Another 110 tragic and needless deaths from Covid-19 reported in the UK today. I would remind you that we didn't lock down a week or even a fortnight too late, we locked down three weeks late. Here's what the overall graph looks like now. (1) 
110 is slightly under trend, which is good news, but not by far. Here's a closer look at the recent trend (2)
Indeed if we look at Tuesdays, they've been perhaps the most chaotic with regard to reporting, with two separate declines having occurred. Goodness knows why (3)
I mentioned in another thread earlier on today that because daily variation is getting bigger rather than smaller, you've got to be cautious of any analysis of that. To reiterate, here'e each days total as a percentage of the 7 day rolling total. (4)
But we can still apply smoothing and discern trends. Sequences of days when each day was above the week before are bad, sequences where each is below are good. And we can still put a line through where R was 23 days ago (average day for those who died today to have caused it (5)
The first time R went back up was VE day weekend, then again when lockdown was eased, and we're just bumping down again from a rise that times precisely to shops reopening and the big demos. Yes, we're still waiting to see what pubs have done to this (6)
...but we haven't yet definitely everted to the lower R we had before or even after VE day. That is of concern, it means we're still in this for a long while yet unless we can bring it down (7)
The 7 day running total remains on an ever so slow downward trajectory, one that is dangerous because the more infection we take in to a potential second wave, the worse that will be. This is too slow. (8)
But for the most part, todays data are neither cause for guarded optimism nor new alarm. They are however crucial in determining what trajectory we're on going in to the weekend, and this weekend is a big one (9)
23 days. Thats the average time from catching Covid-19 to death (for the poor folk who die). Any new spread from the pubs opening will start to show some time around this weekend (10)
...or possibly at the end of next week. Whether thats a slowing in the death total decline or even an increase, we can't yet say. And, of course, anyone in those pubs who took the virus home with them will start to spread it around 4-5 days later... (11)
...its quite interesting that the first 'blip' at VE (immediate neighbours and families mixing) was narrower than subsequent blips associated with rushing to public (you'll remember Durdle Door) and the shops, where much bigger, more diverse mixing occurred (12)
We would (admittedly only on a sample size of three upticks and some fairly solid reasoning) expect any change in infection from pubs opening to more resemble the later ticks than the first. We shall know soon enough (fin)
