# why are deaths dropping but infection rate stabilizing? **URL:** \ **Category:** The Quarantine Zone\ **Tags:** covid-quarantine\ **Created:** [June 16, 2020, 3:38pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779 "2020-06-16T15:38:14Z")\ **Posts on this page:** 20\ **Page:** 3
**Author:** ![RealityCheck71](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/realitycheck71/32/3493_2.png) [@RealityCheck71](https://boards.straightdope.com/u/RealityCheck71)\ **Post date:** [July 8, 2020, 1:40pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/41 "2020-07-08T13:40:54Z")
I don’t know which numbers you’re talking about but almost all countries are showing a largish gap between counted covid deaths and “extra deaths” compared to previous years. ---
**Author:** ![Half\_Man\_Half\_Wit](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/half_man_half_wit/32/21766_2.png) [@Half\_Man\_Half\_Wit](https://boards.straightdope.com/u/Half_Man_Half_Wit)\ **Post date:** [July 8, 2020, 2:21pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/42 "2020-07-08T14:21:37Z")
> [@casdave](#): > > That resource has been consistenty wrong, underplaying predicted death rates by around 50%. Thanks, that’s useful information! I just ended up there today, as it’s linked from the Worldometers-site—if you drill down to the US state info, there’s a new (to me) column titled ‘projections’ that links to the healthdata site. Do you have any source which has similar information, but a better track record? ---
**Author:** ![RealityCheck71](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/realitycheck71/32/3493_2.png) [@RealityCheck71](https://boards.straightdope.com/u/RealityCheck71)\ **Post date:** [July 8, 2020, 2:30pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/43 "2020-07-08T14:30:39Z")
> [@Half\_Man\_Half\_Wit](#): > > Do you have any source which has similar information, but a better track record I doubt it. The reason that model was hated on was because it was consistently low; others were mostly too high and no more accurate. > **[Comparing Covid-19 models](https://flowingdata.com/2020/05/05/comparing-covid-19-models/)** > > FiveThirtyEight compared six Covid-19 models for a sense of where we might be headed. With different assumptions and varying math, the trajectories vary, but they at least provide clues so that pol… On May 5, the IHME was one of few whose range covered reality. ---
**Author:** ![RealityCheck71](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/realitycheck71/32/3493_2.png) [@RealityCheck71](https://boards.straightdope.com/u/RealityCheck71)\ **Post date:** [July 8, 2020, 2:36pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/44 "2020-07-08T14:36:54Z")
Retract that last line! It was a misread from trying to hurry the edit. ---
**Author:** ![Monocracy](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/monocracy/32/4493_2.png) [@Monocracy](https://boards.straightdope.com/u/Monocracy)\ **Post date:** [July 8, 2020, 2:41pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/45 "2020-07-08T14:41:10Z")
> [@RealityCheck71](#): > > Right, and all the other models were plugging random variables into their models and wildly over predicting, for the most part. Do you have a citation for this statement? Because in a [538 article](https://fivethirtyeight.com/features/best-case-and-worst-case-coronavirus-forecasts-are-very-far-apart/) from Apr. 2, the average of the experts’ predictions was 263,000 U.S. coronavirus deaths in 2020. The White House Coronavirus Task Force estimated between 100,000 and 240,000 (total?) deaths if federal guidelines where followed. From the chart in that article, about half of the 15 experts’ best estimate were between 100,000 and 240,000. Six of those best estimates have been surpassed, and a seventh will be soon. Only four experts’ best estimate was above 250,000 (actually five, but 538 eliminated an expert that predicted 1 million), and only two were between 500,000 and 1 million. And if you go back even further, to [an article from March 20](https://fivethirtyeight.com/features/infectious-disease-experts-dont-know-how-bad-the-coronavirus-is-going-to-get-either/), the average of the experts’ best estimates was 200,000 deaths and 14 of the 18 experts predicted under 250,000 deaths. There was one number of 2.2 million that was being bandied about early on, but that was based on an assumption of no changes in American lifestyle, and my belief is that would have been an accurate number. This is my first post on the new boards so forgive any formatting gaffes (no preview?). ---
**Author:** ![RealityCheck71](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/realitycheck71/32/3493_2.png) [@RealityCheck71](https://boards.straightdope.com/u/RealityCheck71)\ **Post date:** [July 8, 2020, 3:03pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/46 "2020-07-08T15:03:10Z")
I’ll try to research that but I was talking shorter term predictions like 2-4 weeks. Models that talk about predicting how second and third waves will go, ie. through next flu season and to the end of the year, aren’t worth reading. imho. ---
**Author:** ![RealityCheck71](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/realitycheck71/32/3493_2.png) [@RealityCheck71](https://boards.straightdope.com/u/RealityCheck71)\ **Post date:** [July 8, 2020, 3:15pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/47 "2020-07-08T15:15:10Z")
I should also point out, that 538 article isn’t really a prediction model as such. It’s a survey of experts’ opinions. In one of the other articles in the series they specifically mentioned less than half of the survey answers are based on models the experts were working on. ---
**Author:** ![HMS\_Irruncible](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/hms_irruncible/32/7394_2.png) [@HMS\_Irruncible](https://boards.straightdope.com/u/HMS_Irruncible)\ **Post date:** [July 8, 2020, 3:30pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/48 "2020-07-08T15:30:35Z")
I wonder if a big contributor to the death rate is because the quality of treatment suffers when medical facilities get overwhelmed. Errors, shortages, cross-contagion, etc. ---
**Author:** ![RealityCheck71](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/realitycheck71/32/3493_2.png) [@RealityCheck71](https://boards.straightdope.com/u/RealityCheck71)\ **Post date:** [July 8, 2020, 4:07pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/49 "2020-07-08T16:07:56Z")
@Monocracy Heh. Just as a coincidence, but I found that older 538 expert survey article I was thinking of and it mentions the above disdained IHME model. > [@](#): > > The expert consensus is that the U.S. will have reported around 47,000 COVID-19 deaths by May 1, with a 90 percent chance of having between 32,000 and 82,000. > > This is very close to the latest forecast from the [Institute for Health Metrics and Evaluation](https://covid19.healthdata.org/united-states-of-america) at the University of Washington, which projects roughly 51,000 deaths by May 1, with an uncertainty range from about 28,000 to 112,000. (Actual fatalities on May 1 were ~66k, an example that they didn’t “consistently” underpredict) > **[Experts Think The U.S. COVID-19 Death Toll Will Hit 50,000 By The End Of April](https://fivethirtyeight.com/features/experts-think-the-u-s-covid-19-death-toll-will-hit-50000-by-the-end-of-april/)** > > The expert consensus is that reported deaths will increase to around 47,000 by May 1, although they think there could be as many as 82,000 by that time. ---
**Author:** ![Shalmanese](https://avatars.discourse-cdn.com/v4/letter/s/45deac/32.png) [@Shalmanese](https://boards.straightdope.com/u/Shalmanese)\ **Post date:** [July 8, 2020, 4:58pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/50 "2020-07-08T16:58:25Z")
What specifically doesn’t add up? People have been back to work since April, traffic levels are normal, pollution levels are higher than they were this time last year, Shanghai Disneyland is open, Starbucks is open, most Chinese are going about their lives as if everything is ok except when there’s a publicized outbreak in which case, there’s a massive, instantaneous, overnight reaction. There was suspicion that imported salmon caused the most recent outbreak and people across the entire country stopped buying salmon literally overnight. That doesn’t sound like the reaction of a peoples who are trying to hide an ongoing epidemic. The simple reason why they wouldn’t lie about it is, as civilizations have learnt time and time again over history, you can’t lie your way out of an epidemic. By many measures, China’s been extraordinarily more transparent about this epidemic than almost any other country. By late Jan, they started publishing the location histories of every single confirmed case outside of Hubei on official websites and by early Feb, you could open up the Chinese equivalents of Google maps and see a dot for every place a confirmed case has travelled as well as advice on where to get tested if you crossed paths with them. That’s not something you can just spin up on a whim if you want to hide fake data. We’re over 3 months into China being back to normal and people still refuse to accept that the pandemic there went largely as the official government narrative portrayed. ---
**Author:** ![Monocracy](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/monocracy/32/4493_2.png) [@Monocracy](https://boards.straightdope.com/u/Monocracy)\ **Post date:** [July 8, 2020, 5:18pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/51 "2020-07-08T17:18:26Z")
> [@RealityCheck71](#): > > (Actual fatalities on May 1 were ~66k, an example that they didn’t “consistently” underpredict) That article was posted on Apr 17 making predictions for May 1. There were ~35k deaths as of Apr 16, so IMHE was predicting 16k more deaths. In actuality, there were 31k more deaths. I would call that an underprediction. But to be fair, every expert under predicted in that circumstance. Which also counters your point that every other model has been over predicting deaths. As for models vs predictions, I don’t see a meaningful distinction for the purposes of our discussion. ---
**Author:** ![Monocracy](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/monocracy/32/4493_2.png) [@Monocracy](https://boards.straightdope.com/u/Monocracy)\ **Post date:** [July 8, 2020, 5:30pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/52 "2020-07-08T17:30:49Z")
As for the OP, I think another factor is that people are underestimating the lag time between events and changes in policies and the reported cases and deaths. It takes time for the the virus to work its way through a population before you start to see the effects it’s having. ---
**Author:** ![RealityCheck71](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/realitycheck71/32/3493_2.png) [@RealityCheck71](https://boards.straightdope.com/u/RealityCheck71)\ **Post date:** [July 8, 2020, 6:31pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/53 "2020-07-08T18:31:12Z")
> [@Monocracy](#): > > As for models vs predictions, I don’t see a meaningful distinction for the purposes of our discussion I do. Those experts are anonymously guesstimating. Models are plugging formulas with variables that can be adjusted with new information. I don’t want to talk about guesstimates, personally. ---
**Author:** ![RTFirefly](https://avatars.discourse-cdn.com/v4/letter/r/c77e96/32.png) [@RTFirefly](https://boards.straightdope.com/u/RTFirefly)\ **Post date:** [July 10, 2020, 11:22am UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/54 "2020-07-10T11:22:20Z")
Re: thread title, the deaths aren’t dropping anymore. Per Worldometer, the 7-day rolling average of U.S. Covid-19 deaths is back up to 625 deaths per day, compared with 627 deaths/day on June 22. That stat had been in the 550s before the Fourth of July weekend. We’re screwed. ---
**Author:** ![RitterSport](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/rittersport/32/6326_2.png) [@RitterSport](https://boards.straightdope.com/u/RitterSport)\ **Post date:** [July 10, 2020, 12:41pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/55 "2020-07-10T12:41:32Z")
Were they ever dropping, if you removed NY and NJ from the calculation? ---
**Author:** ![casdave](https://avatars.discourse-cdn.com/v4/letter/c/c6cbf5/32.png) [@casdave](https://boards.straightdope.com/u/casdave)\ **Post date:** [July 10, 2020, 1:11pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/56 "2020-07-10T13:11:50Z")
I seem to remember Michigan & Massachusetts also having a relatively high death rates early on and this has dropped markedly too. Most other places were just bumping along, it would be more accurate to say they were steadily brewing, like any exponential curve nothing much seems to happen for some time and then it appears to come from nowhere – but this is entirely due to a lack of understanding - add in identity politics and the related ignorance or lack of respect for scientific values and you can easily see how complacency can lead to tragedy - just look at UK. It really has only just got into second gear in the US, the current death rate just climbed to almost 1000 per day again, but this is based on a reservoir of infections acquired three or four weeks ago when it was running at around 25k per day, its now running at over 60k per day and this will be reflected in deaths in the next thee or four weeks. I think further complacency by the science deniers was fueled by sheer willful ignorance of the time lag between infection and death, which was used to suggest that the mortality rate was falling to lower than 1% and infections were rising only because of more widespread testing. That is not a view that can be taken until after the event when irrefutable evidence is avaiable, the only sensible strategy is to deal with the situation as it has shown up to now and not go off into the realms of wishful thinking. ---
**Author:** ![RealityCheck71](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/realitycheck71/32/3493_2.png) [@RealityCheck71](https://boards.straightdope.com/u/RealityCheck71)\ **Post date:** [July 10, 2020, 1:15pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/57 "2020-07-10T13:15:54Z")
Someone can do the math but I would guess, yes it was going down in the Rest of America (ROA). Those two states seem to have been a small part of the numbers since the end of May. > **[An interactive visualization of COVID-19 | 91-DIVOC](https://91-divoc.com/pages/covid-visualization/?chart=countries&highlight=United%20States&show=highlight-only&y=both&scale=linear&data=deaths-daily&data-source=jhu&xaxis=left&extra=New%20York%2CNew%20Jersey#countries)** > > An interactive, data-forward visualization of COVID-19 data by Prof. Wade at The University of Illinois. Updated daily. I don’t know if it’s useful to think of the US as a single entity for this anymore than the different timing of surges in Europe. ---
**Author:** ![Great\_Antibob](https://avatars.discourse-cdn.com/v4/letter/g/e47c2d/32.png) [@Great\_Antibob](https://boards.straightdope.com/u/Great_Antibob)\ **Post date:** [July 10, 2020, 1:39pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/58 "2020-07-10T13:39:59Z")
> [@RitterSport](#): > > Were they ever dropping, if you removed NY and NJ from the calculation? Yes they were but, as you might guess, the US numbers are not as encouraging if you take those states out of the statistics. You can look at state level numbers, and deaths were generally falling across the nation on a state by state level. The national numbers are now climbing again, mainly due to the sunbelt states. They more than offset the still falling death counts in the NE states. Even in states like Texas or Florida, deaths appeared flat or, at worst, slightly increasing, only a week ago, so it wasn’t just a matter of combining statistics. Just more evidence that deaths, as an indicator, lag far, far behind other metrics as a useful gauge in evaluating virus containment. ---
**Author:** ![RitterSport](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/rittersport/32/6326_2.png) [@RitterSport](https://boards.straightdope.com/u/RitterSport)\ **Post date:** [July 10, 2020, 1:51pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/59 "2020-07-10T13:51:26Z")
Thanks, all! ---
**Author:** ![RealityCheck71](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/realitycheck71/32/3493_2.png) [@RealityCheck71](https://boards.straightdope.com/u/RealityCheck71)\ **Post date:** [July 10, 2020, 2:02pm UTC](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779/60 "2020-07-10T14:02:42Z")
> [@Great\_Antibob](#): > > Just more evidence that deaths, as an indicator, lag far, far behind other metrics as a useful gauge in evaluating virus containment Well, we didn’t need evidence. Obviously death lags behind catching the disease. But it’s my understanding that positive tests lag about 2 weeks and deaths about 3 weeks. On a lark, I did another chart like the one above only instead of showing USA/NY/NJ it shows Europe/Spain/Italy. Not crazily different shapes: > **[An interactive visualization of COVID-19 | 91-DIVOC](https://91-divoc.com/pages/covid-visualization/)** > > An interactive, data-forward visualization of COVID-19 data by Prof. Wade at The University of Illinois. Updated daily. [Previous page](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779.md?page=2) [Next page](https://boards.straightdope.com/t/why-are-deaths-dropping-but-infection-rate-stabilizing/855779.md?page=4)