A casino-resort does not earn money through one ledger. Gaming, rooms, food and beverage, entertainment, retail, meetings, parking, and other services can all contribute to—or consume—the value of a visit. That makes a familiar promotion question harder than it appears: when is a discounted room, meal, free-play offer, or event ticket a sensible loss leader, and when is it simply an expensive giveaway?
The useful answer is not found in the retail price of the comp and not in gaming theoretical win alone. It is found in incremental portfolio yield: the risk-adjusted contribution created across the whole property because the offer was issued and redeemed.
This framework is aimed at operators, analysts, and managers. It does not propose that every visit can be reduced to a perfect score. It proposes a more disciplined question: What changed because this customer accepted this offer?
The loss leader is a means, not a result
A loss leader is a product or service deliberately priced below its standalone economic value to stimulate profitable activity elsewhere. A complimentary room can be a rational loss leader when it creates additional gaming and resort contribution. The same room can be a poor decision when the guest would have visited anyway, when the room displaces a high-paying customer, or when the guest’s total activity does not cover the variable and opportunity costs.
The phrase “free room” therefore describes the customer-facing price, not the operator’s economics. The room still carries housekeeping, linen, amenity, utilities, reservation, and service costs. On a sellout night it may also carry a much larger displacement cost: the contribution lost because another customer could not buy it.
Portfolio contribution starts with incremental behavior
The first analytical split is between total activity and incremental activity. A guest who already planned to visit may redeem an offer without changing the trip. In that case, attributing the entire visit to the promotion overstates its value. Conversely, an offer may extend a stay, shift the trip into a low-demand period, bring companions, or increase non-gaming purchases. Those changes may be genuinely incremental.
A practical trip-level model is:
Every term must be measured on a contribution basis rather than gross revenue. A $200 restaurant bill does not create $200 of contribution because food, beverage, labor, and service costs must be deducted. A $200 room rate likewise differs from the room’s marginal contribution.
Why ADT is useful but incomplete
Average daily theoretical win, often shortened to ADT, estimates the casino’s expected gaming win from a rated customer for a gaming day. A simplified form is:
If a customer averages $100 per decision, receives 60 decisions per hour, plays four hours, and faces a blended 1.5% house edge, the estimated theo is:
This is useful for comparing gaming exposure under stated assumptions. It is not the same as actual casino win, customer profitability, or portfolio value. It ignores restaurant margin, room contribution, entertainment purchases, the cost of benefits, collection risk, operational burden, and whether the trip was incremental.
ADT can also be distorted by incomplete ratings, unusual short sessions, credit decisions, game mix, errors in recorded average bet, or a visit made specifically to redeem an offer. It should inform a decision, not replace it.
An illustrative portfolio-yield example
Consider two illustrative customers offered the same complimentary room. The amounts below are examples, not industry benchmarks.
| Component | Guest A | Guest B |
|---|---|---|
| Incremental gaming contribution | $240 | $150 |
| Room contribution before comp | $0 | $0 |
| Food, beverage, and entertainment contribution | $25 | $180 |
| Variable room and offer cost | −$85 | −$85 |
| Displacement cost | −$140 | $0 |
| Incremental portfolio contribution | $40 | $245 |
Guest A has the higher gaming value but arrives on a high-demand night and contributes little outside gaming. Guest B has lower gaming theo but visits in a soft period and produces meaningful non-gaming margin. A gaming-only ranking would favor Guest A. A portfolio-yield view favors Guest B.
The example shows why one universal reinvestment percentage can be misleading. The economically sensible offer depends on date, capacity, customer behavior, benefit cost, and confidence in the estimate.
Displacement and capacity change the answer
Capacity is perishable. An unsold hotel room tonight cannot be stored and sold next month. A restaurant seat during a quiet period has a different opportunity cost from the same seat on a fully booked holiday. The value of a comp should therefore be conditional on demand.
A simple displacement calculation is:
If a comped room has a 70% chance of displacing a sale that would have generated $180 of room contribution, the expected displacement cost is $126. During a low-occupancy midweek period, the probability may be close to zero. The physical room is identical; the economic cost is not.
Risk adjustment prevents false precision
Portfolio models rely on uncertain inputs. The guest may not redeem. The trip may happen without the offer. Average bet may be misrated. Actual gaming time may be shorter than expected. Non-gaming purchases may be shared across several guests. A model that reports one precise number without showing uncertainty can create false confidence.
A practical risk-adjusted measure is:
The uncertainty reserve can be larger for new customers, volatile high-limit play, incomplete ratings, disputed credit, or offers with high fixed cost. Mature customers with consistent, well-documented behavior may justify a narrower reserve.
Scenario ranges are often more honest than a single point estimate. Managers can review a conservative case, base case, and upside case, then make the comp decision against a defined hurdle.
Public-company reporting shows the portfolio logic
Integrated-resort companies report gaming alongside rooms, food and beverage, entertainment, retail, and other operations because the economic model is broader than the casino floor. MGM Resorts’ annual filing, for example, reports distinct operating measures for casino, hotel, food and beverage, and entertainment/retail activities. The filing does not prescribe a comp formula, but it demonstrates why property performance cannot be understood through gaming revenue alone. See the company’s 2025 Form 10-K filed with the U.S. Securities and Exchange Commission.
A managerial scorecard
A portfolio-yield decision should record enough information to be reviewed later:
- the offer and its variable cost;
- the capacity period and expected displacement cost;
- baseline probability of a visit without the offer;
- incremental gaming and non-gaming contribution;
- credit, collection, service, and operational risks;
- the uncertainty range;
- the decision owner and approval threshold;
- actual results after redemption.
The post-redemption review matters. Without it, the organization keeps refining assumptions but never tests whether offers changed behavior or merely subsidized activity that would have occurred anyway.
What portfolio yield does not mean
Portfolio yield is not a justification for unlimited reinvestment. It does not mean every non-gaming dollar should be credited to the casino department, nor that a customer’s worth is identical to the amount spent. It is an internal economic framework for deciding how scarce benefits and capacity should be allocated.
It also does not eliminate judgment. Service recovery, contractual commitments, relationship history, regulatory requirements, and reputational considerations may affect a decision even when the immediate financial return is weak. Those exceptions should be documented rather than hidden inside an inflated value score.
The operational conclusion
The shift from loss leaders to portfolio yield is a shift from departmental thinking to incremental economics. The operator asks not whether the room, meal, or ticket was “free,” and not whether gaming theo alone was high. The operator asks whether the offer created risk-adjusted contribution across the relationship after service cost, displacement, and behavior that would have happened anyway.
That approach does not make comp decisions effortless. It makes the assumptions visible, comparable, and open to correction.