How to Think About What a Coin-Flip Payout Difference Actually Costs You Over Time
When two bets look equally likely but pay slightly different amounts, that small payout gap compounds into a measurable long-run cost you can calculate exactly.
Two bets can share the same probability of winning and still cost you very different amounts of money over time. The difference lives entirely in the payout, and because that difference repeats with every single wager, it accumulates in a way that is easy to underestimate at first glance.
This article walks through exactly how to measure that cost, using plain arithmetic and a worked example you can adapt to any game.
The Setup: Two Bets With the Same Probability
Imagine two separate wagers, each with a 50% chance of winning. That is the simplest probability possible: one fair coin flip.
- Bet A pays $1.00 for every $1.00 risked on a win, and loses $1.00 on a loss.
- Bet B pays $0.90 for every $1.00 risked on a win, and loses $1.00 on a loss.
The only difference is the payout on a win. The probability of winning is identical. Yet these two bets are not equivalent, and the gap between them is not trivial.
Expected Value: The Tool That Makes the Difference Visible
Expected value (EV) is the average result per bet if you repeated the same wager an enormous number of times. You calculate it by multiplying each possible outcome by its probability, then adding those products together.
For Bet A:
- Win outcome: +$1.00, probability 0.50. Contribution: 0.50 x $1.00 = +$0.50
- Loss outcome: -$1.00, probability 0.50. Contribution: 0.50 x -$1.00 = -$0.50
- EV = $0.50 + (-$0.50) = $0.00
Bet A is a fair bet. On average, you neither gain nor lose per wager.
For Bet B:
- Win outcome: +$0.90, probability 0.50. Contribution: 0.50 x $0.90 = +$0.45
- Loss outcome: -$1.00, probability 0.50. Contribution: 0.50 x -$1.00 = -$0.50
- EV = $0.45 + (-$0.50) = -$0.05
Bet B costs you, on average, five cents per dollar wagered. That is a 5% house edge, generated entirely by the payout being 10 cents short of even.
Why "Slightly Short" Feels Harmless but Is Not
Five cents per bet sounds small. Here is what it looks like across a session.
Suppose you make 200 bets of $1.00 each in an evening, which is a moderate pace for many table games. Your total amount wagered is $200.
- Bet A expected loss: $200 x 0.00 = $0 (fair game)
- Bet B expected loss: $200 x -$0.05 = -$10.00
That is $10.00 in expected losses from a payout difference of just ten cents per dollar. The probability never changed. The game structure never changed. Only the payout changed.
If you raise the bet size to $5.00 per wager at the same pace:
- Total wagered: $1,000
- Expected loss on Bet B: $1,000 x 0.05 = -$50.00
The payout gap scales directly with bet size and volume. Neither variance nor luck changes this calculation at the scale of hundreds of bets; the expected cost is arithmetic.
Payout Gaps in Real Games
This pattern appears in actual gambling settings in several common forms.
Even-money bets paid at less than even. In some casino card games, a specific result that would ordinarily pay 1-to-1 is paid at 6-to-5 or 7-to-5 instead. Converting those fractions helps you see the gap clearly. A 6-to-5 payout on a $10 bet returns $12, while a true even-money result at true odds of that event would return more. The shortfall is a per-occurrence cost.
Rounding in payouts. Some games calculate a theoretical payout that falls between two dollar amounts and round down. That rounding is not random over time; it always rounds in the same direction, and it represents a consistent payout reduction.
Commission structures. Certain games charge an explicit fee on winning bets rather than adjusting the payout ratio. A 5% commission on a winning bet that pays even money is mathematically equivalent to being paid $0.95 on a $1.00 win, the same structure as Bet B above.
In every case, the mechanism is identical: the payout on a win is reduced below what a fair bet would pay, and the expected value calculation captures that cost exactly.
How to Calculate the House Edge From a Payout
You do not need a game's published house edge figure to find the cost. If you know the probability of winning and the payout on a win, you can calculate EV directly.
General formula for a two-outcome bet:
EV = (probability of win x payout) + (probability of loss x -1)
All dollar amounts are expressed per unit wagered. If EV is negative, the size of that negative number is your average cost per unit.
For example, suppose a game has a 48% chance of a winning outcome and pays $0.95 per dollar on a win:
- EV = (0.48 x $0.95) + (0.52 x -$1.00)
- EV = $0.456 + (-$0.52)
- EV = -$0.064
That is a 6.4% expected loss per dollar wagered. Both the probability shortfall and the payout shortfall are contributing, and the formula separates their combined effect into a single number.
What This Does Not Tell You
Expected value describes the long-run average. In any single session, your actual result will differ from that average because of variance: the natural randomness in short sequences of outcomes. You can win above expectation in an evening on Bet B. You can lose well below expectation on Bet A. The expected value calculation does not predict what will happen tonight.
What it does tell you is the direction and size of the structural cost you are accepting every time you place a bet. A payout that is 10 cents short of even on a coin-flip bet is not a rounding error or a cosmetic detail. It is the entire mechanism of the house edge, and over enough repetitions, it will appear in your results.
Understanding that cost before you sit down is a more useful starting point than any strategy applied after the fact.