Nantucket Beat the World Cup

Part of our FIFA Friday series, where we use real flight data to unpack what the 2026 World Cup meant for private aviation. The forecast we are grading here is the full interactive simulation we published before the tournament.

---

During the eight days that held both World Cup semifinals, the third-place match and the Final, Nantucket Memorial Airport took 47 business jets a day.

Nantucket hosted no matches. It is a sandbar off Cape Cod. And it moved more private jets per day than the single busiest airport in eight of the eleven American host cities, including Miami, Boston, Houston and Los Angeles.

Only New York, Dallas and Atlanta had a bizav field bigger than a vacation island. Everywhere else, the World Cup lost to summer.

We are not reporting this from a press release. We run two live aviation feeds, ADS-B position data and the FAA's SWIM flight feed, and those are our own arrival counts. They also happen to be a fairly brutal grade on our own homework.

We said 1.67x. Here is the receipt.

Before the tournament we published a projection across 74 airports in the 16 host metros. The headline numbers were a P50 uplift of 1.67x over baseline and a P90 of 3.53x. Those multipliers are what drove the ramp-will-fail framing, ours included.

Here is what our feed actually recorded, each metro measured against its own normal:

Three of the four came in below what we forecast, and it is not close.

The Final at MetLife, the single largest sporting event on the planet, lifted New York business aviation 29% above a normal day. That is a real bump. It is also the kind of bump the New York area gets from a nice weekend in the Hamptons.

Miami's third-place match managed 19%. Statistically that is a Saturday.

Dallas is the one place we have to show our work. Our SWIM history starts mid-morning on July 14, so the daily total for the semifinal is missing its early arrivals and reads artificially low at 1.48x. Across the hours we did capture, 10:00 to 18:00 Central, Dallas ran 176 arrivals against 92 in the same hours a week later. Call it 1.9x. That squares with what operators there told us: the semifinal was the one day that felt like the brochure.

And Atlanta hit 2.42x, comfortably inside the band. We will come back to Atlanta, because Atlanta is the interesting one.

So: our P50 was roughly right for one venue out of four. The P90 that everybody quoted, including us, never came close to happening anywhere.

Teterboro's biggest day was the following Wednesday

Here is Teterboro, arrival by arrival, across finals week:

Teterboro's busiest day of the week was the Wednesday after the tournament ended. Final day came in at 102, below Teterboro's own eight-day average of 114.5.

Sunday is a quiet day in business aviation, which is exactly why we index everything to share of national traffic instead of eyeballing raw counts. Do that properly and the Final still only gets you to 1.29x. The joke survives the statistics.

Westchester is the one New York field that genuinely worked. It went to 133 arrivals on Final day against a weekly average in the eighties. If you were the line service manager at HPN that Sunday, you had a day. If you were at Teterboro, you had a Tuesday.

Atlanta is the one that was real

On July 15, Atlanta did what the model said all sixteen host metros would do. The metro hit 2.42x normal, and the spike landed almost entirely on the reliever fields:

That cascade is the exact pattern we described in When Teterboro Is Full: the primary fills, the overflow fields absorb the rest, and the small field nobody planned around takes the worst of it proportionally. Fulton County running five and a half times normal for a single afternoon is a real operational event.

So the shape of the forecast was right. The magnitude was wrong almost everywhere it was applied.

One caveat worth naming rather than burying: our SWIM history starts July 14, so everything here is finals week. That is the strongest case for the surge, not the weakest. If it was going to show up anywhere, it was the two semifinals, the third-place match and the Final, in four of the richest bizav markets on the continent, and it came in under forecast in three of them.

Why we were wrong

Three things, and none of them are complicated in hindsight.

The fans flew commercial

Eight. For a World Cup Final. The international demand we modeled showed up at JFK in a seat, not at Teterboro in a Global.

Sixteen cities is dilution, not concentration

Every uplift curve in this business, ours included, was borrowed from single-venue events. Spread the same demand across 16 metros and 39 days and you get a run of ordinary busy days instead of one catastrophic one.

What did show up did not stay

People flew in, watched a soccer match, and left.

In New York, the slowest tenth of Final-day arrivals cleared in 34 hours. On the four days before it, that same slowest tenth sat for 45 to 71. Atlanta is starker: 22 hours on semifinal day against 40 to 64 on the days around it. Both metros had their heaviest traffic and their fastest turns on the same day.

The World Cup produced movements. It did not produce parked airplanes.

The number that actually costs money

Several operators told us they rented extra ground support equipment for the tournament and never unwrapped it.

That is not really a story about arrival counts being wrong. Arrivals are the number everybody forecasts. Dwell is the number that decides whether the ramp works. A hundred jets that turn in seven hours is a good day with a lot of fuel uplift and no parking problem. Forty jets that sit for three days will close your ramp.

Every projection published for this tournament, ours included, forecast the wrong variable. The money got spent on parking capacity for aircraft that were never going to park.

The annoying part is that dwell is the easier number to know. It is sitting in the ground times from every event you have already worked. We had ours too, and we published a multiplier instead.

We will keep publishing the forecasts. We will also keep publishing what actually landed, including the times the two do not match. This one did not match.

Nantucket, meanwhile, is undefeated.

---

This is part of our FIFA Friday series. Want to know what your own ramp actually does under load, rather than what a multiplier says it should? See your layout, or dig through the full World Cup simulation we are grading here.