# Surprise! The Quarterfinal That Shook the Internet More Than Any 2026 World Cup Match

Hi, I'm Shiichan! Today I found a story that's less about soccer and more about how soccer shook up the whole internet. It's the kind of data-lover's dream I couldn't resist digging into, so let's go!

## What was announced?

Cloudflare's Blog published a report analyzing how global HTTP traffic shifted during the 2026 World Cup, held from June to July 2026. Cloudflare's network spans 120+ countries and 330+ points of presence, so the team could observe exactly how kickoff times and halftime habits moved traffic in each country.

## Why it matters

Internet traffic usually moves for different reasons in different places, and it's rare for the whole world to change its online behavior for the same reason at the same moment. The World Cup, a 96-year-old tradition, is one of the few events that pulls that off. Watching how traffic shifts around kickoffs and match action gives us a rare window into genuinely global, collective online behavior, not just a fun trivia angle.

To measure this, Cloudflare defined "normal" as the median traffic over the four preceding weeks. Deviations from that baseline are expressed as a log₂ ratio (0 = normal, +1 = twice normal, -1 = half normal), which keeps the comparison fair across countries of very different sizes.

## What changes

This report isn't a new product or feature you can try out tomorrow. But for people designing CDNs or infrastructure, or running streaming services, it's a useful reference for how much traffic can spike around a major live event, and when. It's also simply a fascinating read on how the world's internet behavior moves together.

## Dive Deep

Kickoff time turned out to matter a lot. Matches played between roughly midnight and 8am local time saw the biggest swings, with traffic sometimes more than doubling, since normally very few people are online at that hour. Matches during normal daytime hours (9am to mid-afternoon) barely moved the needle, since most viewers would already have been online. Early-evening matches produced a smaller second bump, most visible on weekdays.

Bosnia and Herzegovina is a clean example of this contrast: when its match kicked off at 2am local time, traffic more than doubled versus normal. When it played in the evening instead, traffic dropped to about 70% of normal, as people set their devices aside to focus on the game.

The Round of 32 match between Brazil and Japan on June 29, 2026 (Brazil won 2-1) is another great case. Brazil's kickoff fell during normal waking hours in its GMT-3 time zone, while the same match landed in the dead of night in Japan (GMT+9). Over the same 90 minutes, Japan's traffic rose to roughly double normal (+1), while Brazil's traffic dropped below normal (around -0.4), the same match producing opposite curves.

To find which match "moved the internet" most, Cloudflare measured each country's traffic deviation in the two hours after kickoff and took the median absolute deviation across countries. The top spot didn't go to the final or a semifinal. It went to the July 11 quarterfinal between Argentina and Switzerland (Argentina won 3-1), with an impact factor of about 1.26, edging out the France vs. Spain semifinal at 1.21. By team, Argentina topped the list at 1.17x, a 17% average swing wherever they played, likely fueled by their status as defending champions and the possibility that each match could be Lionel Messi's last with the national team. France, Brazil, Portugal, Morocco, and Spain followed, along with Norway thanks to the Erling Haaland effect; Haiti and Iraq also stood out due to unusually large swings relative to their typical traffic.

Traffic to gambling sites also rose after the tournament kicked off, and its usual weekly pattern flattened out into a steadier daily rhythm, likely because matches were happening almost every day.

Countries also split into three distinct clusters based on how their traffic curves behaved during pauses in play:

- 44 countries, 101 matches: traffic rises during hydration breaks and halftime, most visible in Croatia and Bosnia and Herzegovina
- 8 countries, 18 matches: the mirror image, traffic falls during those same pauses, in Algeria, Tunisia, Jordan, Egypt, and DR Congo
- Iran's 3 matches: an outlier, because its May baseline was measured while the country was still recovering from an internet shutdown, skewing the comparison

Looking at the June 28 group-stage match between Algeria and Austria by MIME type, the label that tells a browser whether it's receiving an HTML page, an image, or a video stream, Algeria's traffic was driven largely by streaming, and many viewers seemed to close the stream at halftime, while Austria, which streamed less, actually saw traffic rise during the break. Neither country showed much change during the much shorter, three-minute hydration breaks, since closing a stream for that short a pause isn't worth it.

The final kicked off at 21:00 UTC on June 19 (Argentina vs. Spain), and the bronze-medal match kicked off at 23:00 UTC on June 18 (England vs. France). During the final, traffic rose by up to 20 percentage points compared to a similar period, and the shape of the HTTP request volume even traced out the game's individual stages: kickoff, halftime, hydration breaks, and the final whistle.

## Wrap-up

- Cloudflare analyzed global HTTP traffic across the 2026 World Cup
- Kickoff time, especially overnight versus daytime, drove how much traffic shifted
- The same match could produce opposite reactions in different countries, like Brazil vs. Japan
- The single biggest traffic mover wasn't the final or a semifinal, it was the Argentina vs. Switzerland quarterfinal
- Even three-minute hydration breaks were enough to nudge internet usage in most countries

Anyone designing internet infrastructure or streaming services will find this useful, and so will anyone simply curious about how global events move the internet.