23,000 Texans dead by end of August?

15,632 Views | 94 Replies | Last: 3 yr ago by PJYoung
The Fall Guy
How long do you want to ignore this user?
AG
https://cbsaustin.com/news/local/ut-model-predicts-over-23000-covid-19-deaths-in-texas-by-end-of-august
Aggies2009
How long do you want to ignore this user?
AG
Oh hey. Another model. They've been SO accurate so far!
jeffdjohnson
How long do you want to ignore this user?
Seems like a poor analysis. Detected cases are already starting to dip down. Deaths per day in Texas never even reached 300 per day at peak. Even if we hit ~300 per day (seems unlikely) that would only add ~6000 more total deaths by the end of August. That would still be way short of this projection.
DeBoss
How long do you want to ignore this user?
AG
Can you help me understand how masks are working if deaths are spiking?
- If you are going to do something stupid, be smart about it.
Old RV Ag
How long do you want to ignore this user?
AG
DeBoss said:

Can you help me understand how masks are working if deaths are spiking?
You're assuming everyone is wearing a mask. Kind of like a condom - having it stay in your wallet doesn't do any good.
beerad12man
How long do you want to ignore this user?
AG
https://covid19-projections.com/us-tx

Way too high. This says 19549 by Nov 1. Much more reasonable and accurate. two different graphs coming up with numbers so drastically different are why it's so hard to believe anything right now. It's just unfathomable to think they could come up with such different conclusions this far into this thing, but that's where we are.

That would take 714 deaths per day. We aren't at half that, and cases begin to trend down a couple of weeks ago. Deaths should start to trend down in the next week or so.

The only way this could be even close is if there are 10k deaths we don't know about that will get added. Ones that occured over the last few months. Otherwise this makes no sense, and the covid19 projects seems far more reasonable.
GAC06
How long do you want to ignore this user?
AG
Old RV Ag said:

DeBoss said:

Can you help me understand how masks are working if deaths are spiking?
You're assuming everyone is wearing a mask. Kind of like a condom - having it stay in your wallet doesn't do any good.


Japan is seeing a lot of new cases despite the use of masks.
Old RV Ag
How long do you want to ignore this user?
AG
GAC06 said:

Old RV Ag said:

DeBoss said:

Can you help me understand how masks are working if deaths are spiking?
You're assuming everyone is wearing a mask. Kind of like a condom - having it stay in your wallet doesn't do any good.


Japan is seeing a lot of new cases despite the use of masks.
Yes, but not to the same level. They also have >10x population density. Still supports that masks help.
Aggies2009
How long do you want to ignore this user?
AG
Old RV Ag said:

DeBoss said:

Can you help me understand how masks are working if deaths are spiking?
You're assuming everyone is wearing a mask. Kind of like a condom - having it stay in your wallet doesn't do any good.
Considering stores don't let people in without them, they're being enforced pretty hard.

Yet cases didn't drop.
P.U.T.U
How long do you want to ignore this user?
AG
GAC06 said:

Old RV Ag said:

DeBoss said:

Can you help me understand how masks are working if deaths are spiking?
You're assuming everyone is wearing a mask. Kind of like a condom - having it stay in your wallet doesn't do any good.


Japan is seeing a lot of new cases despite the use of masks.
A lot is relative, they never really had a huge spike that lasted weeks. They rarely have had more than 5 deaths a day since May

So many fear mongering articles right before school that are not based on solid models or facts.
dragmagpuff
How long do you want to ignore this user?
AG
If you look at the actual model, it is trying to tie to the 7-day average for Texas. The issue is that there was the one day where TX reported like 700+ "missed" deaths in one day. This tripled the 7-day average instantaneously.

As to why the deaths are continuing to grow in their model, they are going of cell data as a stand in for social distancing.

Bruce Almighty
How long do you want to ignore this user?
AG
All of these models have been wrong during this entire pandemic. They've been too high, adjusted, then too low, and adjusted again. Texas isn't going to more than double their deaths in 3 weeks. That's asinine. 23,000 by end of August 2021 would be more realistic than this.
Old RV Ag
How long do you want to ignore this user?
AG
Aggies2009 said:

Old RV Ag said:

DeBoss said:

Can you help me understand how masks are working if deaths are spiking?
You're assuming everyone is wearing a mask. Kind of like a condom - having it stay in your wallet doesn't do any good.
Considering stores don't let people in without them, they're being enforced pretty hard.

Yet cases didn't drop.
But going to a party or gathering after without one. Kind of like wearing a condom with one girl and not wearing one with another and wondering why girl two is pregnant (or you've got an STD) - but I wore a condom!
Harry Stone
How long do you want to ignore this user?
AG
Bruce Almighty said:

All of these models have been wrong during this entire pandemic. They've been too high, adjusted, then too low, and adjusted again. Texas isn't going to more than double their deaths in 3 weeks. That's asinine. 23,000 by end of August 2021 would be more realistic than this.


It's simple, put out an overreaching prediction to scare everyone. When it doesnt happen,ignore the original prediction cause nobody will give a *****
culdeus
How long do you want to ignore this user?
AG
Are they using excess mortality maybe? Otherwise hard to see it getting that big.
beerad12man
How long do you want to ignore this user?
AG
Old RV Ag said:

GAC06 said:

Old RV Ag said:

DeBoss said:

Can you help me understand how masks are working if deaths are spiking?
You're assuming everyone is wearing a mask. Kind of like a condom - having it stay in your wallet doesn't do any good.


Japan is seeing a lot of new cases despite the use of masks.
Yes, but not to the same level. They also have >10x population density. Still supports that masks help.
Does it? Was that the only variable? Need more information. Did they lockdown harder? Are they counting/reporting things the exact same way? Are they in better health than we are? Have they had as many protests? As much travel into their country? Does being an island make a difference? I really don't know. They are wearing masks now and seeing the spike. Who knows where they will be in 6 weeks. They may just be 3-4 months behind us and have to go through it all eventually anyways.

I've always cautioned people to point to one success in this and say that's what we should have done. What they did might not have worked as effectively here.

Either way, this was never about a competition. This was about not overwhelming the hospitals. After a scare in NYC, we have succeeded. To this day, some of us still would rather have the freedoms than added safety. If this were truly a human extinction level event, or even a 5-10% IFR rate, you may have more on board. For 0.25% or 0.65%, not as many fall in line thinking it's worth it to society as a whole to do what we have done to it. Even if others think we are monsters for doing so. It's how we are wired.
culdeus
How long do you want to ignore this user?
AG
Excess death figures look a little scarier now than they did a few weeks ago: Enough so that you can see where there is nearly certainly some undercounting. However, you need to apply a ton of these deaths over to get where this number is.

https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm

Looking forward, deaths and hospitalization has dropped so much for all causes too, so you need another huge 3rd? wave to pull this off?

beerad12man
How long do you want to ignore this user?
AG
culdeus said:

Excess death figures look a little scarier now than they did a few weeks ago: Enough so that you can see where there is nearly certainly some undercounting. However, you need to apply a ton of these deaths over to get where this number is.

https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm

Looking forward, deaths and hospitalization has dropped so much for all causes too, so you need another huge 3rd? wave to pull this off?


Maybe I'm reading it differently, or just being optimistic, I'm reading this as lowering over time, meaning that some of the deaths from Covid might very well have been from the expected numbers any given year to begin with. It seemed to be way above the line on April 11th, but now, getting closer and closer to normal. I also don't think you can conclude that this means there is nearly certainly some undercounting even if we are in the excess. Any given year can change by 100k or so in the US, and......

Wouldn't you also have to factor in more potential deaths due to suicide, drug overdose, alcohol, violent crimes, other health issues neglected or cancelled due to Covid, etc. many other reasons people may have died that weren't because they contracted covid, but rather because of a covid world associated with lockdowns and the fear driven hysteria created by the media.

On the other hand, you may have less potential accidents or flu / pneumonia deaths from distancing and less people on the roads.

DadHammer
How long do you want to ignore this user?
AG
How many of those were already expected to die

In the next four months
six months

According to actual counts that is actually the case. People close to death were killed months before they were expected to pass.

Though sad, it should not be used to Scare the population
Charpie
How long do you want to ignore this user?
AG
Lots of models work IF you don't do any behavior modifications.

Texas cases will go down. However, I don't see border cases going down. Right now, it appears they are plateauing.
culdeus
How long do you want to ignore this user?
AG
beerad12man said:

culdeus said:

Excess death figures look a little scarier now than they did a few weeks ago: Enough so that you can see where there is nearly certainly some undercounting. However, you need to apply a ton of these deaths over to get where this number is.

https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm

Looking forward, deaths and hospitalization has dropped so much for all causes too, so you need another huge 3rd? wave to pull this off?


Maybe I'm reading it differently, or just being optimistic, I'm reading this as lowering over time, meaning that some of the deaths from Covid might very well have been from the expected numbers any given year to begin with. It seemed to be way above the line on April 11th, but now, getting closer and closer to normal. I also don't think you can conclude that this means there is nearly certainly some undercounting even if we are in the excess. Any given year can change by 100k or so in the US, and......

Wouldn't you also have to factor in more potential deaths due to suicide, drug overdose, alcohol, violent crimes, other health issues neglected or cancelled due to Covid, etc. many other reasons people may have died that weren't because they contracted covid, but rather because of a covid world associated with lockdowns and the fear driven hysteria created by the media.

On the other hand, you may have less potential accidents or flu / pneumonia deaths from distancing and less people on the roads.


All this is broken out by the CDC. I would take a look at their statistics for yourself. There's no need to bring in things like car deaths and suicides, it's easy enough to split things out with that aside.

They are tracking pneumonia like deaths by state, then you can subtract out the actual Covid deaths to see where there might be an undercount. This is accelerating a bit. That's the easiest. The ones that may be concerning are dementia which are drastically up over prior year, and all cause cardiacs. Again drastically up. Those are very suspicious, but not so much as to freak out over. I do think that a rational person would/could ascribe the excess pneumonia deaths to Cov this year safely.

Even if you do choose to ascribe 100% of other semi-related cause (cardiac, dementia) excess mortality to Covid, it still doesn't look like it will get anywhere close to this figure and I don't see that they are trying to do this. (apply EM data)



beerad12man
How long do you want to ignore this user?
AG
I'm not expert, but it looks like the gap has closed dramatically since April 11th, but I guess I'm just reading it wrong?

I'm pretty sure excess pneumonia deaths have happened a lot if you look back over the years. It tends to fluctuate, or at least has had a few years with big spikes. I seem to remember a post about that on this board, but I may be wrong. I don't think it's any guarantee that it's necessarily covid increasing those.
culdeus
How long do you want to ignore this user?
AG
beerad12man said:

I'm not expert, but it looks like the gap has closed dramatically since April 11th, but I guess I'm just reading it wrong?

I'm pretty sure excess pneumonia deaths have happened a lot if you look back over the years. It tends to fluctuate, or at least has had a few years with big spikes. I seem to remember a post about that on this board, but I may be wrong. I don't think it's any guarantee that it's necessarily covid increasing those.
Yes, and I pointed to the gaps in some of that logic. That being said you still have to reconcile alot of deaths over what will be the 9 year high in pneumonia deaths even taking 100% of all Cov deaths as pneumonia induced. The year isn't really over yet. Back of envelope there seems to be at least 2500 deaths (probably 4000) that are coming from somewhere that wasn't expected over the prior 9 year highs in Texas. That's still quite a bit, but again nowhere near what this model would be tracking towards (if all were really cov).

The last refresh of this seems to be 8/4. I'll take a look at it Wednesday when they in theory will have refreshed it to get the most realistic look and put real numbers to it.
Aust Ag
How long do you want to ignore this user?
AG
Aggies2009 said:

Old RV Ag said:

DeBoss said:

Can you help me understand how masks are working if deaths are spiking?
You're assuming everyone is wearing a mask. Kind of like a condom - having it stay in your wallet doesn't do any good.
Considering stores don't let people in without them, they're being enforced pretty hard.

Yet cases didn't drop.
Here's my take. I think little transmission would be happening at stores with or w/o masks. Think about it: How close to you get to strangers to talk? Especially these days....it's sort of ingrained at this point, stand away. With those shields at every damn checkout, you really wouldn't even transfer to the clerks. Seems like sort of a waste.

I think masks at grocery stores, etc are pretty useless. Most of the time, I'm the only person in the aisle. I feel like "guy driving with mask".

I think what we're seeing is the biggest issue is still in families, with friends (parties, get togethers, etc) and well, bars when they open (basically, a form of a party). Not sure how you stop that, but those that get it under those circumstances sort of got whatever comes....they know better. Everyone knows better at this point, whether they'll admit or not.

Another Doug
How long do you want to ignore this user?
AG
Stupid local news story.

The model is by a statistics PHD candidate that at this point only has a BS in statistics.

His model is largely based on cell phone usage.


The max of any model I have seen is 14,000
fullback44
How long do you want to ignore this user?
AG
This thread needs to be brought up at the end of August ..

Fear. Mongering x 10,000
bigtruckguy3500
How long do you want to ignore this user?
Did y'all read the article?

This model is an automated model that is constantly changing projections based on data it pulls. On July 27th there was a reporting error in its source data that listed 1202 deaths, compared with 88 the day before and 223 the day after.

That erroneous spike caused the model to have such a high projection.

The author of the article is just using the number, which the creators of the model acknowledge is faulty, as click bait.

"At the beginning of August, the model actually had a more modest projected death toll of 7,176, compared to the real-world total of 7,471."

Any model/projection is only as good as the data that goes into it, and the data is constantly changing.
aglaes
How long do you want to ignore this user?
AG
beerad12man said:

For 0.25% or 0.65%, not as many fall in line thinking it's worth it to society as a whole to do what we have done to it. Even if others think we are monsters for doing so. It's how we are wired.


Especially if you're under 40 and relatively healthy and IFR is really 0.001%
AggieBiker
How long do you want to ignore this user?
AG
bigtruckguy3500 said:

Did y'all read the article?

This model is an automated model that is constantly changing projections based on data it pulls. On July 27th there was a reporting error in its source data that listed 1202 deaths, compared with 88 the day before and 223 the day after.

That erroneous spike caused the model to have such a high projection.

The author of the article is just using the number, which the creators of the model acknowledge is faulty, as click bait.

"At the beginning of August, the model actually had a more modest projected death toll of 7,176, compared to the real-world total of 7,471."

Any model/projection is only as good as the data that goes into it, and the data is constantly changing.

And yet it was reported on the news here in Dallas last night as a credible projection. People do not study the data or question the facts on any of the information about this entire event. They just accept what is said if it fits their personal view of the risk of the virus affecting them or their community. And this shows again how the media thinks selling fear is what draws in viewers. Thanks for your insight on this.
aTm papi
How long do you want to ignore this user?
Old RV Ag said:

DeBoss said:

Can you help me understand how masks are working if deaths are spiking?
You're assuming everyone is wearing a mask. Kind of like a condom - having it stay in your wallet doesn't do any good.
Pull out worked for me back in college.
tysker
How long do you want to ignore this user?
AG
Old RV Ag said:

Aggies2009 said:

Old RV Ag said:

DeBoss said:

Can you help me understand how masks are working if deaths are spiking?
You're assuming everyone is wearing a mask. Kind of like a condom - having it stay in your wallet doesn't do any good.
Considering stores don't let people in without them, they're being enforced pretty hard.

Yet cases didn't drop.
But going to a party or gathering after without one. Kind of like wearing a condom with one girl and not wearing one with another and wondering why girl two is pregnant (or you've got an STD) - but I wore a condom!
There is no "safe" sex just like there is no covid risk-free "safe" interaction. Condoms are 99% effective only when used correctly. I've read that in realworld applications condoms are about 85-90% effective. Just like condoms, masks are a tool not a solution. The only real solution far stopping transmission is abstinence from in person interactions.
BiochemAg97
How long do you want to ignore this user?
AG
From the statement in the article by the spokesperson...

" The mortality data do not yet reflect the slowdown in transmission that occurred following mid-June changes in COVID-19 policies and behavior. If mortality data begin to decline, similar to recent declines in COVID-19 case counts and hospitalizations, then the model will naturally update to reflect the trend and its projected deaths will decrease."

That suggests the model fails to consider the relationship between new cases and deaths. Makes me question if it accounts for any epidemiological factors.

The model considers cell phone mobility data and current trend of death. Cell phone mobility is a reasonable proxy for how well people are staying home and isolating, but it fails to account for things like social distancing and masks.

It isn't surprising that cell phone mobility might be increasing as teachers return to schools to prep for the new school year. However, with so many school districts starting with online only and generally having the teachers teach from the classrooms while the students stay home, going to school to hang out in a large classroom by yourself doesn't really increase the chance of spread.
BiochemAg97
How long do you want to ignore this user?
AG
tysker said:

Old RV Ag said:

Aggies2009 said:

Old RV Ag said:

DeBoss said:

Can you help me understand how masks are working if deaths are spiking?
You're assuming everyone is wearing a mask. Kind of like a condom - having it stay in your wallet doesn't do any good.
Considering stores don't let people in without them, they're being enforced pretty hard.

Yet cases didn't drop.
But going to a party or gathering after without one. Kind of like wearing a condom with one girl and not wearing one with another and wondering why girl two is pregnant (or you've got an STD) - but I wore a condom!
There is no "safe" sex just like there is no covid risk-free "safe" interaction. Condoms are 99% effective only when used correctly. I've read that in realworld applications condoms are about 85-90% effective. Just like condoms, masks are a tool not a solution. The only real solution far stopping transmission is abstinence from in person interactions.
Abstinence only sex Ed leads to higher teen pregnancy rate. We as a society are very bad at abstinence.
PJYoung
How long do you want to ignore this user?
AG
I saw this stat this morning. It took the world 6 months to get to 10 million confirmed cases and only 6 weeks to get to 20 million.
PJYoung
How long do you want to ignore this user?
AG
PJYoung said:

I saw this stat this morning. It took the world 6 months to get to 10 million confirmed cases and only 6 weeks to get to 20 million.

This space reserved for post getting blue stars stating "Good, the quicker we get to herd immunity the better"
Last Page
Page 1 of 3
 
×
subscribe Verify your student status
See Subscription Benefits
Trial only available to users who have never subscribed or participated in a previous trial.