Do hydration breaks really kill momentum?
FIFA made in-match hydration breaks mandatory at the 2026 World Cup. Coaches and pundits call them momentum killers. Across all 104 matches the team on top really does sag after a break, but it sags nearly as much with no break at all.
momentum points: the average swing away from the team on top in the five minutes after a hydration break, though the same teams sag about 17 with no break at all.
01 — The claim: hydration breaks change the game
Every match it's the same beat. The referee blows for the mandated break midway through each half, both sides jog to the touchline for three minutes of water and instructions. Plenty of people have an opinion about it — players, pundits, fans — but almost no one has measured whether it actually moves the game.
"Hydration breaks are a bit interesting … every time going to a commercial is a bit — not really that I like it. If it's really hot it would be good to put them in. But you have to look at it in every game, separately."
Van Dijk's gripe is flow and ad breaks. The real debate, however, is tactical, and even the TV booth has a name for it: Emma Hayes (the USWNT head coach, on ITV) calls them momentum breaks: "advantageous for the team losing momentum … when you're on top, you don't want it; when you're losing, you do." [ESPN] The data is where that claim gets tested.
02 — What every stoppage does
An index of which side is on top, built from the flow of attacks, shots and dangerous moves, not the scoreline. Positive means the home team is pressing, negative the away team. The project reads the number from FotMob's per-minute momentum, it doesn't compute one; all it does is measure how it moves in the five minutes either side of a stoppage.
Take the team on top before a break and track what it loses over the next five minutes, split by what kind of stoppage cut in. Every stoppage type leans the same way, though the hydration break leans the furthest.
The bars show the 95% margin of error, computed match by match; all of them fall on the negative side. The effect doesn't hinge on the window we picked (it shows up the same at 4, 5, or 6 minutes), but even across the full tournament the interval stays wide, so read it as a signal, not proof. And looking only at the team that was on top doesn't rule out that team simply drifting back to its normal level on its own; that's what the no-break comparison below controls for. The causal claim goes unmade: the gap never cleared chance.
95% INTERVAL (CLUSTER BOOTSTRAP)
A hydration break costs the leading side about 1.3× a VAR review and roughly 1.3× an injury stoppage with no sub. Based on 205 breaks across the tournament.
03 — Match by match
The result is built from every match, all filterable by stage. The wave rises when the home side is on top, drops when the away side takes over, and dashed lines mark detected stoppages. Click any match for the full interactive chart.
The extremes
Where a hydration break landed hardest, and where it barely registered. Each row is a match's biggest swing for the side that was on top; the figure after "from" is how high they were riding when the whistle blew. Click any to open the chart.
Group A6
Group B6
Group C6
Group D6
Group E6
Group F6
Group G6
Group H6
Group I6
Group J6
Group K6
Group L6
Round of 3216
Round of 168
Quarter-finals4
Semi-finals2
Third-place play-off1
Final1
Per-minute momentum rendered from FotMob (derived analysis only; raw payloads not redistributed). One panel per scraped match; stoppage markers from the reconciled FotMob + ESPN commentary feed.
04 — How is it different from other stoppages?
Line the stoppages up by how much of a coaching window they create, and the momentum swing roughly tracks them. The more a break looks like a timeout, the harder the leading team tends to fall.
Three scheduled minutes. Full organised huddle, water, tactical reset.
A long pause, players idle, but no formal touchline instruction.
Unscheduled, but a substitution was made: fresh legs and a word from the bench.
Quick stoppage, play resumes before anyone regroups.
ESPN times every break from the whistle to the resume: a median of 2:52 minutes (from 1:36 to 3:56), two to three times longer than a typical injury stoppage, and remarkably consistent. Does a longer break hit the leader harder? Here the data can't say: the long breaks dip a little deeper (about −22 vs −22 for the short ones), but that gap is too small to separate from noise. So what seems to matter is that there's a long break at all, not small differences in how long it runs.
If a hydration break were only about rest, a long VAR pause should match it. It doesn't quite, which hints at the coaching window a break creates. Belgium's Rudi Garcia put it plainly: "for me, it's a coaching break more than a cooling break." [Al Jazeera] The comparison is by stoppage type rather than perfectly duration-matched (the breaks themselves are timed, but VAR and injury timings are patchier), and the "with sub" injury split is confounded by the substitution itself. Once pre-break momentum is controlled for, the hydration and VAR intervals overlap. Not a verdict, but suggestive. And a VAR pause still lets a coach shout instructions from the touchline, so at most this comparison understates the coaching window rather than inventing it.
05 — The catch
A team that just had a blazing five minutes tends to cool off anyway, break or no break. This is called regression to the mean, and it's the single biggest threat to reading too much into the results above.
So run the exact same measurement where no break was mandated (on the same FotMob scale) and put the −22 next to it. The cleanest control is the very same 2026 matches measured at ordinary calm minutes instead of at a break — a stand-in for no break at all. The leading team fades there too: −17. Add national-team football with no breaks at all (World Cup 2022 at −15, Euro 2024 at −16, the same-host Gold Cup 2025 at −18), and the no-break baseline lands around −15 to −18. Copa América 2024 lands shallower at −10, but its interval is too wide to anchor a bound, so it's shown alongside rather than folded into that range. So most of the −22 is the team cooling off anyway. But not all of it: the break still sits about 3 points below the same teams with no whistle, but even across the full tournament, that gap's 95% interval (-0 to +7 points) still includes zero, not distinguishable from no effect.
Same statistic, same scale. The gold-standard control is the same 2026 teams at quiet minutes (−17), and Euro 2024, World Cup 2022 and the same-host Gold Cup 2025 land right with it (−16, −15, −18): national-team football regresses about −15 to −18 on its own. Clubs swing more (the Club World Cup drops −23), which is why an earlier club-only comparison flattered the "same drop" read. The break (−22) sits about 3 points below the same-teams control once their pre-break level is matched, but even across the full tournament, that gap's 95% interval (-0 to +7 points) still includes zero, not distinguishable from no effect. Even with the full tournament in, that gap stays within the noise — suggestive, not proven. (An event-xT cross-check on 2022 agrees the slide is real.)
That caution is why there's no headline causal claim. Two things hold it back: even the full-tournament sample leaves wide intervals, and a stricter test — one that accounts for how far ahead a team already was — finds no clear difference between a hydration break and any other stoppage. What's left is the ~3-point gap between a break and no break: too small to tell apart from chance, even across all 104 matches. That's where the finished data leaves it. The full model is in the methodology.
06 — Did they even need them?
If the momentum case against the breaks is thin, the heat case for them is barely there. FIFA mandates a break in every match. But cooling breaks were built for genuine heat stress, and most of these games never came close.
So most matches got a mandatory three-minute interruption with neither a momentum effect nor a heat reason. The breaks may still be worth it for the handful of genuinely brutal afternoons. But a fixed, every-match rule is hard to justify as a heat measure in most matches. And even this WBGT, estimated in the shade, runs cooler than the pitch.
And altitude is a different argument: 9 matches sat above 1,500 m (Mexico City, Guadalajara), where thinner air is its own fatigue load. That might justify a breather, but it's a separate stressor this momentum analysis doesn't measure, and it isn't what a cooling break is for.
A different objection is the heat itself: maybe momentum swings more because players from cool leagues wilt in the US summer, not because of any whistle. So the project checked, mapping every player to his club's home city and comparing that heat to match day. Teams furthest from their home climate didn't drop harder; across clubs and across national teams alike, the link came out flat or slightly backwards. The big drops aren't acclimatization. They're regression to the mean. How it was tested →
WBGT (wet-bulb globe temperature) approximated from Open-Meteo temperature + humidity at each venue and kickoff. Altitude (Mexico City and Guadalajara sit above 1,500 m) shapes fatigue, not hydration need; the signal for a cooling break is heat and humidity.
Recomputed at each daily snapshot through the tournament. As of the final data (22 Jul 2026), the hydration swing sits at −22 across 205 on-top breaks. The line shows how it settled as the full tournament came in.
On the surface it's a momentum killer. But the team on top fades almost as much with no break at all — and even across the full tournament, the break's extra bite never separated from noise.
The leader sags about −22 after a break, yet the same teams sag −17 at quiet minutes with no whistle. That leaves a gap of about 3 (95% CI -0 to +7).
Most matches never reached the heat a cooling break is built for, so a mandatory break in every game is hard to justify on heat alone.
The tournament is complete. Across all 104 matches the gap never firmed up — it stayed within the noise, so the extra momentum bite from a break goes unproven.