Event-Anchored Weekends Are Outpricing Seasons

There is a belief, widely held and rarely examined, that the events calendar has replaced the seasonal one. A fixture on a Saturday does more for weekend rate than the month it sits in. Book the stadium, the argument goes, and the rate takes care of itself.

The summer of 2026 was the largest test that belief has ever had. Forty-eight teams, sixteen cities, three countries, and a month of scheduled demand that every operator could see coming two years out.

It did not go the way the forecasts said.

What happened in Mexico

Mexico hosted matches in three cities. In June 2026, according to figures published by the Asociación Nacional de Cadenas Hoteleras and the Asociación Mexicana de Hoteles y Moteles, this is what the World Cup did to their hotels:

Hotel occupancy and rate, June 2026 against June 2025. Source: ANCH and AMHM, reported July 2026.

City	Occupancy	Change	Rate
Mexico City	56.5%	−2.0 pts	+46.7%
Guadalajara	56.0%	−7.0 pts	+41.4%
Monterrey	53.5%	−5.8 pts	+39.8%

Rates up by roughly forty to forty-seven percent. Occupancy down in all three. The industry had been forecasting 80 to 90 percent; the three cities averaged 55.3. Jorge Paoli Díaz, chief executive of Grupo Diestra, put the realised June range at 57 to 65 percent.

The short-term rental market, working the same weekends in the same cities, did better on both measures: 63.3 percent occupancy in Mexico City, 57.5 in Guadalajara, 56.9 in Monterrey — with rates up 56.5, 78.4 and 44.7 percent respectively. Apartments raised prices harder than hotels did and still filled more of their inventory.

Rates up forty-seven percent. Occupancy down two points. That is not a demand story. It is a pricing decision that did not work.

A contradiction worth noting

Halfway through the tournament, Mexico’s tourism ministry described a different event. Sectur reported occupancy of roughly 95 percent on match days across the three host cities, around 75 percent across the host states, and Mexico City up 16 percentage points against the same period in 2025.

Those numbers cannot be reconciled with the ones its own hotel associations published a month later. One says Mexico City gained sixteen points; the other says it lost two. Nor do they sit comfortably with air traffic: IATA’s booked-ticket data for June and July showed Mexico City down 2.2 percent year on year and Guadalajara down 3.4, with only Monterrey up, at 7.3 percent.

Neither side published a methodology. Different windows, different panels and different definitions of a host city would each move a number, and any of them could explain part of the gap. But the gap is roughly eighteen percentage points on the single most-quoted figure of the tournament, and it has not been explained by anyone. That is worth saying plainly, without deciding who is right.

Monterrey has its own version of the same problem. The state hotel association reported June occupancy of 64 percent, up four points, with match-day averages of 72 percent and peaks near 80, and an average rate of 2,619 pesos, up 38.3 percent. The national associations put the same city at 53.5 percent and falling. Both are published; they do not agree.

It was not only Mexico

The most careful public work on the tournament comes from HVS, which built a counterfactual: it took each host market’s January-to-May growth rate for 2026 over 2025, applied it to the equivalent 2025 match-week performance, and measured the actual result against that baseline. The window ran from three days before each city’s first match to three days after its last.

Across eleven United States host markets, the World Cup produced 680 million dollars of incremental rooms revenue. New York alone accounted for 339.2 million, with average rate up 62.27 dollars.

And occupancy fell in seven of the eleven.

Occupancy point change against the counterfactual baseline, World Cup match weeks. Source: HVS.

Market	Occupancy	RevPAR
Seattle	−6.3 pts	+15.7%
Kansas City	−4.6 pts	+43.1%
Miami	−4.6 pts	+18.9%
Philadelphia	−3.5 pts	+14.8%
Atlanta	−2.6 pts	+9.4%
Houston	−1.8 pts	+22.2%
Boston	−0.5 pts	+22.0%
New York	flat	+23.5%
San Francisco	+0.3 pts	+13.4%
Dallas	+0.7 pts	+30.1%
Los Angeles	+1.0 pts	+15.5%

HVS is unambiguous about the cause, and it is not a demand shortfall. It is the pricing. As hotels raised rates in anticipation of World Cup visitation, the firm writes, individual business and leisure travellers who would ordinarily have booked those markets in those weeks were priced out, and either shifted their dates, chose another destination, or did not travel. Meeting planners did the same thing at scale: they postponed events or moved them to markets without the rate spike.

In New York, where the money actually landed, the local hotel association still revised its own impact estimate down from 300 million dollars to 100 million.

The premium has a half-life

If the World Cup is a single observation, Las Vegas is a time series. The Formula 1 Grand Prix has now run three Novembers in the same city, on the same streets, with the same published monthly data from the Las Vegas Convention and Visitors Authority.

Las Vegas average daily rate, month of November. Source: LVCVA monthly tourism indicators.

November	ADR	Occupancy	Visitors
2023	$249.31	81.9%	3.29m
2024	$199	81.4%	3.31m
2025	$193.04	79.4%	3.14m

The 2023 figure was the highest average room rate in the city’s recorded history. Strip hotels averaged 270.17 dollars that month, up 37.7 percent; on race weekend they reached 629. Two years later the November average is 193.04 — down roughly 22 percent from the peak, and, adjusted for inflation, below where it stood the year before the race existed.

The race did not go anywhere. The demand did not collapse. What disappeared was the operators’ ability to charge for it.

You can watch it happen on the consumer side too. The Las Vegas Review-Journal tracked the same four-night stay at the same three hotels across booking windows. Rooms that opened at 2,694.87 dollars for the 2023 race opened at 1,310.45 for 2024 — and in 2023 had settled at 903.58 by race week. One property opened at 4,336.61 and sold at 1,003.24. The gap between the launch price and the race-week price is the revenue manager’s misjudgement, printed and dated.

The race did not go anywhere. What disappeared was the ability to charge for it.

Not every event is the same event

The aggregate hides three different products. CoStar and IDeaS, looking at booking behaviour by event type, describe conference attendees who commit months ahead and are barely price sensitive; sports fans who book about a week out and will pay almost anything; and concert-goers who book late, stay the shortest time, and generate almost no shoulder nights.

Which is why the most spectacular rate numbers in the record belong to concerts, and why they mean the least. On the Eras Tour, Vancouver saw average rate up 260 percent, with Friday nights at 599 dollars. New Orleans saw RevPAR up 226 percent. Indianapolis recorded rate up 106 percent and Saturday rate up 120 — and Sunday occupancy down 17 percent year on year. The concert did not add a weekend. It moved one night’s demand forward and hollowed out the night behind it.

Geography beats the fixture

The other thing city-level data cannot see is that the winners and losers are often inside the same city.

During the first Las Vegas Grand Prix, while Strip hotels were printing 629-dollar Thursdays, off-Strip properties ran Saturday rates down six percent and occupancy down 15 to 24 percent through the race days. At Super Bowl LX in February 2026, hotels within twenty miles of Levi’s Stadium ran 88 percent occupancy on Sunday and the Santa Clara and San Jose central business district ran 93 — while San Francisco, which had hosted the same event a decade earlier, sat at 82. Against Super Bowl 50 in 2016, San Francisco’s occupancy was down six percent and its inflation-adjusted RevPAR down five. Same region, same event, ten years apart, and the money moved forty miles south.

Monterrey shows the pattern in miniature: a city at 64 percent for the month, match days averaging 72, and individual properties near the stadium peaking at 80.

This is the honest limit of everything above. Every figure in this piece is a city average, because city averages are what gets published. The unit that actually decides whether an event was good for a hotel is the sub-market, and sometimes the block.

What the pattern actually says

Three years of published data across two continents point at the same mechanism, and it is not the one the industry describes.

Events move rate. They move it hard and they move it fast. What they do not reliably do is fill the building, because the rate move itself evicts the demand that would have been there anyway — the corporate traveller, the meeting, the couple who would have come that weekend for no reason at all. HVS measured that eviction directly. Mexico lived it: rates up forty-seven percent, occupancy down two points, a tournament that was supposed to deliver ninety.

And the premium is not durable. It survives the first year, when nobody knows what the event is worth. By the third year the guest has learned the number, the inventory has learned the number, and the number is gone. Las Vegas took two Novembers to give back everything Formula 1 delivered in the first.

The forecasts, meanwhile, were wrong in a consistent direction. Eighty to ninety percent against 55.3 delivered. Three hundred million dollars against one hundred. A tourism ministry at 95 percent against its own hotel associations at 56.5. The errors do not scatter. They all point the same way, and they are all produced by parties with an interest in the number being large.

An event on the calendar is not demand. It is a reason to believe demand is coming — and the belief is priced long before the guest arrives to confirm or deny it.