Sorry to be a bringer of sad tidings, I really am, but todays figures for Covid-19 deaths in the UK are way above trend. 74. Here's the overall picture. (1) 
The death total isn't going down. The seven day trend is upwards. Have a look at the most recent data here. The last month or so. We're looking at deaths per day remaining static, at present trending slightly up, over nearly two weeks (2)
The problem with big curves like the top one is that the human mind doesn't really take that in and see changing trends very easily. Log graphs help. As I keep saying, drop a log on it, if its straight then the trend is proportionally constant. Here's what that looks like now (3)
And if that isn't convincing just look at the 7 day rolling total. Its static, near enough (4)
We're in the fourth,now by far the longest, stalling of the fall in daily fatality rate. The new normal, right now, is deaths per day are not going down. Take the 7 day total, divide it by the one before, average that over 7 days and its a very, very clear trend (5)
In that graph above, the lower the line, the faster the rate deaths are falling. They are now trending upwards. (6)
Deaths going up now means R was rising 23 days ago, the peak day people who died caught it. You may recall I predicted three possible routes for this week, and now we know which was correct (7)
The rise was due to pubs opening, which tracks back perfectly to when death rate decline slowed. One possibility was we'd just tip over the top of that peak before going up again as people who caught Covid when the pubs opened passed it on 5 days later (8)
Remember, peak infectivity is 5 days from infection, peak death is another 18 days later. You're most infections 5 days after catching it, most likely to die 23 days after. Here's what R was doing, calculated using deaths, at a 23 day delay. This is unsmoothed data (9)
One day doesn't make a trend. And on Tuesday, the next day we'll get a statistically significant proportion of the weeks death (daily variation is still massive, see graph in this tweet) we may see something else happen (10)
But as things stand, 23 days after the pubs opened we see the death total started to rise, R was rising at that point. 3 days later deaths started to fall, only to start going up again 5 days later. (11)
The precision with which this trend matches the night that pubs across the UK were rammed full of revellers, and the exactness of the 5 day peak infectivity coinciding with a second rise in death rates... It seems rather more than a coincidence (12)
What we don't know is what happened afterwards. If people who caught it from pubgoers kept the disease within those households we'll see a bump and by the end of next week deaths will be falling again. If they didn't... This is getting messy (13)
There are times I like being right. This has not been one of them. Sorry to be the bearer of bad news, but we now have proof that opening pubs, those crowds of drinkers, that caused a new small peak in Covid-19 deaths. (14)
Lets all hope that it goes no further. Wear a mask. Keep apart from people as best as you can. Do not be among the victims of this disease if you can help it. Spread the word that we're far from being out of this. (fin)
