The Reverse Martingale, commonly called a Paroli-style positive progression, increases the stake after wins instead of after losses. That difference matters for bankroll behavior: a losing spin usually sends the next wager back to the base unit, so the system does not create the same escalating loss chase as a classic Martingale.
What it does not change is the roulette wager itself. A red bet remains a red bet, with the same probability and payout whether the previous spin won, lost, or landed on zero. Paroli changes the path of the stake through a session. It does not change expected value.
What Paroli actually changes
A simple version starts with one base unit, doubles after a win, doubles again after a second win, and then resets after either a loss or a third consecutive win.
With a $10 base unit, the planned ladder is:
$10 → $20 → $40 → reset
That is a positive progression because the stake grows after favorable results. Compare it with the Martingale system, where the stake grows after losses in an attempt to recover them.
The distinction is psychologically important. Paroli tends to cap a cold sequence at repeated base-unit losses instead of demanding $10, $20, $40, $80, $160 after successive defeats. But “less likely to explode during a losing streak” is a risk-management description, not evidence of a mathematical edge.
The standard wheel prices remain those shown in the Wizard of Odds roulette basics: about 2.70% on normal single-zero bets and 5.26% on normal double-zero bets. Official rule documents such as the Nevada roulette rules of play define the wager and payout; no rule awards a better probability because the player is pressing a win.
A three-win cycle, spin by spin
Assume a player bets on black with a $10 base unit and uses a three-step Paroli.
| Stage | Stake | What happens after a win | What happens after a loss |
|---|---|---|---|
| Start | $10 | Bet $20 | Reset at $10 next cycle |
| One win | $20 | Bet $40 | Reset at $10 next cycle |
| Two wins | $40 | Bank the cycle and reset | Reset at $10 next cycle |
There are only four ways the cycle can end:
- Lose the first bet: -$10
- Win $10, then lose $20: -$10
- Win $10 and $20, then lose $40: -$10
- Win all three: +$70
That payoff pattern is why Paroli feels elegant. Three consecutive wins create a large cycle profit, while any first loss in the ladder produces the same $10 net cycle loss.
But elegance does not make the sequence fair. The rare +$70 outcome has to compensate for many -$10 endings, and roulette’s zero pocket makes it fall slightly short.
The cycle mathematics exposes the missing edge
On a European single-zero wheel, an even-money wager wins on 18 pockets and loses on 19. Let:
$$p=\frac{18}{37},\qquad q=\frac{19}{37}$$
For a three-step 1-2-4 Paroli cycle, the probability of the +7-unit finish is $p^3$. The three -1-unit endings occur with probabilities $q$, $pq$, and $p^2q$.
So the expected result per cycle is:
$$EV_{cycle}=7p^3-(q+pq+p^2q)$$
Substituting the European probabilities gives approximately:
$$EV_{cycle}\approx -0.0789\text{ base units}$$
That number becomes intuitive when you look at expected action. The cycle always wagers 1 unit, wagers another 2 units only after the first win, and wagers 4 units only after two wins:
$$Expected\ Action=1+2p+4p^2\approx 2.9196\text{ units}$$
Apply the standard single-zero edge of $1/37$ to that action and the expected loss is again about 0.0789 base units per cycle.
The progression did not create a new price. It merely created a variable amount of action.
“Playing with house money” is an accounting illusion
A common defense of Paroli is that the second and third bets use winnings, so the player is supposedly risking “casino money.” That phrase is emotionally useful but financially wrong.
Once a $10 winning wager has been paid, the extra $10 belongs to the player. Betting $20 on the next spin is a new $20 wager. If it loses, the bankroll is $20 lower than it was a moment earlier. Calling part of that stake “house money” does not change ownership.
This matters because Paroli encourages players to mentally discount the cost of the later steps. The $40 wager can feel almost free after two wins. Mathematically, it is still $40 of fresh action subject to the same house edge rate.
The expected loss calculator is useful here because it treats every dollar wagered as action, regardless of where the chips came from.
Paroli versus Martingale: different failure modes
Paroli and Martingale are often presented as mirror images, but their bankroll risks are not mirror images.
A Martingale concentrates danger in a losing streak. Stakes can grow rapidly while the player is already behind, and table limits or bankroll limits eventually stop the sequence.
Paroli concentrates danger in a winning streak. The player is ahead before the large wager appears, so the sequence usually feels less desperate. A loss late in the ladder gives back the earlier wins and leaves the cycle down one base unit.
| Feature | Martingale | Paroli / Reverse Martingale |
|---|---|---|
| Stake rises after | Loss | Win |
| Cold streak behavior | Escalating stake | Repeated base losses |
| Hot streak behavior | Usually reset after recovery | Escalating stake |
| Table-limit pressure | High during losses | Lower, but still possible |
| Expected value | Negative | Negative |
If the objective is simply to avoid loss-chasing behavior, Paroli is structurally easier to cap. If the claim is “this beats roulette,” both systems fail for the same reason: neither changes probabilities or payouts.
For a broader comparison, use roulette betting progressions compared and flat betting.
Wheel quality still matters more than the progression
A player who insists on using Paroli is still better off applying it to a cheaper underlying game. A three-step progression on European roulette is exposed to about a 2.70% standard edge on its action; on ordinary American double-zero roulette the standard rate is about 5.26%.
French rules can matter further. If La Partage or a favorable En Prison rule applies to the even-money bet, the effective price can be reduced when zero appears.
Notice what changed in those examples: the wheel or settlement rule changed. The progression did not.
This is an important way to evaluate any betting system. If the explanation spends pages describing stake movements but never identifies a change in probability, payout, or rule, it has not identified a source of positive expectation.
What a casino sees when a player presses wins
From the table’s perspective, Paroli is ordinary variable staking. A dealer may see $10 on red become $20 and then $40 after consecutive wins. The operational concerns are mundane: was the larger wager placed before the betting cutoff, is it within the table maximum, and was it paid correctly?
The casino does not need to fear the sequence because the price of each accepted wager remains intact. In fact, a progression can increase total handle during a favorable run because the player recycles winnings into larger bets.
For ratings and theoretical win, the relevant quantities are average wager, time or decisions, and the game’s edge—not whether the player calls the sequence Paroli, Reverse Martingale, or simply “pressing my wins.”
A disciplined way to test the system without believing the sales pitch
If you want to use Paroli as a session format, define it as a bankroll rule, not a profit system.
Write down the base unit, the maximum number of presses, the maximum stake, and the reset point before play. Do not extend a three-step plan to four, five, or six steps because the wheel “looks hot.” Do not switch to a loss-chasing progression after a failed cycle. And do not confuse a successful three-win run with evidence that the next cycle is more likely to succeed.
A sensible test question is not “Did I win tonight?” It is “What exactly did the system change?” The correct answer is stake distribution and session variance. If you want to see that difference visually, use the variance simulator.
The clean conclusion on Reverse Martingale
Paroli has one legitimate advantage over a classic Martingale: it does not require the player to keep increasing the bet while losses accumulate. That can make the bankroll path easier to control and the emotional pressure lower.
Its mathematical claim stops there. The next roulette spin does not know that you won the previous one. A $40 pressed red bet on a single-zero wheel still has 18 winning pockets and 19 losing pockets. A progression can decide how much you wager after each result; it cannot decide how the wheel is priced.
Continue with roulette expected value for the pricing math, roulette payouts for settlement, and why roulette systems fail for the broader system argument. The roulette odds calculator can verify the underlying bet probabilities before any progression is layered on top.