Doh, just seen this one. Ignore previous

 LLTNP 105m asl 

3,082 new cases

Norman Smith telling R4 PM that he 'suspects' the Prime Minister's condition is considerably more serious than Number 10 is letting on.

As of 9am 6 April, 252,958 tests have concluded, with 13,069 tests on 5 April (excl. Northern Ireland). 208,837 people have been tested of which 51,608 tested positive. As of 5pm on 5 April, of those hospitalised in the UK who tested positive for coronavirus 5,373 have died.

Does anyone know what the number of people tested was yesterday?

Update: 439 UK deaths in last 24hrs. With Scotland only reporting 4 deaths in the last 48hrs, I wonder if we are going to see a big spike tomorrow?

Lets hope that we've reached the peak and that tentative signs of improvement in Italy and Spain are mirrored here very soon.

Best wishes to Boris for making a speedy recovery.

I make it 13,303 tests with 3,802 confirmed as positive 

3 weeks of restrictions starting to filter through?

Not good at all!

Losing the Prime Minister could do one of two things I suspect: 1) Give people a kick up the arse and make them realise no-one is safe from this virus, so they obey the lockdown, or 2) Morale drops and people stop caring, and just start ignore the lockdown rules.

Just watched this video posted by J.K. Rowling. It is presented by a Dr who describes a breathing technique which allegedly reduces the chance of developing pneumonia. 

Any views? If it's not authentic or misleading I'll delete my post. 

Please watch this doc from Queens Hospital explain how to relieve respiratory symptoms. For last 2 weeks I've had all symptoms of C19 (tho haven't been tested) & did this on doc husband's advice. I'm fully recovered & technique helped a lot.https://t.co/xo8AansUvc via @YouTube

— J.K. Rowling (@jk_rowling) April 6, 2020

Latest JFF model results.

Note I'm still using the variation that increases testing linearly until the end of the month. After about a week I'll change to use the data to come up with trend for testing so I can more robustly predict how testing will change. To be clear, any increase being projected by the model is only a result of testing ramp ups. The model predicted a peak yesterday when you account for testing increase. Provided the blue curve remains below the orange curve that prediction will not have been invalidated.

Anyway upshot is this variation model has not performed that well today. It has overestimated the number of cases. Although the 3dayperiodicity variation got it pretty much bang on. I would not be suprised to see a 'big jump day' in a couple of days time. This should not be a major cause for concern though.

Yesterday positives were ~45% and today ~28%. Are we definitely comparing apples with apples here?

For those that want the '3dayperiodicity' variant of the model. Here it is:

Today was a 'big drop day', though you can see the peridocity has gone a bit out of whack in the few previous days. If it returns I will try and combine it with the testing increase to provide one 'unified model'.

I don't think 2) is much of an option. The lockdown remains massively popular with the public (94% support it and only 3% opposes it).

Well the CT, ECG and blood tests would have taken a few hours yesterday. If they're keeping him in tonight there is clearly a degree of concern.

I am no expert, but why not? 

If anything today's figure is so low compared to what I'm expecting that I am actually considering adjusting my model parameters to bring it down a little. Will wait to see if today was a 'big drop day' before doing that though.

It's one of the worst results for my model in the last couple of weeks, but worst in a good way. Big overestimates are definitely welcome!

Can you change the record?  This comment is actually completely bonkers since I was explaining why I didn't think we'd have a problem and why I was fairly relaxed; yet you decide to go back down the same tedious rabbit hole for the thousandth time.

As you seem unable to avoid the petty name calling I feel obliged to point out that you were indeed behaving like a headless chicken back in January, demanding ludicrous, unachievable and disproportionate measures. I remember the frequent rants about cruise ships in which you completely failed to grasp the situation.

Still, I suppose we can expect you to keep trotting out this dross ad nauseam. But I will keep putting you back in your box in similar vein.

17.9K patients are now in hospital.

I thought the update on the PM was very vague. 

Tests results today were excluding NI - I wonder if that's having an impact?

Updated chart for daily deaths. 

UK peak remains 708. Total today 439.

Hospital admissions have risen across the UK by 7% and 25% in East England. London has seen a rise for the second successive day.

Also, can anyone confirm. The Midlands line seems to have been correct. I though that dropped at one point as well? 

Italy have reported 3,599 new cases and 636 new deaths.

• New cases down by 717


• New deaths up by 111

The recommendation to lay on your front is something I heard from someone dealing with patients in intensive care.

Breathing in more deeply means more oxygen reaches further into the lungs so that would also make sense. I'm not sure that it matters whether the cycle is five or slightly more, as long as you don't hyperventilate - which he mentioned getting the symptoms of in the video.

I am not convinced with the updates on the PM.

Today's charts suggest to me that today's data is not complete and will be horrific tomorrow.