Not Your Demographic

Assertions so counter intuitive and nonsensical - they must be true!

Cogent insights by Benjamin Scherrey

Prev Next


MORE TESTS, FEWER POSITIVES, FEWER DEATHS. Some promising news indeed. It looks like those long promised test capacities are really happening. The number of tests completed yesterday was 223,690 - way more than the ~140k/day we've been stuck at since about April 2nd. Other good news is that the % positive test rate has dropped as well now to 14% after hanging just under 20% for the last 3 weeks. Does this mean we're testing more asymtomatic people or that there's just fewer actual new cases? Too soon to say but all curves, however slight, point downwards. Unfortunately now we're entering into weekend time so we expect to see exagerated downward curves followed by a bump up on Tuesday and so it'll be Wednesday before we likely have our next snapshot for which we'd have any real confidence.

For Friday, rates of growth (rog) for new incidents is and deaths are lower. Our new total for incidents is 890,816 (+31,498 - on track for 1Million on Apr 28), an increase of 3.67% over the previous day (-0.25% from yesterday). Deaths have totaled 45,786 (+1,772) which is an increase of 4.03% (-0.51%) over the previous day and on track for 50,000 tomorrow although weekend reporting latency might delay this til Sunday. Since the peak in Mar 19, rates of growth for incidents has fallen by 47.97% and for deaths has fallen by 38.83%.

Some rather frustrating news, not because I think it's at all accurate, but rather because it seemed to be on track until they completely re-worked the model which now bears no resemblence to any actual data I can find. now shows all but ten states above the Rt 1.0 level and presumably rising. Since this is a lagging indicator (unless they changed that aspect of the model too, I'll have to check), it would seem to be counter to what the data from demonstrates and THIS IS THE SAME DATA SOURCE THEY USE! Their data is entirely based on reports of new incidents, not deaths. If their data was correct, however, we should see rogs for deaths and actual daily death counts going higher - significantly higher. Instead, states like Arkansas are the 2nd worst according to the Rt level yet Arkansas hasn't seen a single death in days! If I have time this weekend I'll try to roll back their model and reproject to see if their original model fits what the data is showing. Either I'm wrong and we're about to see a 2nd wave to dwarf the first, or the data I'm seeing is right and this thing is slowly but surely in a real decline.

Meanwhile, I hear rumours of states starting to kick off random anti-body testing. I really hope this is true as we desperately need it.

Prev Next