Updated IHME model now predicts only 957 deaths in Texas

28,000 Views | 152 Replies | Last: 5 yr ago by Stymied
Keegan99
How long do you want to ignore this user?
AG
https://covid19.healthdata.org/united-states-of-america/texas

This is the 4/16 update, and is a drastic cut from the 4/12 model, which projected 2,200+ deaths in The Lone Star State.
Keegan99
How long do you want to ignore this user?
AG
Also, the peak in Texas was moved forward to roughly this weekend.
HouAggie2007
How long do you want to ignore this user?
Whoa, on the backside of that peak!
74Ag1
How long do you want to ignore this user?
AG
Fantastic thanks
scoodogg
How long do you want to ignore this user?
AggieAuditor
How long do you want to ignore this user?
AG
I feel like these model experts are just running a goal seek in excel.

Expert 1: "Our models suggest 5,000 will die in Texas"

Expert 2 "Did you refresh formulas?"

Expert 1 "Our models now predict 900 will die in Texas*"

*based on a range of 10 to 50,000,000
ham98
How long do you want to ignore this user?
Look at hospital usage. We need to put our medical personnel back to work as PPE needs are met.
Captain Pablo
How long do you want to ignore this user?
AG
Keegan99 said:

https://covid19.healthdata.org/united-states-of-america/texas

This is the 4/16 update, and is a drastic cut from the 4/12 model, which projected 2,200+ deaths in The Lone Star State.
Not a Bot
How long do you want to ignore this user?
AG
ham98 said:

Look at hospital usage. We need to put our medical personnel back to work as PPE needs are met.


Yes. Our same day surgery and operating room folks are on mandatory paid time off but are facing furloughs without pay if we don't get this going very soon. Some hospitals are already there.

but it isn't just healthcare workers, it's everyone that needs to get back to work.
txaggie2010
How long do you want to ignore this user?
How is it remotely possible that there are no more deaths after May 20th? Even if we were sheltering in place, people will continue to catch it and die.
thelaw4
How long do you want to ignore this user?
AG
The model is based off sheltering in place throughout the month of May...
dermdoc
How long do you want to ignore this user?
AG
Keegan99 said:

https://covid19.healthdata.org/united-states-of-america/texas

This is the 4/16 update, and is a drastic cut from the 4/12 model, which projected 2,200+ deaths in The Lone Star State.
Great news
No material on this site is intended to be a substitute for professional medical advice, diagnosis or treatment. See full Medical Disclaimer.
TxAG#2011
How long do you want to ignore this user?
I am not sure how they could be projecting a complete shutoff in COVID deaths. The virus will still be circulating with a lockdown, and even moreso when the lockdown gets reduced.
RandyAg98
How long do you want to ignore this user?
AG
That is great news. It says
Quote:

Current social distancing assumed until infections minimized and containment implemented
Does it assume we continue our current social distancing for this model? Til what date? Seems vague.
txaggie2010
How long do you want to ignore this user?
In order to have no deaths in 4 weeks, there would have to be almost no cases starting in 2 weeks. That obviously would not have happened even with shelter in place.
AG81xx
How long do you want to ignore this user?
AG
Whoever is building these models aren't backing up and using common sense to check some of the state models.....look at the Wyoming projections. So far there has only been 2 deaths in Wyoming , but they are projecting 243 by August 4th. With Wyoming's population of only ~500,000 that equates to a death rate of ~500 per million... way out of line with other states..... puts them on par with New Jersey right now. When the most populated city is only 65,000, it doesn't given come close to passing a smell test.

Be careful what experts tell you.
MasterAggie
How long do you want to ignore this user?
AG
We have not peaked nor will we this weekend.
Keegan99
How long do you want to ignore this user?
AG
Based on...?
Dad
How long do you want to ignore this user?
AG
I'm one of those people that thinks the virus death rate is way overestimated, but I think it kills a lot more than that number and I don't think the deaths will fall to almost nothing so soon like that.

It doesn't make sense to me.
Keegan99
How long do you want to ignore this user?
AG
If you want a head scratcher, look at the IHME model for Wyoming.

https://covid19.healthdata.org/united-states-of-america/wyoming

Wyoming has about 300 current cases and 2 current deaths. They do not have anything close to exponential growth in their reported case counts. Their 7 day moving average of new cases is under 10 per day and is declining. (https://www.nytimes.com/interactive/2020/us/wyoming-coronavirus-cases.html)

Nevertheless, the IHME model projects an eye popping 240+ deaths in Wyoming. Or about 1/4 as many as Texas, which has 50x the population.

I'm thinking the model really struggles with rural areas and accounting for Wyoming's lack of a stay-at-home order. Which isn't needed because, ya know, it's Wyoming.
AG81xx
How long do you want to ignore this user?
AG
Check my similar post 3 or 4 up. I lived in Wyoming for 16 years and there are zero similarities between it and NYC.
The models need to take into account such things as use of mass transportation and living density. Half of the pandemic is NYC and New Jersey.
BowSowy
How long do you want to ignore this user?
AG
Could it be that the model sees that Wyoming hasn't put in place a stay at home order and extrapolates the data from that? I'd bet it's looking at growth without a stay at home order and applying it to their population but isn't fully accounting for the fact that they have such a sparse population density.

I will say that this model is somewhat questionable to me. I've been loosely following it for Texas and its peak date has been all over the place the past few weeks, even when things haven't changed much here.
txaggie2010
How long do you want to ignore this user?
The error bar is equally ridiculous. We know with much more certainty what will happen tomorrow vs 4 weeks from know yet it has a huge error band for tomorrow and tiny one for 4 weeks later. This models says they have almost absolute certainty 0 will be killed in 4 weeks.
The model for New York looks the same as Texas. Going from 800 deaths a day to 0 4-weeks from now with absolute certainty. Crazy!
Dad
How long do you want to ignore this user?
AG
txaggie2010 said:

The error bar is equally ridiculous. We know with much more certainty what will happen tomorrow vs 4 weeks from know yet it has a huge error band for tomorrow and tiny one for 4 weeks later. This models says they have almost absolute certainty 0 will be killed in 4 weeks.
The model for New York looks the same as Texas. Going from 800 deaths a day to 0 4-weeks from now with absolute certainty. Crazy!

The model is either way off or they know something we don't know... maybe the virus plans on taking a vacation in the summer. I would think it would continue to spread and keep killing people for the rest of 2020, just at a lower daily rate than we see now hopefully. It seems like this herd immunity we all are hoping or would a year or two away and not four more weeks.
Keegan99
How long do you want to ignore this user?
AG
"As of Saturday, 1,321 coronavirus patients are known to be hospitalized in Texas. That's a decrease of 201 patients from Friday. At least 176,239 tests have been conducted."

From The Texas Tribune on Facebook.
PJYoung
How long do you want to ignore this user?
AG
Keegan99 said:

"As of Saturday, 1,321 coronavirus patients are known to be hospitalized in Texas. That's a decrease of 201 patients from Friday. At least 176,239 tests have been conducted."

From The Texas Tribune on Facebook.

It feels like Texas has peaked already but I guess opening things back up might change that.

We bought the time we need for PPE and hospital capacity, let her rip in Texas and see where that leads us.
cone
How long do you want to ignore this user?
AG
gonna let her rip without increased testing capacity

sounds like a good plan
Bondag
How long do you want to ignore this user?
AG
Keegan99 said:

If you want a head scratcher, look at the IHME model for Wyoming.

https://covid19.healthdata.org/united-states-of-america/wyoming

Wyoming has about 300 current cases and 2 current deaths. They do not have anything close to exponential growth in their reported case counts. Their 7 day moving average of new cases is under 10 per day and is declining. (https://www.nytimes.com/interactive/2020/us/wyoming-coronavirus-cases.html)

Nevertheless, the IHME model projects an eye popping 240+ deaths in Wyoming. Or about 1/4 as many as Texas, which has 50x the population.

I'm thinking the model really struggles with rural areas and accounting for Wyoming's lack of a stay-at-home order. Which isn't needed because, ya know, it's Wyoming.
I have ready articles about the soccer match that was ground zero for Italy, but have seen nothing about the Staples Center.

In Italy they have public transportation, live in multigenerational houses, greet with a kiss, and take common communion.

In California there was an event a day in staples center for like 30 straight days before the shutdown, yet we have not seen the spike they had from one outdoor soccer match.

The models have no idea what they are doing.
PJYoung
How long do you want to ignore this user?
AG
cone said:

gonna let her rip without increased testing capacity

sounds like a good plan


I agree it's not nearly ideal. We are about to find out how deadly this virus really is but Texas looks fine right now, and people aren't going to tolerate being locked down while things look fine.
DadHammer
How long do you want to ignore this user?
AG
PJYoung said:

cone said:

gonna let her rip without increased testing capacity

sounds like a good plan


I agree it's not nearly ideal. We are about to find out how deadly this virus really is but Texas looks fine right now, and people aren't going to tolerate being locked down while things look fine.

Very true. Protests are only going to get worse. People have to keep a roof over their families and food on the table.
aginlakeway
How long do you want to ignore this user?
AG
Agree. But how do protests help? No one thinks gathering in groups is a good idea.
DadHammer
How long do you want to ignore this user?
AG
I didn't say it was a good idea and I am definitely not doing it.

But we are only delaying the inevitable now. The virus is here and we can't hide from it. The bottom line we have to go out. There will be a spike but what is the alternative? We can't stay locked up much longer. I don't think many understand the economic impact and devastation the lock down is going to cause.
CT75
How long do you want to ignore this user?
AG
Bondag said:



In Italy they have public transportation, live in multigenerational houses, greet with a kiss, and take common communion.
I was in Italy last fall and this correct...the lifestyles are completely different. Compared to Italy we HAVE BEEN practicing social distancing for decades! Hell.....compared to NYC we HAVE BEEN practicing social distancing for decades.
aginlakeway
How long do you want to ignore this user?
AG
I actually think people do understand the economic impact.

It's a bad situation.
DadHammer
How long do you want to ignore this user?
AG
Yes it is.

As soon an individual gets laid off, their outlook on the lockdown is going to change quickly.
 
×
subscribe Verify your student status
See Subscription Benefits
Trial only available to users who have never subscribed or participated in a previous trial.