You're not full — you're spiky. Stop managing the average.

Peak nights feel sold out. The weekly average says otherwise. The venues that make money manage capacity yield — not the story they tell themselves about being busy.

It is Thursday at 6:40pm. Court 1 is waitlisted. Court 2 is a private rental that paid full price. The group chat is already arguing about next week's slot. You walk the floor and think: we're crushed.

Open the same week on Monday at 1pm. Or Friday at 8pm. Or that "prime" hour you gave to a last-minute rental that would have booked tomorrow anyway. Half the shelf is still sitting there. You were never full. You were spiky.

Sports facilities do not fail because the building is empty. They fail because owners manage the feeling of the busiest hour instead of the economics of the other thirty-eight.

The average is a trap

Industry chatter still quotes studio-class utilization targets like they transfer to courts. They do not. A 2026 analysis of 888,000+ bookings across hundreds of sports facilities (Baseline) puts hard numbers under what every honest operator already suspects:

Median established facility: about 19% average prime-time occupancy.
Two-thirds still sell out at least one prime hour a year.
Median facility's busiest prime hour: 100% booked. Its 95th-percentile prime hour: about 67%.

Both statements are true: you get crushed at the summit, and most of the mountain is inventory. "We're slammed" and "most shelves are empty" are not a contradiction. They are the same week.

If you only manage the slam — more ads, more desk staff, more marketplace listings for the hours that already waitlist — you are optimizing the wrong hour.

Two yield numbers. Stop mixing them up.

Quiet facilities often look "premium" on a vanity metric: revenue per used space-hour can run around $80 when almost nothing sells. Packed facilities trade sticker for volume and land closer to $47 per used hour — and still triple what they earn per prime capacity hour (about $23 vs $7 in that same study, against a median ~$50 list price for a 60-minute rental).

Read that again.

High rate on few hours is not a business model. It is a building that feels exclusive while the rent still runs on every empty cell. The growth opportunity is the gap between what a sold hour posts and what the average prime hour actually earns.

Illustration, not a customer result: ten courts × thirty-nine prime hours a week × fifty-two weeks is a lot of shelf. If you only obsess over the nights that already sell out, you are arguing about the tip of a demand mountain while leaving the slopes unsold.

Why the spike feels like the whole story

Three habits keep operators trapped in the average:

You measure occupancy as a vibe. "We were busy Thursday" is not a KPI. Average prime occupancy, sold-out prime hours per month, and revenue per capacity hour are.
You release prime rentals too early. Rentals book on a ~1-day horizon in that same research set. Programs, enrollments, and group formats sell weeks out. Giving Tuesday 7pm to a walk-up rental on Monday morning is not hospitality — it is selling the wrong product on the scarce shelf.
You treat 1:1 as the hero format. Packed facilities in the data run roughly 72% of hours as group formats versus ~42% in the quiet tier. One coach, one court, many payers. That is how capacity yield moves — not by raising drop-in rates on empty mornings.

September makes this worse. Fall joins, tryouts, and indoor restarts load the peaks. You feel successful because four nights scream. Then October hits and you discover the mid-day product was never designed — only hoped for.

Cost of inaction

Every week you manage the average instead of the spike:

Waitlists stack on hours you already sold, while shoulder slots never get a product besides "hope someone calls."
Staff invent local rules for who gets prime. When that person quits, inventory policy walks out with them.
You keep buying demand for the summit (boosts, marketplace cuts, more phone hours) instead of redesigning the slopes.
You misread lesson grosses, rental "fills," and membership free hours as the same P&L line — and price the wrong thing.

Inaction is not neutral. It is leaving the only inventory you cannot manufacture — this week's prime capacity — priced by anecdote.

Switch test: can your stack tell the truth about the shelf?

Judge any system — including the one you already pay for — on three questions:

Revenue / mix. Can you see occupancy and yield by daypart and product (private rental vs pick-up vs clinic vs league), not one blurry "utilization" number? Can you hold prime rental inventory until 48–72 hours out so programs claim the scarce shelf first?
Recurring players. When a shoulder product finally fills, do those players book you next week on your domain — or disappear into a group chat and a rented marketplace?
Experience. Does the player see honest status (waitlist, soft-hold, seat reserved pending fill) — or a fake "confirmed" that gets yanked when a better rental shows up?

Scoreline is a white-label venue OS for that mix: your brand on the player site, private rentals and pick-up on the same zones, soft-hold rules so you can recruit without lying, and a desk that is not the system of record. The point is not a prettier calendar. It is running capacity like inventory that expires every hour — under your name.

You do not need a case study to run the test. Pull last week's prime window. Count sold-out hours. Count empty prime cells. Count which products occupied which. If "busy" is still the report, the report is the problem.

What you forfeit if you stay put

You keep the story that Thursday proves the building works. What you forfeit is the rest of the mountain: the capacity hours that never had a product, the prime shelf you gave away a day early, and the recurring book that only forms when the calendar has gravity beyond the peak.

See how Scoreline prices per location, or book a walkthrough and bring last week's prime occupancy by daypart — not the vibe from Thursday night. We will map fill rules to your courts.