Realized hold is the percentage of a defined wagering base that the casino actually retained during a completed period of play. It is an observed result, not the game’s mathematical target.
The term is useful only when the denominator is clear. In slot reporting, realized or actual hold is commonly statistical win divided by coin-in. In table games, hold is often win divided by drop. Those are different activity measures and should not be compared as though they were identical.
The basic calculations
For slots:
[ \text{Actual slot hold %}=\frac{\text{statistical win}}{\text{coin-in}}\times100 ]
If a bank of machines records $2,000,000 coin-in and $120,000 statistical win:
[ \frac{120{,}000}{2{,}000{,}000}\times100=6% ]
The realized hold is 6%, and the corresponding realized return for that sample is 94%.
For table games, a common operational measure is:
[ \text{Table hold %}=\frac{\text{table win}}{\text{drop}}\times100 ]
If a blackjack pit records $500,000 drop and $75,000 win, realized table hold is 15%.
The table percentage is not the blackjack house edge. Drop is not total wagering volume, because chips can be wagered repeatedly after the initial buy-in.
Realized hold and theoretical hold answer different questions
Theoretical hold is the expected long-run percentage based on game rules, player behavior assumptions, paytables, and game mix. Realized hold is what actually occurred in the observed sample.
A slot with a 7% theoretical hold can realize 2%, 10%, or a negative result over a short period. A table game can have a 1% house edge on wagers while showing a double-digit hold against drop.
The difference does not automatically indicate a malfunction, advantage player, or reporting error. It can arise from normal variance, sample size, jackpot timing, player mix, game selection, promotions, meter issues, or incomplete records.
Percentage variance and dollar variance
A useful comparison is:
[ \text{Hold variance}=\text{realized hold}-\text{theoretical hold} ]
If realized slot hold is 6% and theoretical hold is 7.2%, the variance is (-1.2) percentage points.
The projected dollar difference is:
[ \text{Projected dollar variance}=\text{coin-in}\times(\text{realized hold}-\text{theoretical hold}) ]
For $2,000,000 coin-in:
[ 2{,}000{,}000\times(0.06-0.072)=-$24{,}000 ]
The actual result is $24,000 below the theoretical expectation for that sample. It does not prove that $24,000 is missing.
Sample size changes the interpretation
A hold percentage based on 100 spins is unstable. A percentage based on millions of spins is more informative, but even a large sample must use clean data and the correct configuration history.
Management should consider:
- number of decisions or spins;
- coin-in, drop, or handle;
- volatility and jackpot exposure;
- time period;
- game and denomination mix;
- paytable changes;
- progressive contributions;
- free play and promotional treatment;
- meter and reporting completeness.
A lifetime percentage can hide a recent configuration problem. A daily percentage can overreact to ordinary variance. Good review uses multiple windows.
See Sample Size, Short-Term Variance, and Expected Hold.
Slot hold requires consistent definitions
Nevada’s slot internal-control procedures define actual hold as slot statistical win divided by coin-in and require comparisons with theoretical hold, percentage variance, and projected dollar variance. The current procedures are available in the Nevada slot control document.
The same document notes that statistical win used for performance analysis may not always equal tax-reporting win because promotional or bonus treatment and reporting methods can differ. That is why reports must state their definitions.
Table hold has its own limitations
Table hold can move sharply because drop is influenced by buy-in behavior. Two players can wager the same total amount but produce different drop if one repeatedly buys in and the other recycles chips.
Operational review can pair hold with:
- estimated wagering volume;
- average wager;
- decisions per hour;
- hours played;
- game mix;
- fills and credits;
- opening and closing inventory;
- known credit or chip movements;
- player concentration.
A high realized hold can be a favorable random result rather than excellent management. A low hold can occur during correct, profitable long-run operations.
Negative hold is possible
A casino can have negative realized hold over a period when players win more than the measured base or when timing and accounting create a temporary result. For slots, a large jackpot can produce negative daily statistical win. For tables, a major player win can exceed drop for the period.
Negative hold is not evidence that the game lacks a house edge. It is evidence that short-term outcomes can be large relative to the selected denominator.
What can make realized hold misleading?
Common problems include:
- mixing different paytables or theoretical percentages;
- comparing gross revenue with statistical win;
- including some promotions but excluding others;
- missing coin-in or meter data;
- unrecorded jackpots, fills, or adjustments;
- changing the period after seeing the result;
- comparing table hold with slot hold;
- treating a percentage without volume as meaningful;
- using actual result to rate player value when theo is the approved measure.
A 12% hold on $10,000 and a 12% hold on $10 million are numerically equal but operationally very different.
How managers use realized hold
Realized hold can support:
- game and floor performance review;
- comparison with theoretical configuration;
- identification of unusual machines or periods;
- revenue forecasting;
- investigation of data or meter exceptions;
- evaluation of denomination and game mix;
- communication of actual results to finance.
It should trigger questions, not automatic conclusions. A meaningful review asks whether the result is statistically plausible, operationally explainable, and supported by complete records.
The practical definition
Realized hold is the casino’s actual retained percentage over a stated sample and activity base. It must be reported with the game type, denominator, time period, volume, and theoretical comparison.
The percentage tells you what happened. It does not, by itself, tell you why it happened or what will happen next.
Weighted theoretical hold matters for mixed products
A floor or bank containing different games and paytables should not compare realized hold with a simple average unless each product receives equal wagering volume. A weighted theoretical hold is:
[ \text{Weighted theo}=\frac{\sum(\text{coin-in}_i\times\text{theoretical hold}_i)}{\sum\text{coin-in}_i} ]
If Game A receives $800,000 at 6% theo and Game B receives $200,000 at 10% theo, weighted theo is 6.8%, not the simple average of 8%.
Concentration can dominate a period
One jackpot, one high-limit player, or one short table session can determine a daily percentage. Management should review concentration alongside aggregate hold: share of volume from the largest players, largest jackpots as a percentage of win, and the number of active machines or tables.
A stable floor-wide result can also hide offsetting extremes. Segment review by game, denomination, paytable, location, and time period helps distinguish normal portfolio averaging from a specific control or performance issue.