A RM200 balance may support RM1,600 in cumulative stakes if money returned from earlier rounds is wagered again. At a theoretical 3% house edge, RM1,600 in stakes corresponds to RM48 in expected aggregate loss, not RM6. The opening balance and the total wagered are different quantities.
House edge is the expected net loss to the player as a proportion of the amount wagered, under a defined set of rules and, in some games, a specified strategy. It is not a fee deducted from each round, and it does not tell anyone what their next session will produce.
A calculation needs three ingredients
To calculate a house edge, we need the probability of each possible result, its net payout, and the amount initially at risk. The net payout is positive for profit, negative for a loss and zero for a returned stake.
Consider a purely fictional game that costs RM10 per round. A win pays RM8 net profit and occurs with probability 50%; otherwise the RM10 stake is lost. The expected net result is:
(0.5 × RM8) − (0.5 × RM10) = −RM1 per round.
The house edge is therefore RM1 ÷ RM10 = 10%. Winning half the rounds does not make the game fair: the asymmetry is in what a win pays compared with what a loss costs.
If the same game paid RM10 net profit on a win, the mathematical expectation would be zero. Paying RM11 profit on a win would instead make the expected result positive for the player, assuming the same win probability. This is why win frequency alone is not an adequate measure of value.
The percentage is attached to a wager and a rule set
Published house-edge tables are useful only when the conditions match. Here are three examples:
| Defined wager | Approximate theoretical edge | What must be specified |
|---|---|---|
| European roulette, standard red/black | 2.70% | Single zero, standard 1:1 payment, no La Partage |
| Baccarat Banker | 1.06% | Standard eight-deck rules, 5% commission on winning Banker bets |
| Baccarat Tie | 14.36% | Standard eight-deck rules, Tie pays 8:1 |
The values above are standard reference calculations, not universal rates for games carrying those names. The separate European-versus-American roulette guide addresses how the extra zero changes that wheel’s arithmetic; the baccarat guide explains its drawing and payment rules.
The relevant question is often more specific than “What is the house edge of baccarat?” A Banker bet at one table, a no-commission Banker variation, and a Tie side bet are different propositions. Their probabilities or payouts—and therefore their edges—are not interchangeable.
Turnover can grow faster than a player’s balance suggests
Suppose an imaginary player places 80 wagers of RM20 each on a game with a stable 3% edge. The total amount staked is 80 × RM20 = RM1,600. The expected aggregate net loss across comparable wagers is RM1,600 × 0.03 = RM48.
If the player began with RM200, the initial balance is not the denominator for the 3% calculation. Money returned from one wager may be risked again on the next. The example does not predict a final balance of RM152: actual outcomes vary, and the player may stop earlier or change the size of later wagers.
Here is a useful distinction:
| Measure | Meaning in this example |
|---|---|
| Starting funds | RM200 initially available |
| Turnover | RM1,600 in total placed wagers |
| Theoretical expected loss | RM48 across those wagers |
| Actual session result | Unknown until the wagers settle |
A theoretical edge stays defined even when the observed result is a large win. Conversely, a large loss over a short session does not prove the mathematical edge was higher than advertised.
A percentage can hide different ways of defining the stake
Some table-game calculations assume a fixed opening wager. Others involve extra bets after the initial decision, such as doubling or splitting in blackjack. A figure expressed as a percentage of the initial wager need not equal the percentage of all money eventually put at risk.
Similarly, house-edge comparisons may assume a particular strategy. If one blackjack calculation presumes basic strategy while another presumes arbitrary decisions, comparing their headline percentages directly is misleading. Any side bet should be evaluated as a separate wager, not folded into a claim about the entire table.
These distinctions are more consequential than rounding a published edge from 1.06% to 1.1%.
Theoretical edge is not the same number as observed hold
For a fixed slot configuration, theoretical RTP plus theoretical house edge equals 100%. A published 96% theoretical RTP corresponds to a 4% theoretical edge under that configuration. The dedicated RTP guide explains why actual play can depart sharply from that expectation.
Observed hold is an accounting ratio, not a direct measurement of a game’s theoretical edge. For slot machines, a reported win percentage may use total wagers as its denominator. For table games, a commonly reported hold percentage uses the amount exchanged for chips (the drop), which is not the same as total wagers. Neither figure should be compared with a published house edge without first checking the accounting definition and reporting period.
Published percentages require a clearly defined stake base and reporting period. PlayDash provides brand information; it does not establish the accounting denominator used in a particular casino report.
What house edge cannot tell you
Even an accurately calculated edge says nothing about which outcome will happen next. It also does not specify how volatile the distribution is: two games can have the same expected cost but very different patterns of small and large payouts.
A quoted 1.06% edge on a standard Banker wager and a casino’s reported 20% table hold are not contradictory on their own. The percentages can use different denominators, so comparing them as though they measured the same quantity creates a false discrepancy.