Corona deaths in France:
- week to 3rd September: 130
- week to 16th March (date of lockdown): 123

But I think both headbanging "It's Only Cases Twitter" and bedwetting "Stay Locked Down Forever Twitter" are both wrong, and you can see why if you look at the numbers correctly.

You've probably seen these graphs: cases are way up, even higher than the first wave, and yet deaths are hardly up at all. (All graphs pinched from Our World In Data, because why wouldn't you) https://ourworldindata.org/coronavirus-data-explorer?zoomToSelection=true&year=latest&time=2019-12-31..2020-09-05&country=~FRA&region=World&casesMetric=true&interval=smoothed&smoothing=7&pickerMetric=location&pickerSort=asc

So, the more excitable Toby Young stans use the word casedemic a lot, and retweet things about viral load or false positives to justify their belief that the virus has gone away or is much less dangerous now. I'm not convinced.

Part of the point is that cases are not infections. We have a lot more tests now, and so the first peak may have been 20-50 times higher than it looks. But there's a more important point about cases, which you can see if you plot them correctly.

The point is, and I can't believe I have to say this after 6 months, but PLOT YOUR NUMBERS ON A LOG SCALE. What really matters isn't the height of the peaks, but how fast they went up, and you can't see that on linear axes.

Whereas on a log scale all becomes clear: cases *are* going up, but nothing like as fast as they did first time round. By eye, I reckon they were doubling every 3.5 days in March and are doubling every 15 days now.

It's a big difference from a public health point of view: if you lock down, nothing much changes for 3 weeks or so so further doubling is baked in: the current situation is a bit of a car crash, but it's more of a slow motion one than before.

Ah, but what about deaths? Deaths haven't gone up, have they? Actually, yes they have. It took a while after cases started going up, as younger people got infected first, but again, using a log plot you can see deaths are up lately.

Doesn't looks much, but it isn't helped by the fact that whoever put out the French data yesterday had 2 bottles of Puligny-Montrachet for lunch, and managed to add 19 to 30706 and get 30686 (so ignore the dip in the last two day's figures).

So, overall I think France has serious problems: their test positivity rate keeps going up too, and it's not obvious how they put the brakes on. Of course, the interesting question is how much of this is our future? /FIN

None of yesterday's numbers look good either. (Only 12 deaths, but was 6 last Saturday). Compare positivity and number of clusters with 10 days ago https://twitter.com/BristOliver/status/1298695694463705088/ for example. I'd say France are boiling a frog, but they'd probably take it as a recipe suggestion.