Here is the updated of US data using the latest. Chart below shows new deaths & new confirmed in a 10:1 ratio (cases to the left scale & deaths to the right).

Will look at the states w/ highest new confirmed cases recently (California, Texas, Florida). https://twitter.com/Trinhnomics/status/1277819144449089536

Latest date is 30 June (It's 1 July in HK but still 30 June in the USA). Anyway, here is California. The absolute increase yesterday was:

+37 deaths
+4,293 cases (yesterday+15,861)

That said, data tends to be volatile on the day, so better to smooth using a 7-day moving average

Absolute deaths are down in California (right scale blue line) & cases wise seem to be trending down recently (left scale orange line) although cases & deaths diverge in trends & so we should look at ratios of fatality to see trends!

2 ratios:

The blue line shows fatality of new daily deaths (7-day moving average)/ new cases (7-day moving average) = 1.1% & same as previous day.

Here's the good news:

Current fatality/ cases of TWO WEEKS AGO to take care of lag issue = 1.7% for 30 June& down from 1.9 yesterd

Texas, also a state w/ high surge of Covid, shows this:

New deaths = +15
New Cases = 5303

But new deaths using 7-day moving average = 39 & cases =4170. I do this to smooth out vol & u see that deaths are down in the blue line while orange surges!

Texas fatality shows that fatality is falling if u use Covid deaths now/ Covid cases two weeks ago = 1.7%, down from 1.9% previously & the trends are showing below.

Blue is current deaths/current cases (7-day moving average of course).

Florida, another state w/ high surge of Covid-19 cases:

New deaths = +58 (yesterday 81)
New cases = +6,093 (yesterday was 23,381)

So clearly volatile so u do a 7-day moving average to show trends & it shows deaths are flat.

CA, Florida & Texas fatality all lower than national.

Florida daily change fatality:

Current fatality is very low at 0.5%
If you lag using current deaths/current fatality of two weeks ago then 1.6% from 1.8% yesterday & trend is downward even if we lag.

So it means fatality is falling!

This is a chart of deaths & percentage change per day (using 7-day moving average new daily) & you can see that we still haven't seen a massive spike.

And u can see that already in the absolute numbers I showed earlier but this is the math of the rate of daily change .

Deaths & cases: We compare deaths rate of change daily & also rate of change of new cases.

You can see that the orange line & blue line diverge in rate of growth & that's already in the level chart earlier & also the fatality ratios declining. Just another way to analyze.

In California, Texas & Florida:

*Deaths are down even if cases are up
*Relative deaths are down (fatality ratios), and even if we us lag of 2 weeks, deaths are down.

Okay, someone asked me to do Arizona (a fellow finance person over IB ) so here it is:

Deaths +47
Cases+4683

As previously stated, because daily changes can be so volatile, we do 7-day moving average daily change & you can see that:

a) Cases rising rapidly
b) Deaths also rising but slower.

Let's look at ratios & rate of change!

Arizona is different in that absolute deaths are rising, although let's not forget that its fatality ratios are still lower than the USA average.

Fatality of current deaths/current cases = 1.2%, same as yesterday

Current deaths/confirmed 2 wks =2.2% down from 2.3%

Arizona Covid deaths % daily change. Chart to the left is daily % change using the 7-day moving average & chart to the right is smoothing further by taking the 4-day moving average. Still shows the same:

Increase & some trending down but no spike (well hopefully ever & not yet)

Let's look at some other factors for Arizona (CA testing ratio is about 10% of population & the hit ratio is about 5%).

Here is Arizona. It has tested in total 302k w/ a cumulative hit ratio of 6.7% positive & so it has tested less as a share of population than CA (pop = 7.5m)

Basically it has tested 4% of total population (the US has tested about 10% total) so it is behind the USA by a long shot and so when it does test people, it is finding out more hit ratios as well.

The latest positive is 8.4% so higher. Let's look at hospitalization ratios.

There are 2 ways that I do it:

a) Total # of people now in hospital = 300/(total cases - total recovered =) = 6191 so 300/6191 = 4.8%

or

b) Change of cases/ change in hospital = 1.5%

Either way, u get a trend that is downward!

So new cases = less hospitalization!

Next, let's look at ventilators usage. Arizona doesn't have ICU data so let's use vents. So far we know that:

a) Deaths rising but rate is not spiking & gently upward from a low base
b) Fatality is falling
c) Hospitalization falling
d) Vents use is up to 63.
Let's look at ratios

So there are two ways to take the ventilator ratios:

Either cumulative vents use change / cumulative case change = 0.6%

or taking the change of current vents/ daily new cases = 0.05%

What we care about is trends & so:

New Covid = less hospitalization & less vents use

Meaning, we may get an increase of deaths but that increase isn't going to be proportion to an increase of cases for a wide range of reasons, including:

*Better treatment (hospitalization ratio falling & vents/ICU falling)
*Earlier detection of virus & likely many asymptomatic

For those interested:

Covid Data is free & widely available. And you can get it here & download & make all the charts you like all day everyday !!
https://covidtracking.com/ 

Sincerely,

@Trinhnomics

Going for a hike now! Free & healthy & wholesome & what a clear day in HK!