A baker's dozen day trend in slowed acceleration of new cases

[Ms. Toad]

(i.e. flattening the curve)

Daily multiplier (prior day's cases x daily multiplier = current day's cases). Two very minor blips up (bolded)

3-22 - 1.385797497
3-23 - 1.303732189
3-24 - 1.253526924
3-25 - 1.244204075
3/26 - 1.252510592
3/27 - 1.218774507
3/28 - 1.186812131
3/29 - 1.149452168
3/30 - 1.15344921
3/31 - 1.150667708
4/1 - 1.140417971
4/2 - 1.138946898
4/3 - 1.131837616
4/4 - 1.123379552

All the usual caveats about trends and all the stuff that can interrupt them (new testing that reveals tons of new cases, high population areas taking off at a higher rate of increase than the national average, all of us getting stir-crazy and ignoring stay-at-home orders, and so on).

But those of you who have been skeptical,and have been suggesting that this is just like China blips, or previous 1-3 day downturns in US cases, go here: https://www.worldometers.info/coronavirus/country/us/

Scroll down to the new cases graph, and click on the logarithmic scale. The downward curve is much more visible now than when I (and others) started calling it out on around the 25-27th of March. Line up a ruler along the curve from abotu 2/29 to 3/22 - from there to the right you should be able to clearly see it pulling away (down) from your ruler.

Now: That doesn't mean we're out of the woods yet - it just means the coronavirus gods no longer have the accelerator floored (that's what the flattening the curve means - and what the logarithmic curve shows). We're still speeding up (more new cases each day than the day beffore), we've just eased off the accelearator a bit so there aren't as many new cases each day as if the accelerator was still floored - if it keeps going (i.e. we keep up these restrictions) we will be closer to having enough resources to care for the people who need them.

The daily multiplier for the # of deaths curve is not as smooth, but also trending down:

3/24 - 1.405405405
3/25 - 1.316666667
3/26 - 1.260954236
3/27 - 1.308880309
3/28 - 1.315044248
3/29 - 1.158815612
3/30- 1.216027875
3/31 - 1.290035021
4/1 - 1.259378085
4/2 - 1.190672154
4/3 - 1.17198815
4/4 - 1.186911951

[Takket]

1. if we assume........

that the rate of new cases decreases by about 0.01 a day, we are still about 13 days away from the peak of the curve, and the point where we start seeing fewer new cases "today" than we did "yesterday". and even then we must maintain rigid stay at home discipline to keep the curve on the downward slope.

[Ms. Toad]

2. My curve has us at about 16 days away

As we flatten more, the peak gets farther away (it was on th 15th, on my curve it's on the 20th now).

But you are absoutely right that we have to stick to the stay-at-home discipline. It's like we're hanging out under the accelerator pushing up against the COVID foot. As we wander back out (and there are fewer of us pushing back on the accelerator), it will creep back down to th floor.

[Pobeka]

3. But you won't get .01 a day decrease. The data shows the daily decrease is decreasing...

The decrease will asymptotically approach 0.0 a day until the peak is reached. More than 13 days away. Probably more like 26 days away, if the asymptotic reduction is linear (which I haven't looked into...)

[Igel]

4. multiple peaks, not a peak.

And that's a good thing, because it means that by the time parts of the country need ventilators, those using them in other parts of the country are likely to have died and the ventilators prepped for the next pre-corpse.

Most of those who go onto ventilators die. When the doctor says, "Oh, it's serious, we have to intubate" just make funeral arrangements. Easier to cancel later than make them under duress.

Yeah, harsh. That's how it goes. We're in this situation because people didn't want to make hard, harsh choices. The universe really doesn't care.

https://covid19.healthdata.org/projections

Better than the OP, with all the usual caveats. Better than the OP, it includes calculated error out to the 95th percentile. Given random variation and inherent error, the likely death toll (equipment usage, etc.) is within the shaded blue. The dotted line is most likely course, but by "most likely" don't think "very likely". Sometimes it might just be a 15% chance it'll be that value, but less everywhere else. That dotted line should really be just a darkening of the background color that suggests a line.

And the reason I include that link here is because it's state by state. (Area by area would be better--seriously, NYC is on one trajectory, Buffalo on another, Rochester on a 3rd, and upstate NY is its own critter. Same for Texas. There is no "Texas" curve. It's backwater areas + Dallas/FTW + Houston + El Paso + Austin + San Anton. Some paths are similar, some aren't.

(Tensors were made for this kind of thing. And the fact that I wrote that statement means I either have had too much borovicka or not nearly enough.)

[Demsrule86]

6. That is not true...some have lived who have been on ventilaters...a 12 year old from Georgia for

example...and older folks too...

[greenjar_01]

8. "people didn't want to make hard, harsh choices."

Who?

[Demsrule86]

5. So many people can't get tested, I don't see how we can know for sure.

[aikoaiko]

7. Thank you for the data presentation.

[JCMach1]

9. Prior week TX was around 55K tests conducted and at 63K today....

That is still extremely low. Even first responders here are having to wait 8-10+ days for test results. This was reported on the local Fox affiliate in DFW this evening. It is sidelining whole groups of responders who have to quarantine.

If we knew the actual availability of testing we could actually get a good look at this...

I still think we are on the outer tracks of the Imperial College projections.

[BrightKnight]

10. I head on the local news that you could only get one if you were admitted

To a hospital or a first responder. I had a fever and body aches for 7 days and did not try. My sister is a nurse and was not given a test when she got sick. The total positive cases number is not at accurate.

[JCMach1]

11. My dad's cousin in Florida died the day before she went for testing.

Be very careful

[uponit7771]

12. We have less than .5% testing rate per million in the US. Unless we're par with Germany, Hong Kong

... Greenland, S Korea how do we know we're getting in or out of the soup?

Of course our infection rate goes down when we're not testing enough people in large or medium cities.

Thx in advance for any input on this and VERY much appreciate and follow your numbers

[Ms. Toad]

13. Were not getting out of it just easing off the accelerator

There is still a whole heap of hurt coming our way. But we've slowed our acceleration enough that instead of slamming dead on into a brick wall straight on, we might be able to turn it into a glancing blow.

But testing has gotten better, we're testing more of the ones that are likely to die, since many states are testing people sick enough to be hospitalized more than those with few symptoms, and the curve for the number of deaths is also turning down. If the curve was limited by testing, I would have expected it to still be accelerating as fast as the actual new case curve (which we can't see because we aren't testing).

No guarantees. But I'm hopeful.