People are desperate to force the "it was football" narrative, no matter the facts.
Here's the latest. The FT has usually been a source of well presented info but this is terrible. They started with the conclusion and rigged the chart to fit.
A quick glance and it looks like infections coincide with football matches. But the little footnote gives the trick away. The football dates are adjusted by 4 days, i.e. to show daily cases four days after the match.
This is of course wrong. If you crowd into a pub to watch a match on Sunday evening, breathing in the Covid fumes, there's no way you'll show up in the stats on Thursday - 3.5 days later. The time from infection is in the range 2 to 12 days, with 5 being typical (possibly the median for Delta is a bit quicker at 4 days). Then there's the time to do a LFT and arrange a PCR and get the positive and get into the stats pipeline and be reported. Normally cases today tells us about infections 7 to 10 days earlier. The lines on the chart need to be shifted 3 to 6 days to the right.
It's clear the men vs women rates diverged before the Euro 2020 started. Probably the explanation is worse vaccine rates for men in the <50 groups. Men vs women narrowed in Scotland before Scotland went out.
And of course there's the basic fact that people go to the pub on a non-football Saturday night as well as a big match on Sunday night. And it's not just men 20-34 who watch football in pubs. And slightly more men than women go to pubs overall.