PHMACAO – Live Roulette Variants and Why European Beats American

One Pocket Is the Whole Difference

A European wheel has 37 pockets: the numbers 1 to 36 plus a single zero. An American wheel has 38: the same numbers plus both a zero and a double zero. Both pay a straight-up winner at 35 to 1. That single design decision is where the money goes.

On the European wheel a straight-up number wins once in 37 spins. A fair payout would be 36 to 1. You receive 35 to 1, so the expected return is (1 ÷ 37) × 36 = 0.9730, and the house edge is 1 ÷ 37 = 2.70%. On the American wheel the same bet wins once in 38 spins against the same 35 to 1 payout: (1 ÷ 38) × 36 = 0.9474, and the edge is 2 ÷ 38 = 5.26%. The extra pocket did not change what you are paid. It only changed how often you are paid.

If a live lobby offers both wheels at the same stake, playing the American one is a decision to pay 95% more for the identical experience. There is no compensating feature. The extra pocket buys you nothing.

Every Bet on the European Layout Prices to the Same Number

Players often assume the outside bets are 'safer' in the sense of cheaper. They are not. On a single-zero wheel every bet on the standard layout carries an identical 2.70% edge, because every payout is short by exactly the same proportion. Only the shape of the outcome changes.

Check any row and the sum collapses to the same thing. A corner covers 4 of 37 pockets, so it wins 10.81% of spins and returns 0.1081 × 9 = 0.9730. A dozen covers 12 of 37, wins 32.43%, and returns 0.3243 × 3 = 0.9730. Even money covers 18 of 37, wins 48.65%, and returns 0.4865 × 2 = 0.9730. Spreading chips across the felt changes your variance – how bumpy the ride feels – and never your expectation. Covering 30 numbers does not make the wheel cheaper; it just means you lose small amounts often instead of large amounts rarely.

French Roulette, La Partage and En Prison

The one genuinely cheaper roulette on a live floor is French roulette carrying la partage. The wheel is the same 37-pocket European wheel, but when the ball lands in zero, even-money bets lose only half their stake instead of all of it. That single rule halves the edge on those bets from 2.70% to 1.35% – the lowest price on any roulette table anywhere.

The arithmetic is short. Put ₱1,000 on Red on a standard European wheel: you win ₱1,000 on 18 of 37 spins and lose ₱1,000 on 19 of 37, giving ₱486.49 − ₱513.51 = −₱27.03, or 2.70%. Add la partage and the zero result changes: 18 of 37 win ₱1,000, 18 of 37 lose ₱1,000, and 1 of 37 loses ₱500. That is 0 − (1 ÷ 37 × ₱500) = −₱13.51, or 1.35%. En prison is a variant of the same idea – the stake is held for one further spin instead of half-refunded – and works out to the same 1.35% on standard rules.

• La partage applies only to even-money bets: red/black, odd/even, high/low. A straight-up on a French wheel still costs 2.70%.
• Not every table advertised as 'French' carries the rule. Open the game rules panel and confirm it before you sit.
• At ₱500 a spin for 40 spins an hour, la partage saves ₱27 an hour against a plain European wheel and ₱78 an hour against an American one.
• There is no rule anywhere on any roulette layout that reduces the edge below 1.35%, and none that reaches zero.

Lightning and Multiplier Roulette: What the Fireworks Cost

Multiplier roulette formats keep the 37-pocket European wheel and add a second layer: each round, a handful of numbers are struck with random multipliers, typically ranging from 50× up to 500× the stake. It looks like free upside. It is funded by cutting the base straight-up payout from 35 to 1 down to 29 to 1.

That trade is close to neutral but not quite. Providers publish a headline RTP of 97.30% on outside bets in these formats – identical to a plain European wheel, because outside bets are untouched by the multiplier mechanic – and 97.10% on straight-up bets, which is where the multipliers land. So a straight-up on a Lightning-style table costs 2.90% instead of 2.70%. You pay a fifth of a percentage point for the chance at a 500× hit, and you accept a far more lopsided distribution: many more dead spins, a small number of very large ones.

None of this is a criticism of the format – it is a well-priced entertainment product and the multiplier maths is disclosed. It is simply not a discount. If someone tells you the multipliers 'boost the RTP', the published figures say otherwise: they slightly reduce it and dramatically increase the variance.

What an Hour Costs, by Table Speed

The edge is a percentage of turnover, so turnover is the variable you actually control. A dealer-run live table gets through roughly 40 spins an hour. An automated wheel with no croupier manages about 60. A speed auto table cycling every 25 seconds runs near 140. Same wheel, same 2.70%, wildly different bills.

Read the corners of that table. The cheapest cell is ₱108 an hour and the most expensive is ₱736 – nearly seven times the cost, with no change in stake and no change in skill. Picking a single-zero wheel and a slower table is the only decision in roulette that measurably moves your result, and it moves it before the ball is ever released.

The Martingale, Worked Out in Pesos

The Martingale is the most persistent idea in gambling: bet ₱100 on Red, double after every loss, and the first win recovers everything plus one unit of profit. It is easy to test with real numbers, so let us test it on a European wheel with a ₱10,000 even-money maximum.

The doubling ladder runs ₱100, ₱200, ₱400, ₱800, ₱1,600, ₱3,200, ₱6,400. That is seven steps and ₱12,700 at risk to win ₱100. The eighth step would need ₱12,800 and the table will not take it. The chance of seven consecutive losses on Red is (19 ÷ 37)⁷ = 0.94%, or about one sequence in 106.

Here is the part that settles the argument. The expected loss per sequence, −₱20.52, is exactly 2.70% of the ₱759.38 that passes through the table on an average sequence. The progression did not reduce the edge, alter the odds, or exploit anything. It restructured the same 2.70% into 105 pleasant small wins and one catastrophic loss that is slightly larger than all of them combined. Every progression ever published – Fibonacci, d'Alembert, Labouchère, reverse Martingale – produces the same identity, because expected value adds linearly across bets and each spin is independent of the last.

Hot Numbers, Wheel Bias and Dealer Signature

Live roulette tables display a statistics panel: hot numbers, cold numbers, the last 500 results, how many spins since each number appeared. It is a record, not a forecast. A number that has not landed in 200 spins has exactly a 1 in 37 chance on spin 201, and a number that has landed three times running has exactly a 1 in 37 chance too. The wheel has no memory and no obligation to balance itself.

Physical wheel bias was real in the era of poorly maintained mechanical wheels, but exploiting it required thousands of logged spins on one specific wheel to establish that a pocket was genuinely favoured rather than randomly ahead, and then a licence to keep playing that wheel. Modern studio wheels are inspected, rotor-swapped and cross-checked, and many live tables are air-driven automatic wheels with no croupier at all – which also removes any question of dealer signature, since there is no hand releasing the ball. Where a human dealer is present, ball release, wheel speed and deflector strikes are not reproducible to pocket precision.

So the honest summary: there is no prediction software, no sequence, no bankroll trick and no seat that produces a positive expectation at roulette. The single-zero wheel returns 97.30% and keeps 2.70% of everything staked, forever. PHMACAO is an independent guide, not an operator and not a licence holder; the casinos we write about are PAGCOR-licensed operators, and their roulette studios are supplied by providers regulated in their own jurisdictions. Players must be 21 or older. Set a loss limit before you open the table, not after. The NCMH Crisis Hotline on 1553 is free, confidential and available 24 hours.