Refurbishment Crowd Redistribution

The line that came back from nowhere

On July 22, 2025, Test Track reopened at EPCOT after a 13-month refurbishment and immediately posted a two-hour wait. Over its first ten operating weeks it averaged 80.7 minutes between 10 a.m. and 6 p.m. — the longest standby line in the park, longer than Guardians of the Galaxy: Cosmic Rewind.

Here is the part that should change how you plan around closures. In the five weeks before Test Track reopened, the rest of EPCOT’s attractions averaged 25.9 minutes. In the five weeks after, with an 80-minute monster now running next door, they averaged 25.4 minutes. Nothing moved. Guardians dropped 2.4 minutes. Frozen Ever After dropped 0.6. Remy’s Ratatouille Adventure went up half a minute.

Test Track’s crowd didn’t come from the other rides, because it was never sitting in the other rides while Test Track was closed. When a headliner goes dark at Walt Disney World, its would-be riders don’t pile onto the next-best coaster. They largely disappear from the wait-time economy entirely. We found the same pattern in all three of 2025’s biggest refurbishments — Test Track, Big Thunder Mountain Railroad, and Buzz Lightyear’s Space Ranger Spin.

How we tested it

We used Lightning Brain’s archive of posted standby waits, recorded at five-minute intervals, plus operating-status logs that flag exactly when each attraction switches to REFURBISHMENT. All wait comparisons are restricted to the midday-to-early-evening window (10 a.m.–6 p.m.) and to posted waits above zero, so we’re comparing like with like. The three closures give us three natural experiments: Test Track reopening (July 22, 2025), Big Thunder closing (January 6, 2025), and Buzz Lightyear closing (August 4, 2025). Where we compare across years, we hold the calendar month constant and check all four parks as controls.

Finding 1: The reopening is the real crowd event — and it’s fully self-contained

The Test Track reopening is the cleanest test we have, because it runs the experiment in reverse. If a headliner closure redistributes crowds to similar rides, then reopening it should pull that crowd back — the other EPCOT thrill rides should fall when Test Track returns. They didn’t.

Add it up and the story sharpens. Sum the average wait across every EPCOT attraction and you get a “total queue burden” of 285 minutes while Test Track was closed and 360 minutes after it reopened — a 26% jump. Test Track alone accounts for 81 of those 75 added minutes; the rest of the park actually shed a minute or two. On day one back in service, a single attraction was carrying 22% of the entire park’s standby demand, and it materialized without borrowing from anywhere.

This wasn’t an opening-week novelty spike that faded, either. Test Track opened at 91.6 minutes, settled into the mid-70s to low-80s, and was still averaging 78–83 minutes in late September, two months later. A reimagined headliner is a durable demand magnet, not a weekend curiosity.

The lesson for your trip: the crowd event isn’t the closure — it’s the reopening. A refurbished marquee ride can hold 80+ minutes for months, so that’s exactly when you want to see its daily low or grab Lightning Lane. Lightning Brain tracks every reopening in real time and shows you the 5-minute window when each ride bottoms out. Available at lightningbrain.app and on the App Store.

Finding 2: Big Thunder’s closure didn’t feed its neighbors

Big Thunder Mountain Railroad shut down on January 6, 2025 for a full track-replacement refurbishment. If a closed mountain sends its riders to the other mountains, then Magic Kingdom’s thrill rides should have run hotter in early 2025 than in early 2024, when Big Thunder was open. We compared the same calendar month — February — across both years.

Every comparable ride fell 21–29%. Not one absorbed Big Thunder’s displaced riders — Space Mountain and Seven Dwarfs, the two closest substitutes for a family thrill coaster, dropped the most. That’s the opposite of redistribution.

Now, honesty check: was that decline caused by Big Thunder, or was early 2025 just softer than early 2024 across the board? We ran all four parks as controls. February 2025 waits were down 23% at EPCOT, 20% at Hollywood Studios, and 13% at Animal Kingdom versus 2024 — a park-system-wide cooldown that had nothing to do with any single ride. So we can’t pin Magic Kingdom’s drop on Big Thunder. But we don’t need to. The point that survives the confound is this: with a major coaster gone, its natural substitutes did not outperform the broader trend by even a minute. There was no measurable spillover to catch.

Finding 3: Buzz closed, and Tomorrowland got quieter, not busier

Buzz Lightyear’s Space Ranger Spin — a slow-moving interactive dark ride, Tomorrowland’s family anchor — closed August 4, 2025. In the weeks before it closed it averaged 22.5 minutes. Where did that demand go afterward? Nowhere you can see.

Buzz’s own Tomorrowland neighbors — Space Mountain, the PeopleMover, Astro Orbiter — didn’t tighten up to swallow its former guests. They loosened, tracking the normal late-summer slide from July peak toward the September lull. This case is confounded by seasonality in the other direction (crowds fall naturally as school resumes), but the direction is unmistakable: closing a popular ride did not raise the wait at a single comparable attraction. Even the ride most similar to Buzz in ride type and location — the Tomorrowland Speedway — moved half a minute.

So: suppression, or redistribution?

Across three closures, the answer is consistent, and it’s neither of the two things guests usually assume.

Conventional wisdom says a closed headliner “pushes crowds onto everything else,” making the rest of the park miserable. The data flatly rejects that. Not once did a comparable attraction’s wait rise to absorb a closed ride’s demand — not Guardians when Test Track was down, not Space Mountain when Big Thunder was down, not the PeopleMover when Buzz was down.

What actually happens is closer to quiet dispersal. The demand a headliner carries is remarkably self-contained. When the ride is running, it concentrates a huge, specific slice of guest interest — a fifth of an entire park’s queue burden, in Test Track’s case. When it’s gone, that slice doesn’t reroute to the next coaster; it dissolves into the parts of the day that don’t generate a posted wait — World Showcase, shopping, table-service meals, shows, a shorter day, a park hop, or simply a trip booked for a different week. Because those outlets have effectively unlimited capacity, they never show up as a longer line anywhere.

That also reframes the “does it suppress attendance” question. We measure waits, not turnstiles, so we can’t count heads directly. But the fact that Test Track could reappear as the busiest line in the park without lowering a single neighbor tells you the demand it serves is genuinely additive — people organize a visit around riding it. Take it away and you don’t redistribute those guests; you lose the pressure they’d have created, spread invisibly thin across everything that isn’t a ride.

What this means for your trip

• Don’t skip a park just because a headliner is closed. The widespread fear that “everything else will be slammed” is not supported. When Big Thunder and Buzz were down, Magic Kingdom’s other rides ran at normal-to-below-normal waits. A park in the middle of a big refurbishment is often a relatively calm pick, not a crowded one.
• Treat a closure as a mild net positive, not a disaster. You lose one attraction, but you don’t inherit a surge on the others. The rest of your touring plan behaves as if the ride simply never existed.
• Circle the reopening, not the closure, in red. This is where the crowd actually shows up. Test Track came back at 80+ minutes and held it for months. If a reimagined ride is set to reopen during your trip, plan to ride it at rope drop or in the last hour before close, or use Lightning Lane — and don’t expect the “new ride” line to fade after opening week.
• Don’t count on a closure to shorten the line you actually want. If you were hoping Big Thunder’s closure would gift you a quick Space Mountain, it won’t — that demand was never going to move there in the first place.

Limitations

Posted standby waits are a proxy for crowd density, not a direct attendance count, and Disney’s posted numbers carry their own padding. We can measure within-park effects cleanly but can’t directly observe park-hopping or guests choosing a different park entirely, so the “where did the demand go” conclusion is inferred from the absence of spillover, not from tracking individuals. The year-over-year Big Thunder comparison is confounded by a WDW-wide attendance softening in early 2025, and the Buzz comparison is confounded by normal late-summer seasonality — which is exactly why we lean hardest on the Test Track reopening, where the reversal logic controls for those factors. All figures cover December 2023 through September 2025.

The takeaway

A headliner refurbishment does not redistribute crowds to similar rides, and it does not flood the park. Its demand is self-contained: it vanishes quietly when the ride closes and reappears, fully formed, the day it returns. The single most useful thing this tells a planner is that the crowd risk lives on the reopening date, not the closure — so plan your calm days around a park mid-refurbishment, and plan your battle stations around the ribbon-cutting.

Plan smarter: lightningbrain.app · App Store