“'It's tough to make predictions, especially about the future'”

[Chartreuse Wombat]

Forecasting for COVID-19 has failed - International Institute of Forecasters

COVID-19 is a major acute crisis with unpredictable consequences. Many scientists have struggled to make forecasts about its impact [1]. However, despite involving many excellent modelers, best intentions, and highly sophisticated tools, forecasting efforts have largely failed.

forecasters.org

---

Members don't see this ad.

 

[elementaryschooleconomics]

Forecasting for COVID-19 has failed - International Institute of Forecasters

COVID-19 is a major acute crisis with unpredictable consequences. Many scientists have struggled to make forecasts about its impact [1]. However, despite involving many excellent modelers, best intentions, and highly sophisticated tools, forecasting efforts have largely failed.

forecasters.org

"Failure in epidemic forecasting is an old problem. In fact, it is surprising that epidemic forecasting has retained much credibility among decision-makers, given its dubious track record."

I mean...is it really that surprising? We use the best tools we have - people forget that our tools aren't actually that good, that we're barely 100 years out from using cocaine to cure ghosts. But it's better than nothing.

Except in the case of the 2010 Ben Smith paper projecting a shortage of RadOncs. That was actually worse than nothing.

 

[Mandelin Rain]

we're barely 100 years out from using cocaine to cure ghosts.

I still do that. Has something changed? Link?

 

[elementaryschooleconomics]

I still do that. Has something changed? Link?

MDACC has a retrospective review with 56 patients showing crystal meth cures ghosts faster and with less recurrence, p=0.002

 

[Mandelin Rain]

MDACC has a retrospective review with 56 patients showing crystal meth cures ghosts faster and with less recurrence, p=0.002

High Grade 3/4 dental toxicity though.

 

[medgator]

I still do that. Has something changed? Link?

Knew it was schedule II for a reason...

 

[Lamount]

Did anyone really have high hopes for these things being accurate? We can't predict the weather beyond 7 days... and that doesn't even involve human psychology or politics.

 

[ramsesthenice]

MDACC has a retrospective review with 56 patients showing crystal meth cures ghosts faster and with less recurrence, p=0.002

You have to read the methods. 87% of their cohort had teeth. It’s not generalizable to the average meth user.

As to the original post, no sh@t they were bad. Disease modeling requires so many variables that it is inherently garbage in = garbage out. Now we are talking about a novel disease with even more unknowns and an impressively chaotic national leadership. I whole heartedly agree that it is unfathomable why people still put so much stock into them being anything more than a best guess.