Death toll predictions from 19 modelers

[Bongleur]

Graph with a line for each of 19 sources, between May1 to June 1

https://www.sciencemag.org/news/2020...ovid-19-s-toll

Surveys of infectious disease experts aim to predict COVID-19’s toll
By Jeffrey Brainard
Apr. 23, 2020

on average, they estimated that in the most likely scenario, the number of deaths will reach 45,157 on 1 May and 74,631 on 1 June.

The experts’ understanding of the dynamics of disease outbreaks informs their estimates,...

An inspiration for the effort is the Good Judgment Project,
https://goodjudgment.com/
launched by psychologists at the University of Pennsylvania in 2011. It studied ways to improve forecasting using crowdsourcing; its team won competitions sponsored by an organization funded by U.S. spy agencies—the U.S. Intelligence Advanced Research Projects Activity...

He notes that, in the first surveys of disease experts he conducted, the consensus predictions of the number of confirmed U.S. cases were lower than those actually reported later. But as the pandemic has expanded, McAndrew says, the forecasts made by the experts have become increasingle accurate...

[Dtail]

Interesting. So the May 1 average estimate will be off around 40%.
At first estimates in general appeared to be too high, now they seem to be coming in on the lower side.

[Sunset]

It is hard to estimate as it's a pretty dynamic situation.

By the end of today April 30, we will be over 63,000 deaths in USA.

Certainly past all the low estimates I've ever heard.... and it's far from over.

[copyright1997reloaded]

Quote:

Originally Posted by Bongleur Graph with a line for each of 19 sources, between May1 to June 1 https://www.sciencemag.org/news/2020...ovid-19-s-toll Surveys of infectious disease experts aim to predict COVID-19’s toll By Jeffrey Brainard Apr. 23, 2020 on average, they estimated that in the most likely scenario, the number of deaths will reach 45,157 on 1 May and 74,631 on 1 June. The experts’ understanding of the dynamics of disease outbreaks informs their estimates,... An inspiration for the effort is the Good Judgment Project, https://goodjudgment.com/ launched by psychologists at the University of Pennsylvania in 2011. It studied ways to improve forecasting using crowdsourcing; its team won competitions sponsored by an organization funded by U.S. spy agencies—the U.S. Intelligence Advanced Research Projects Activity... He notes that, in the first surveys of disease experts he conducted, the consensus predictions of the number of confirmed U.S. cases were lower than those actually reported later. But as the pandemic has expanded, McAndrew says, the forecasts made by the experts have become increasingle accurate...

Sorry, I find this just about useless. Let me summarize:
The models were off by gads and gads. But don't say that, instead say: " in the first surveys of disease experts he conducted, the consensus predictions of the number of confirmed U.S. cases were lower than those actually reported later. But as the pandemic has expanded, McAndrew says, the forecasts made by the experts have become increasingle accurate."

In my model of stock market predictions, I can now pretty accurately tell you what will happen in April 2020. (My model for predicting April wasn't quite as good when I first put it out in March. )

How about an article showing us how/why the earlier predictiions were so off, or why folks like Neil Ferguson (Infectious disease modeller/epidemiologist. Director of J-IDEA and the MRC Centre for Global Infectious Disease Analysis) refused to make his modeling code available on GitHub. (His excuse was:

Quote:

I’m conscious that lots of people would like to see and run the pandemic simulation code we are using to model control measures against COVID-19. To explain the background - I wrote the code (thousands of lines of undocumented C) 13+ years ago to model flu pandemics...

*Some* of us were calling folks like this out....way back in Mid March.

[6miths]

Yes it looks as if the low and moderate models are mostly on the low side given that US deaths will be nearly 65,000 at the close of the day. And from the CDC data this looks like a significant underestimate of actual deaths. Also, the flatness of most of the curves looks optimistic given the number of new cases and the daily death rates still being in the 2,000 range. And add into that the reduction in containment measures that are starting to happen. We'll see and hope of course.

[ejman]

On April 9th Dr. Fauci said "I believe we are going to see a downturn in that, and it looks more like the 60,000 [range] than the 100,000 to 200,000,".
As of 4/29/20 Worldmeters listed 61,655 deaths. I hope he is right and the number of deaths drop precipitously from this point on but it sure doesn't look like it to my untrained eye.

[audreyh1]

Oh well - crossed 1M cases, crossed 60,000 deaths in late April. Well ahead of some of the projections I had seen. I think it's going to keep going up - as fast? Not sure.