WPRI’s False Death Chart and Games with Models, 5/29/20

RI-COVIDdeaths-perday-7dayaverage-052920-featured

Over the past decade or so, data visualizations have almost become part of the WPRI news department’s brand, with Ted Nesi taking the lead.  The idea of visualizations is to help the public have clarity on complicated and/or emotionally challenging public policy issues.

That’s why it’s especially perplexing to see the reporters using data visualization with false information about COVID-19.  Consider this tweet from Nesi earlier today (emphasis added):

A few stats that stand out to me in today’s @RIHealth data

• Test positivity rate down to a new low (3.7%) on a day with 3,000+ tests reported

• 20 new hospital admissions – bears watching

• 9th straight day with a double-digit death toll

That last bullet just isn’t true.  The “death toll” of a day is the number of deaths on that day.  What WPRI is reporting is the number of deaths being reported each day, and that includes backwards revisions.  From yesterday to today, the state revised death numbers back to May 7, which is three weeks ago.  Nine of the 16 newly reported deaths actually happened two or more weeks ago.

What difference does this make?  Well, let’s visualize some data.  Here’s the WPRI chart that Nesi is using to substantiate his claim.  Note again that they’re charting the deaths on the day that they’re reported by the state, not the day that they occurred:

WPRI-EliSherman-COVIDdeaths-052920
This chart makes it look like deaths were decreasing until about a week ago and are now on the increase… with the 7-day average almost back to its early-May peak.  But that’s simply not the case.  Here is the chart as it appears if we assign the deaths to the day that they actually occurred.

RI-COVIDdeaths-perday-7dayaverage-052920

What is actually happening is that deaths, like hospitalizations, are decreasing.  In fact, the numbers are back to early-to-mid April. Additionally, contrary to WPRI, no single day has seen more deaths than 22, with only two over 20, not the six shown by WPRI’s chart.

To be sure, we should have some skepticism that the numbers will be revised again, but as Rhode Islanders look at the news reports to develop their own “reopening” comfort levels and evaluate the governor’s policies, charts that give the false impression that deaths are on the rise are not helpful.

Turning to the daily update of how well (or poorly!) my predictive model is performing against reported numbers, here’s the chart for today.

COVID19-hospitalizationsandprojections-052920

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

Projections versus actuals (date of report).  Remember that the projected numbers were based on what was reported as of yesterday, so revisions to earlier reports would make the gap bigger.

  • Cases:
    • Projection for 5/29: 14,586
    • Actual for 5/29: 14,635
    • Projection for 5/30: 14,724
  • Hospitalizations:
    • Projection for 5/29: 212
    • Actual for 5/29: 219
    • Projection for 5/30: 208
  • Deaths:
    • Projection for 5/29: 681
    • Actual for 5/29: 693
    • Projection for 5/30: 697


  • Mario

    I have been waiting for this post because it’s the only place I have to correct myself. I think I have been too pessimistic lately, largely because I accidentally neglected to update one of my formulas. I changed it today, actually anticipating that it would make things look worse, and everything largely reverted back to where I had it a week or so ago. Oops. The deaths are still very high, but it is back to ending close to mid-July at 856 deaths, rather than trending inexorably to 1000 as before. I still have no confidence that it’s right, but it’s something to look at, and it fixed some things that I thought looked implausible. The upshot is that testing now seems adequate to effect an end in a way that I couldn’t see before. That was my mistake.

    I see 134 cases tomorrow, assuming almost 3400 tests, 222 hospitalizations (+18, -10, -5), 11 deaths. I also show hospitalizations continuing to drop for a while longer then going back up a week from now (236 on the 6th, which is to say reported on the 8th) before dropping again. I honestly don’t know why, just thought I’d point it out.

    I don’t think the WPRI charts are misleading. They aren’t accurate, maybe, but the real chart is no less inaccurate, since we know that the most recent dates will eventually have higher numbers than they do today. A better chart would probably exclude the most recent days entirely to avoid showing a spurious decrease, but any cutoff would be arbitrary since the daily revisions to all of the numbers have seemingly no relationship to temporal mechanics as I understand it. In any event, if you just exclude the last seven days, the argument that deaths are dropping weakens significantly. I think things are improving (again, mea culpa) but the death chart is the last thing that would show it. The WPRI chart at least has the virtue of being based on something undeniably true, even if only temporarily.

    • Justin Katz

      The chart is absolutely misleading. The title is “7 Day Average of Deaths,” not “7 Day Average of Reports of Deaths.” Those two things are different in a significant way.

  • Lou

    Still nothing on Burrillville? This can be your chance to take a victory lap! NO ONE else called it as a hot spot. Let’s see the numbers.

    Hey, do you think George Will knows about your phony unemployment index? I don’t know how he could write a column like this if he knew the “facts”.

    https://www.providencejournal.com/opinion/20200525/my-turn-george-will-as-vp-raimondo-would-help-restore-adult-supervision-in-washington

    • ShannonEntropy

      If you’re looking for a promising pick in the Joe Biden veepstakes
      because you hate Donald Trump, that’s up to you. But Gina Raimondo has demonstrated some of the worst authoritarian impulses imaginable as Governor, along with conducting business in an ethically questionable fashion and opposing transparency regarding her activities. I hardly think those are the characteristics we’re looking for in someone who could wind up being only a heartbeat away from the presidency. And that’s particularly true when the heart in question is beating in the chest of a man of Joe Biden’s considerable age and dubious grasp on his mental faculties.

      https://hotair.com/archives/jazz-shaw/2020/05/24/george-wills-misguided-endorsement-gina-raimondo/

      p.s. I was “shooting from the lip” in my earlier post and B-ville. Turns out a meat-processing plant there is mostly responsible for the outbreak. For whatever reason, these places tend to be C-19 hotspots:

      https://www.providencejournal.com/news/20200424/about-80-coronavirus-cases-linked-to-burrillville-based-daniele-inc

      • Lou

        Thanks for the info. Are you suggesting we don’t have “authoritarian impulses…along with conducting business in an ethically questionable fashion and opposing transparency regarding…activities” already in the White House?

        • ShannonEntropy

          Let’s face it… if Joe wins whoever he picks as VEEP will be POTUS by 2022

          Yeah, I know…. Orange Man Bad!!

          But replacing him with Queen Gina would be exactly like the Weimar Republic fixing their problems by electing Hitler

          • Christopher C. Reed

            Naah, more like, Argentina fixing their problems by re-electing Christina Kirchner i.e. the b*tch who whelped them.
            But c’mon man…break the Godwin Rule =>lose. Besides, we already have in the WH the worst Hitler in the history of Hitlers, ever! It’s Science. So don’t forget to vote for the Alzheimer’s guy…it’s important.

            Der Führer was wildly popular with the Germans for fixing much of what had been foisted on them by the victors at Versailles, which the WR had been impotent to address.

          • ShannonEntropy

            Why is it that whatever the question, like Should we put Queen Gina in a position where she may very well be the next POTUS ??, the answer from Trump Derangement Syndrome sufferers is always the same:

            ORANGE MAN BAD !!

            C’mon guys !! You gotta do better than that

            p.s. Christopher tried… when Gina was compared to Hitler he noted that Hitler was very popular

            Yer skating on thin ice, Chris… people have lost their careers over saying stuff like that:

            https://www.cnn.com/2020/01/28/us/college-football-coach-hitler-trnd/index.html

          • Christopher C. Reed

            What?! In the home of the brave and land of the free!?Naaah…
            Yeah, that coach def fumbled it…should’ve called a ‘Joe Dzugashvili’ play, he’d be in fine odor now.