How Casinos Make Money: Hold, Handle, Comps & Theoretical Loss
Casinos do not need luck. Learn how house edge, handle, drop, hold and theoretical loss turn millions of small bets into steady profit, and what comps cost.
Key takeaways
- The house edge takes a small, fixed slice of every bet, so casino revenue scales with total money wagered, called handle.
- Hold percentage is measured against drop, the cash that buys chips, which is why it runs far higher than the house edge.
- Across millions of bets, the law of large numbers makes casino results highly predictable even though each player's results are not.
- Comps are a calculated rebate: a fraction of your theoretical loss, which is average bet times pace times hours times edge.
On this page
A casino does not need to win your bet. It needs you, and thousands of other people, to make lots of bets. Every game carries a small built-in house edge, and multiplied across enormous volume, that edge turns a business built on chance into one of the most predictable revenue streams around.
The house edge: a small slice of every bet
Every casino game pays winners slightly less than true odds. On a European roulette wheel, a straight-up bet wins 1 time in 37 but pays 35:1 instead of the fair 36:1. That gap is the house edge, 2.70%, explained fully in house edge explained.
The edge does not tell you what will happen on one bet. It tells you what the casino expects to keep, on average, from every dollar wagered.
That simple product drives the entire business. More bets, bigger bets or higher-edge games all increase it.
Handle, drop, win and hold
Casinos track money with a handful of terms that sound similar but measure different things.
| Term | What it measures | Simple example |
|---|---|---|
| Handle | Total amount wagered, counting every bet | 280 hands at $25 = $7,000 |
| Drop | Cash and credit exchanged for chips at the tables | You buy in for $500 |
| Win | What the casino actually keeps | You leave with $450, so win = $50 |
| Hold % | Win as a share of drop (tables) or of amount played (slots) | $50 ÷ $500 = 10% |
| Theo | Expected win from a player, based on the edge | $7,000 × 0.5% = $35 |
The distinction matters because the numbers can look wildly different. A blackjack game with a 0.5% edge can easily show a hold of 10% or more.
Why hold is so much bigger than the edge
The trick is that chips recirculate. Suppose you buy in for $500 and play $25 hands of blackjack at 70 hands per hour for 4 hours. You have bet $25 × 70 × 4 = $7,000, fourteen times your buy-in, using the same chips over and over.
Slots work the same way. A machine with a 92% RTP keeps about 8% of every dollar played. But if you put in $100 and keep re-betting your winnings until you have spun $1,000 through the machine, the expected loss is $80, or 80% of the cash you inserted.
Volume: why the casino barely gambles
A single player's results are wildly unpredictable. A casino's are not. That is the law of large numbers at work: as the number of bets grows, the actual result clusters ever more tightly around the expected one.
Consider flat $10 even-money bets on European roulette. Each bet has a standard deviation of almost exactly $10, and the spread grows only with the square root of the number of bets.
| Number of bets | Expected casino win | Standard deviation | Chance the casino is ahead |
|---|---|---|---|
| 100 | $27 | $100 | about 57% (plus an 8% chance of a tie) |
| 1,000 | $270 | $316 | about 79% |
| 10,000 | $2,703 | $1,000 | about 99.6% |
| 1,000,000 | $270,270 | $9,996 | effectively 100% |
At a million bets, the expected win is 27 standard deviations above zero. The casino is not gambling in any meaningful sense. It is collecting a fee with a little noise on top.
Notice the asymmetry. For a player making 100 bets, a winning session is common enough to feel normal. For a casino hosting thousands of players, the occasional winner is already priced in, and the total lands close to the expected figure night after night.
Theoretical loss and comps
To decide how valuable a customer is, casinos estimate their theoretical loss, or theo, using the same inputs: average bet, decisions per hour, hours played and house edge.
- A $25 blackjack player for 4 hours: $25 × 70 × 4 × 0.5% = $35 theo.
- A $1 slot player for 4 hours at 500 spins per hour and 92% RTP: $1 × 500 × 4 × 8% = $160 theo.
The slot player bets 25 times less per decision but is worth more than four times as much to the casino. That is why loyalty programs reward slot play so readily.
Comps, such as free meals, rooms, drinks and show tickets, are a fraction of theo handed back to keep you coming. They are a marketing expense, not a gift. The atmosphere that encourages longer play is part of the same strategy, as covered in casino psychology and design.
Table limits, markers and high rollers
Volume protects the casino, but very large bets reintroduce real risk. A single high roller, sometimes called a whale, can swing a casino's results for a month or more. Casinos manage this in several ways.
- Table maximums cap how big a single bet can be. They also break progressions like the Martingale, which need unlimited doubling to work, as shown in the Martingale system math.
- Markers are lines of credit extended to approved players, which require careful credit checks and collection.
- Loss rebates and other special terms are sometimes offered to high rollers, trimming the effective edge in exchange for big volume.
Beyond the gaming floor
Gaming is only part of the picture. In major resort destinations, hotel rooms, restaurants, shows, shopping and conventions can make up a large share of revenue.
Some gambling products earn money without the house taking a side at all. Poker rooms collect a rake, a small cut of each pot, or charge a time fee, while players compete against one another. Sportsbooks build a margin into the odds: on a standard −110/−110 line, the bookmaker's margin is about 4.55% of total handle if action is balanced. The details are in sports betting math.
On the cost side, casinos pay gaming taxes, staff, security, compliance, marketing and the comps themselves. The house edge has to cover all of it, which is one reason edges are rarely as thin as players might hope.
The bottom line
Casinos make money the way insurance companies do: by pricing every bet so the average favors them, then writing enough volume that the average becomes reality. Handle times edge sets the expected win, hold measures it against cash, and theo decides who gets the free buffet.
For you, the takeaway is simple. The house is not trying to win your next hand. It is counting on the thousands of hands after it. The fewer of those you play, and the lower the edge on each, the less the business model costs you. Look up any unfamiliar term in our glossary.
House Edge & Expected Loss Calculator
What a session really costs, per hour and per year. Put this guide into practice right now.
Frequently asked questions
What is the difference between handle and drop?
Handle is the total amount wagered, counting every bet. Drop is the cash and credit exchanged for chips at the tables. Because players re-bet the same chips many times, handle is usually several times larger than drop.
What does hold percentage mean?
Hold is the share of money the casino keeps. At table games it is usually measured as win divided by drop, which is why it is much higher than the house edge on any individual bet.
How do casinos decide what comps to give?
They estimate your theoretical loss from your average bet, pace of play, time played and the house edge of your game, then return a fraction of that as comps.
Can a casino lose money?
On a single night or to a single big player, yes. Over months of steady play across many customers, the house edge makes a gaming loss very unlikely, which is why casinos set table limits to manage large swings.
Keep learning
House Edge Explained: The Number Every Casino Game Is Built On
What the house edge is, where it comes from, and how to turn it into dollars per hour, with worked examples for roulette, blackjack, baccarat and slots.
Read guide →The Law of Large Numbers: Why the Casino Wins Over Time
The law of large numbers explains why a small house edge becomes a near-certain casino profit. See how your odds of being ahead shrink as the bets pile up.
Read guide →Casino Psychology: How Design, Sound & Light Shape Behavior
Chips, missing clocks, celebratory sounds and comps all nudge how you play. See how casino design works on your brain and what each feature costs per hour.
Read guide →