Games with Models, 6/3/20 Data

COVID19-hospitalizationsandprojections-060320-featured

And… the positive news continues on the COVID-19 front.

The number of new positive cases was up 107, while the number of active cases (if we assume an average of two-week infections) went down 106.  In other words, roughly speaking, twice as many people got better yesterday than tested positive.

Reported deaths increased by 10, but once again, that included revisions going back a while.  On a day-to-day basis, we’re still at 5 or less over the last few days, and regularly in the single digits for the past couple weeks.  Two days ago (the last available for hospital information), zero patients died in the hospital, as far as we know at the moment.

That fact, as well as the fact that the number of people discharged from hospitals fell by more than half the day before help account for the modest improvement in hospital numbers, despite the fact that only seven new cases checked in.  (That’s the first single-digit admissions day since March.)

COVID19-hospitalizationsandprojections-060320

(See here for my original methodology and here for a subsequent modification I made.)

Projections versus actuals (date of report).

  • Cases:
    • Projection for 6/3: 15,199
    • Actual for 6/3: 15,219
    • Projection for 6/4: 15,302
  • Hospitalizations:
    • Projection for 6/3: 178
    • Actual for 6/3: 189
    • Projection for 6/4: 181
  • Deaths:
    • Projections for 6/3: 736
    • Actual for 6/3: 742
    • Projection for 6/4: 746


  • Mario

    It seems like the recovery is much more orderly now. I have the last official death (#850 or so) coming in on July 8th. Two weeks ago I thought this would end right at the end of June, so all things considered it is looking very stable. On the other hand I also thought we would get by with fewer than 700 deaths back then. If this ends the way I think it will, I think the government process in this state, as a whole, was a tremendous success, but even then the number of deaths far exceed any reasonable early estimate. It isn’t as bad as it would have been with a no-lockdown approach, but it is really bad. We protected the vast majority of the state, but the ones that weren’t adequately protected were absolutely hammered. I think we saw maybe 2.5x the number of deaths you would expect if the virus were similarly prevalent but more equally spread.

    I am also worried about a protest-related resurgence (it would show up roughly 8 days from now), but that seems like it will be less of a concern for RI than other major cities. It’s not not a problem, but not much we can do about that but wait and see.

    • sandie

      i know nothing about models and numbers, i only want to know when is it pk to see my boyfriend, who is extremely cautious and inky goes to the store, while protecti g my 91 year old mother i live with and care for.

      • Justin Katz

        That’s one of those very personalized questions. People with regular contact with the elderly are going to have to be more cautious for longer than those without it. So, just as in everyday life but moreso, it’s a question of everybody’s priorities involved and everybody’s willingness to take precautions.