NY Update. Our numbers continue to rise by virtually any metric you choose, but NY still remains low by comparison with other states.

In Westchester, as recently as 6 weeks ago, we were averaging less than 300 new cases per week. As of the past week, we’re now averaging 859 new cases per week. Here’s a chart showing the number of cases diagnosed daily, since March 12 — you can see that after a long period of steady but low spread, we’ve started moving higher over the past month or so.

The testing-positive percentage in Westchester has also been rising, and over the past week it has averaged 1.89%. Yesterday we were at 2.26%, but I tend not to put too much weight on individual days’ results given vagaries of the timing of test results, etc.

The rise is not being caused by more testing — as you can see below, Westchester’s testing numbers have been pretty stable (rising slightly) over the past month:

Current Westchester map of cases (20 active cases in the Pelhams):

New York State as a whole is at around 1.6% positivity for the past week. Here’s a handy chart I had not come across previously that shows the test-positivity numbers by region:

Nationwide, the numbers really are trending in a bad direction. The testing-positivity number nationwide is nearing 7%. We’re adding nearly 600,00 new cases per week, with no signs of slowing. Perhaps more importantly, the lagging indicators — hospitalization, ICU admission and death — are rising to levels we have not seen for a while.

Here’s the hospitalization chart for the US, since 3/12/20:

Here’s the ICU chart:

And finally, daily deaths (the graph looks like an EKG because there are big swings between the reporting during the week and over weekends):

Fatalities lag behind hospitalizations/ICUs, so we’ll likely see the daily fatality rate climb quite a bit over the next several weeks. In fact, if you’ll recall, a few months ago I tried to get a sense of the lag time between new cases being reported and the fatalities that result from those cases. I did so by bracketing a theoretical “case fatality rate” and looking at two-week periods to see when increases/decreases in new cases correlated with increases/decreases in deaths.

Although pretty inexact, my ballparking now tends to show that the deaths reported in a two-week period reflect the surge or decrease in cases the prior two-week period (rather than a longer lag). In the chart below, you’ll see color-coding that matches the predicted range of fatalities if the CFR is somewhere between 1.5% and 2% with actual fatalities:

If the numbers hold true, over the next two weeks we may see fatalities in the range we were seeing in late July and early August (between 15,000 and 20,000). I fear that by mid-November we may be even higher, if current trends persist.

To end on an even more sobering note, on average, since March 12, we have lost almost 950 people per day to COVID in the US.

Stay safe, wear a mask, maintain social distancing, and let’s try to keep this thing under control in NY at least.