How Casinos Calculate Theoretical Loss to Set Comps

Published September 7, 2026 · 8 min read · By the OddsLighthouse Editorial Team

Ask a pit boss or a hosts desk how they decided to offer a free room, a meal credit, or a show ticket, and the honest answer is almost never a gut feeling. It's a formula. Casinos track how you play through a player's card or dealer rating, run that data through a calculation called theoretical loss, and use the result — not your actual win or loss on a given trip — to decide what you're worth in comps. Understanding that formula changes how you read every offer a casino sends you, and it explains why two players who lost the exact same amount of money can walk away with very different comp values.

The core formula

Theoretical loss is built from four inputs: your average bet size, how many decisions (hands, spins, or rolls) you play per hour, how many hours you play, and the house edge of the specific game and bet you're making. Multiply all four together and you get a dollar figure representing what the casino statistically expects to win from you over that session, regardless of what actually happens on any individual hand or spin. It's a forward-looking, probability-based number, not a record of your real results.

Written out, the formula looks like this: theoretical loss equals average bet, multiplied by decisions per hour, multiplied by hours played, multiplied by house edge. Each of those four inputs is tracked in real time at a rated table or through a slot machine's player-tracking system, which is exactly why inserting your loyalty card before you play matters — without it, none of this data gets attributed to your account, and there's simply nothing for the comp system to calculate from.

Average bet: the number that matters most

Of the four inputs, average bet size usually swings the final number the most, because it's the only one that scales linearly on its own without any ceiling. A pit boss rating a table doesn't just glance at your bet once — they're expected to observe it periodically across the session and record a representative average, since bet sizes commonly vary hand to hand. Some players deliberately manage this by betting larger when a rating supervisor is actively watching and smaller otherwise, a practice sometimes called "past posting for comps" in casino-industry slang — though most modern pits rate frequently enough, and dealers are trained closely enough, that this matters far less than players sometimes assume.

For slot and video poker play, average bet is tracked automatically and precisely by the machine itself through the player's card, removing the observational guesswork entirely. That's part of why slot theoretical-loss calculations tend to be treated as more reliable inputs than a manually rated table game session, even though the underlying formula is identical either way.

Decisions per hour and why game speed changes everything

The second input — decisions per hour — is where game selection quietly does a lot of work. A hand of blackjack at a full table might produce somewhere around 60 to 70 hands an hour; heads-up against a dealer with no other players, that number can climb well past 200. Slot machines and electronic table games can run through several hundred decisions an hour on their own. Since theoretical loss scales directly with decisions per hour, playing the same average bet at a faster game or a less crowded table generates meaningfully more theoretical loss — and therefore more comp value — over the same number of clock hours, without you having wagered any more per hand.

This is also the mechanism behind our related guide on spin speed and losses per hour: the same math that determines your expected losses per hour is the math a casino uses to size up your comp value per hour. They're two sides of the same formula, just read from opposite directions.

A full worked example

Take a blackjack player betting an average of $25 a hand, playing at a table with a house edge of roughly 0.5% assuming reasonably disciplined basic strategy, seeing about 60 hands an hour, across a three-hour session. Theoretical loss works out to $25 times 60 times 3 times 0.005, which comes to $22.50 for the session. Now compare a slot player wagering $2 a spin at 500 spins an hour with a house edge of 6%, playing for the same three hours: $2 times 500 times 3 times 0.06 comes to $180 for the session — eight times the theoretical loss of the table player, despite a much smaller average bet, purely because of the combination of spin speed and house edge.

That gap is exactly why slot players, even relatively modest ones, are often courted with comps as aggressively — or more — than a table player betting far larger amounts per hand. It has nothing to do with which player the casino values more as a person and everything to do with which player's session produced a larger theoretical-loss figure.

Why theoretical loss, not actual loss

Using actual win or loss to price comps would create an obviously broken incentive: a player who got lucky and won money would receive nothing, while a player who lost far more than expected in an unlucky session would get overcompensated, and neither outcome reflects the value of the business the casino is actually getting from that player's action. Theoretical loss strips the short-term luck out of the calculation and prices the comp off the underlying volume and edge instead, which is a more stable, more defensible number for the casino to build an offer system around across thousands of players and sessions.

It also means your comps aren't really a reward for losing, even though it can feel that way — they're closer to a rebate on the volume of action you generated, calculated the same way whether you happened to win or lose that particular day.

Comp percentage: how theoretical loss becomes an offer

Once theoretical loss is calculated, casinos apply a comp percentage — commonly landing somewhere in a range roughly between 10% and 40% of theoretical loss, depending on the property, the game, and how aggressively that casino wants to compete for a given tier of player — to arrive at an actual comp dollar value. That comp value can be issued directly (a same-day meal or room credit) or accumulated into points redeemable later, and higher-tier loyalty statuses typically unlock a higher comp percentage on the same theoretical-loss number, not just extra perks layered on top.

This is also why comp offers can vary noticeably between properties even for what looks like identical play — the underlying theoretical-loss math is the same everywhere, but each casino sets its own comp percentage based on its own margins and how much it's trying to compete for that segment of players.

Why table games and slots get rated differently

Table game ratings depend on a human pit supervisor's observation and judgment calls about average bet and time played, which introduces some natural variance and, occasionally, room for negotiation with a host. Slot and electronic table ratings are entirely system-generated from the machine's own tracking, leaving essentially no room for interpretation or negotiation on the underlying numbers themselves — the theoretical loss a slot system reports is the theoretical loss a slot system reports. That difference is part of why experienced table-game players often make a point of finding and greeting their pit supervisor at the start of a session: it's less about charm and more about making sure their play actually gets rated accurately in the first place.

What this means for you at the tables

If comps matter to you, the practical takeaways are straightforward. Always play on your loyalty card or get properly rated at a table — untracked play generates zero theoretical loss and therefore zero comps, no matter how much you actually wagered. Understand that faster games and higher house-edge games generate more theoretical loss per hour at the same bet size, which is useful context but not a reason to chase a worse game purely for comp value. And don't mistake a comp offer for a signal that a particular game or machine is a good bet — it's simply a reflection of the volume and edge the casino is capturing from your play, priced the same way regardless of whether that game happens to be a good one for you. For more on how house edge itself varies by game, see our guide on house edge.