Sources

  1. Ethier, S. N. The Doctrine of Chances: Probabilistic Aspects of Gambling. Springer, 2010. Chapter 13, roulette.
  2. Schwartz, D. G. Roll the Bones: The History of Gambling. Gotham, 2006.
reviewed 2026-09-28model: European, single zeromethodology ↗sources ↗