The 2026 World Cup's Hidden Variable, Part II: What the Box Score Will Hide

6 days ago we published an analysis of the geographic and environmental inequality built into the 2026 FIFA World Cup. This is the deeper test of that argument, against the historical record, the peer-reviewed match-tracking literature, and the pre-tournament news cycle of the past week.

It begins not with travel maps but with a single peer-reviewed study from the last World Cup played in summer-equivalent conditions.

WHAT BRAZIL 2014 ALREADY SHOWED

George Nassis and his colleagues published the most rigorous heat-and-performance analysis of an elite international football tournament in the British Journal of Sports Medicine in 2015. They took every match of the 2014 World Cup in Brazil, categorized each match by wet-bulb globe temperature, and looked not at goals or wins but at the underlying movement data. They controlled for opponent and match characteristics, which is exactly the methodological discipline any honest regression on travel or environment requires.

Their finding is the cleanest existing causal evidence that the environment changes how elite teams physically perform. It is also the single most important piece of context for understanding the 2026 tournament.

Total distance covered did not change with heat. Goals scored did not change. Cards did not change. Actual playing time did not change. What fell, and fell significantly, was the number of sprints and the distance covered at high intensity. Players self-modulated. They ran less hard, picked their moments more carefully, and preserved the global shape of the match while paying the physiological cost somewhere quieter. The result on the scoreboard looked identical to a temperate-day match. The result in the players’ bodies did not.

This is the entire foundation of the case for environmental performance modeling in elite sport. The environment is not a small effect that fails to show up in the data. The environment is a large effect that shows up in places most analysts do not look. Total distance, goal differential, and points are not where the signal lives. Sprint count, high-intensity output, late-game closing speed, and pressing-intensity decay are where it lives.

WHY THE BOX SCORE LIES

Without controlling for opponent quality, any travel-versus-results analysis is uninterpretable. We ran the simplest possible version anyway, to see what the raw data shows. Across the 32 teams of the 2018 World Cup in Russia, the Pearson correlation between group-stage travel kilometers and group-stage points is +0.14. The correlation between travel and goal difference is +0.21. Both are effectively 0. The high-travel teams in 2018 included Sweden, Belgium, Spain, and Croatia, who are good. The low-travel teams included Iceland, Iran, and Saudi Arabia. The simple regression cannot separate travel cost from team strength, and so it sees nothing.

That is the methodological problem with travel-versus-results regression, and it is exactly why the Nassis approach matters. When you measure sprints and high-intensity distance with match-tracking data instead of reading the scoreboard, you can see the environmental effect even after controlling for opponent. When you only read the scoreboard, you cannot. The signal does not live in win-loss outcomes because players self-modulate to keep win-loss outcomes looking the same. That is a feature of human physiology, not a missing effect.

THE NBA SAYS ELITE ATHLETES ARE NOT FINE

The strongest counterargument to environmental performance research is that 25-year-old elite athletes, with private jets and sleep specialists, are buffered against the effects measured in older studies on commercial-flight populations. On any single match this is approximately true. The published recovery literature, including Reilly’s position statement for the European College of Sport Science, notes that sleep-wake cycles typically recover within 2 to 3 days for athletes after long-haul travel. At the population level on individual matches, the “they will be fine” argument has merit.

On 25,000 matches it collapses.

The Taylor and Francis analysis of the NBA isolated the circadian effect by holding location of play constant and varying only the direction of travel for the visiting team. When Pacific time-zone teams host Eastern visitors, they win 63.5 percent of the time. When Eastern teams host Pacific visitors, they win 55.0 percent. The 8.5 percentage-point gap is produced by the direction of travel and the circadian state of the visiting roster. The NBA has charter jets, in-flight sleep optimization, blackout schedules, and the most sophisticated travel infrastructure in any professional league. The gap persists.

2 additional facts strengthen the case. The jet-lag literature consistently reports large interindividual variation in adaptation speed. The population mean recovers in 2 to 3 days. The slow-adapting individual on a given roster does not, and tournaments are decided by individual moments. Separately, the 2025-26 European club season is on track to deliver multiple Premier League players to the tournament at 50 to 70 cumulative appearances. Recovery at appearance 55 is not recovery at appearance 30. The travel cost compounds onto seasonal fatigue debt.

QATAR 2022 AS THE REMOVED-VARIABLE CASE

Qatar 2022 is the other half of the historical record, and its result is worth stating carefully. We pulled the match data from StatsBomb’s public archive and computed the great-circle distance between every team’s successive venues across both tournaments. The mean group-stage team in Russia 2018 traveled 2,223 km. The mean group-stage team in Qatar 2022 traveled 39 km. The cross-tournament ratio is 57 to 1.

This figure is real and it is informative, but it is not what we initially claimed. Qatar 2022 produced a 39 km mean travel figure because Qatar is approximately 160 km long. The compactness is geometry, not FIFA equity policy. FIFA cannot manufacture Qatar in North America. The 57x comparison is best read as an existence proof. A recent World Cup, played at the level of the international game by 32 teams across 64 matches, had effectively 0 km of intra-tournament travel. The competitive distortions associated with travel were not observed because the travel was not present to observe. That is not a recipe FIFA can apply to a continental tournament. It is a baseline data point for what football looks like when the variable is removed.

What FIFA does control is the draw. Host geography is fixed. The composition of the 8 groups is not. A travel-equity constraint on the 2026 draw would have required that the 4 teams in each group play within a single regional cluster. FIFA did not impose one. The result, in a single group, is this.

Argentina and Algeria are in the same Group J. They play the same 3 opponents. They get the same rest days, the same match windows, the same broadcast slots. Argentina’s group-stage route covers about 700 km. Algeria’s covers about 5,000 km. The within-group travel ratio is 7.1 to 1. That is not a function of the United States being big or Mexico being far from Canada. It is a function of how FIFA seeded and located the matches of a single group. Apply the Nassis framing to this disparity. Algeria will not arrive at each of their 3 matches with the same physiological baseline as Argentina. Their sprint counts and high-intensity output will not be the same. The scoreboard may obscure it. The bodies will not.

WHAT 2026 STACKS ON TOP

One factual correction to the original piece is in order before the Group H analysis lands. 3 of the United States 2026 venues are climate-controlled, not 1. AT&T Stadium in Arlington, NRG Stadium in Houston, and Mercedes-Benz Stadium in Atlanta all maintain interior conditions near 22°C regardless of outdoor temperature. Treating Atlanta and Dallas as outdoor heat exposures was incorrect on the playing-surface level. The correction matters. Spain’s Group H illustrates why the thesis survives it anyway.

2 of Spain’s 4 venues are indoor matches with outdoor exposure surrounding them. Pre-match warm-ups happen on outdoor training pitches in afternoon heat. Walk-throughs and pitch inspections are outdoor events. The halftime transition from a 22°C bowl back onto an indoor pitch crosses 2 air systems and is itself a thermoregulatory event. Dehumidified indoor air affects respiratory function. Cold-to-warm body temperature swings disrupt the neuromuscular signaling that elite-sport heat protocols are designed to protect. The indoor stadiums do not erase environmental exposure. They redistribute it across the day.

Miami remains a brutal outdoor heat venue. Guadalajara is a moderate-altitude problem. Mexico City is an altitude problem of a different order. Spain’s group does not include Mexico City, but Group A and Group K do. Apply the Nassis framework to each of these venues and the picture is the same. Goals and total distance will look approximately normal. Sprint counts, high-intensity output, and late-game closing speed will not.

BRAZIL 2014, REPLAYING THIS WEEK

None of what we have just walked through is hypothetical. The Nassis pattern is not a historical artifact. 2 days before the tournament begins, it is already showing up in the pre-tournament friendlies and the training-camp footage. Norway, preparing for their group-stage matches, wore ice collars around their necks during a recent friendly against Morocco. European players, mostly from clubs whose entire competitive calendar runs in temperate conditions, have been photographed sunburnt and dousing themselves with water at training. Their bodies are doing the same self-modulation the Nassis paper described in 2015, in real time, before a single competitive minute of the 2026 tournament has been played.

The institutional response is even more telling than the player photographs. Georgia’s state public health director went on the record at the opening of the new USMNT base camp in Fayetteville to predict that heat-related injuries would be an issue at the host nation’s training site. England flew heated training tents into their domestic facility to start acclimatization at home before they boarded a plane. A coalition of current and former professional players published an open letter to FIFA about the conditions, and a player’s description of what heat does to elite-level performance almost exactly restates the published finding from Brazil 2014. The point is not that 1 team is suffering. The point is that the most resourced institutions in elite sport are paying real money to preempt the effect. That is the single strongest piece of evidence that the effect is real, that it is well understood by the people closest to it, and that the published research is not academic.

The tournament begins on June 11. Brazil 2014 has stopped being a paper in the British Journal of Sports Medicine and started becoming the news cycle of this week. Over the next 39 days, it will become the central physical story of the tournament. The question is no longer whether the effect is real. The question is how unevenly it lands across 48 teams, what each team did or did not do to prepare for it, and which knockout matches will be decided by sprint counts that the box score will never explain.

WHY GENERIC PROTOCOLS ARE THE WRONG ANSWER

The most common reading of environmental performance research is collective. Hot tournament, so every team takes the same heat protocol. Altitude matches, so every team takes the same altitude protocol. Long travel, so every team takes the same circadian protocol. That reading is wrong because the inputs are not the same.

A Colombian squad arriving in Mexico City from Bogotá needs no altitude protocol because their physiology is already adapted to 2,600 meters. Their Miami match is the load that matters. An English squad arriving in Mexico City from Atlanta needs maximum altitude protocol and minimal heat preparation. A Senegalese squad playing exclusively on the East Coast needs neither and should spend its protocol budget on recovery between matches and management of cumulative seasonal load. The interventions are evidence-based and individually calibratable. Applying them uniformly across a team is a way to spend resources without buying performance.

The Nassis finding implies this directly. If players self-modulate under environmental stress, and if the self-modulation varies across individuals, then the right unit of analysis is the individual, not the team. Population-mean protocols are wrong by construction because they treat heterogeneous bodies as if they were homogeneous. The data has been telling us this since 2015. The 2026 tournament will be the largest live test of whether anyone listened.

Our original article made a single argument. FIFA was treating a physiology problem as a logistics problem. Everything in this follow-up sharpens that argument rather than softens it. The physiology is real. It is measurable. It has been documented in match-level peer-reviewed research for more than a decade. It all lives in a place the scoreboard will never show.

Brazil 2014 proved the mechanism. Qatar 2022 proved what football looks like when the environmental variables come off the table. The NBA proved that even the league with the most sophisticated travel infrastructure on earth cannot recover its way out of the cost. The 2026 schedule reintroduces every variable Qatar took away, layers altitude on top of 2 stadiums, and runs an unconstrained group draw across all of it.

The conclusion is not that teams need a better universal heat protocol or a better universal travel protocol. They need an individualized one, calibrated to each athlete's actual itinerary, home-climate baseline, chronotype, seasonal load, and personal adaptation curve. The same environmental stress applied to 2 different bodies does not produce the same response. Population-mean protocols treat heterogeneous athletes as homogeneous ones, which is the wrong unit of analysis for a tournament that will be decided by individual moments.

That is the problem we are built to solve.