COVID-19 2nd wave - Texas only - Stats and such

[Wulaw Horn 19560]

2 minutes ago, BradInATX said:

19 minutes ago, Wulaw Horn said:

0.025% of the population so far. Serious question- anyone know how many total deaths (all causes) since March and how many we’d normally expect?  

We've been through the numbers on a couple of threads. There are between 225k and 350k surplus deaths vs an average year so far according to the CDC.

Sorry- I meant specifically texas. I’m just curious to see how many people would normally be expected to die. 
I do the calculations every time I drive by a road sign ( drive safe there have been X number of deaths in the year xxxx). Always try to Figure out- so- I have a 1/15000 chance of dying on the road) etc. 

I calculated my odds for dying from Covid at about 1/4000. Thats certainly not nothing. Of course I have no idea what my actual odds for getting infected or odds of dying if infected are, that’s back of envelope math. 

[smoky 842]

In 2018, approximately 17k Texans died in a given month, according to the CDC.

 

As of a couple of weeks ago, Texas was ahead of the total excess death count from March through July by several thousand.

 

 

 

 

[BradInATX 9208]

Sorry- I meant specifically texas. I’m just curious to see how many people would normally be expected to die. 
I do the calculations every time I drive by a road sign ( drive safe there have been X number of deaths in the year xxxx). Always try to Figure out- so- I have a 1/15000 chance of dying on the road) etc. 
I calculated my odds for dying from Covid at about 1/4000. Thats certainly not nothing. Of course I have no idea what my actual odds for getting infected or odds of dying if infected are, that’s back of envelope math. 

Dunno about Texas specifically. Not sure if they have those numbers.

[justhookit 4205]

We are up about 18% in excess deaths. That number has been climbing quickly this summer.

[achooloco 1086]

On 8/7/2020 at 6:36 PM, Texas Jeff said:

 Is it possible that this was caused by July 4th get-togethers? If so, that sucks, because it means that an event that causes a spike does not show up in the data for a month.  With that kind of lag, any type of opening plan has to go really really slowly.

welcome to every new yorker in march, which is now 5 months behind reality. feel free to catch up. i don't mean that in a pedantic way, it is just the reality of the world we live in now. IF WE HAD quick accurate tests, we could change things but having a fast result means we can get on with life. without it, means draconian shutdowns and collapsed economies in our home state. we're texas, fuck everyone else. lets get the tests and do this right.

[Pig Bellmont 8867]

20 hours ago, Cody2422 said:

Cumulative graphs are scary.  You should lock yourself in your house

1500 Texans died last week. But cool that you’re so tough. I’m sure there’s nothing to worry about 

[Dbeasy 15526]

It’s always so encouraging about the intelligence of our fellow citizens when every few weeks someone wades into the covid threads on Surly and pronounces the virus as no big deal by dividing the number of deaths by the total population - completely ignoring mountains of global scientific evidence that it is an extremely dangerous, highly contagious, serious health risk, AND ignoring the fact that the world “only” has the number of deaths it has as a result of unprecedented actions by governments and citizens around the world. No big deal guys!

[kevwun 16269]

A friend's wife had it and she said the worst part of her symptoms were the headaches.  She said she's never had a migraine, but she guesses that's what they were like.  Pretty much debilitating when they were at their worst.  She lost her sense of taste, but not smell.  Same for their son who had it.

Edited August 9, 2020 by kevwun

[BrazilHorn 6792]

Big Milestone this week. Travis County had its 1st recorded death in the 10-19 year old age bracket. Still a very low mortality rate (0.06%) but a 1st nonetheless.

Only 22 deaths WoW down (prior 3 week avg was 36/week so this is a good reduction)

Mortality Rate by Age bucket below (defined as # died vs total # of cases within age bucket)

0 - 1 (0%)

1 - 9 (0%)

10-19 (0.06%)

20 - 29 (0.03%)

30-39 (0.12%)

40 - 49 (0.30%)

50 - 59 (1.36%)

60 - 69 (4.34%)

70 - 79 (8.76%)

80+ (19.93%) - This is crazy - 1 in 5 chance that if you are >80 and get CV19 you will die.

The three oldest age cohorts make up 81% of all deaths. Not saying anything about lingering conditions in younger, recovered folks but the group dying are those that are older. Stay away from your parents/grandparents.

[Anastasis 23699]

43 minutes ago, kevwun said:

A friend's wife had it and she said the worst part of her symptoms were the headaches.  She said she's never had a migraine, but she guesses that's what they were like.  Pretty much debilitating when they were at their worst.  She lost her sense of taste, but not smell.  Same for their son who had it.

What other symptoms did she have other than ha and lost of taste?

[Liquor and Poker 3659]

49 minutes ago, BrazilHorn said:

Big Milestone this week. Travis County had its 1st recorded death in the 10-19 year old age bracket. Still a very low mortality rate (0.06%) but a 1st nonetheless.

Only 22 deaths WoW down (prior 3 week avg was 36/week so this is a good reduction)

Mortality Rate by Age bucket below (defined as # died vs total # of cases within age bucket)

0 - 1 (0%)

1 - 9 (0%)

10-19 (0.06%)

20 - 29 (0.03%)

30-39 (0.12%)

40 - 49 (0.30%)

50 - 59 (1.36%)

60 - 69 (4.34%)

70 - 79 (8.76%)

80+ (19.93%) - This is crazy - 1 in 5 chance that if you are >80 and get CV19 you will die.

The three oldest age cohorts make up 81% of all deaths. Not saying anything about lingering conditions in younger, recovered folks but the group dying are those that are older. Stay away from your parents/grandparents.

I keep hearing anecdotal evidence about major complications or long term malaria-style debilitations in survivors in various age cohorts, but I haven’t seen any data that isn’t binary “dead” or “recovered.”  

I think it would be very useful in calming fears if someone would break down the percentages by age of “didn’t die, but still fucked up pretty bad months later” or “didn’t die but for 10 days I wanted to.”

I get that it’s a bigger ask but I think it might be very helpful to hear what I can reasonably expect other than “you will probably not die.”

[BrazilHorn 6792]

1 hour ago, Liquor and Poker said:

I keep hearing anecdotal evidence about major complications or long term malaria-style debilitations in survivors in various age cohorts, but I haven’t seen any data that isn’t binary “dead” or “recovered.”  

I think it would be very useful in calming fears if someone would break down the percentages by age of “didn’t die, but still fucked up pretty bad months later” or “didn’t die but for 10 days I wanted to.”

I get that it’s a bigger ask but I think it might be very helpful to hear what I can reasonably expect other than “you will probably not die.”

Challenge is Drs are finding new after effects weekly. Heart issues, lung scarring, cognitive/brain issues, blood clotting etc

There are going to be years of longitudinal studies around recovered patients spanning a lot of possible dimensions. 
 

Just crazy. 

[deter 314]

6 hours ago, Pig Bellmont said:

1500 Texans died last week. But cool that you’re so tough. I’m sure there’s nothing to worry about 

From Covid or total???

[bolverk 30231]

24 minutes ago, deter said:

From Covid or total???

Texas reports over 1,300 coronavirus deaths in the last five days

[Message Board User 12405]

 

[Message Board User 12405]

 

[justhookit 4205]

I’m on the beach so haven’t read anything on here but I guess Corpus just did a data dump.Over 1100 new cases. Hospital and ICU down

[justhookit 4205]

Ok it’s a data dump 

https://www.kiiitv.com/mobile/article/news/county-by-county-breakdown-how-many-cases-of-covid-19-are-there-in-the-coastal-bend/503-d4641cac-7a0c-4561-9e23-7c52dacba91e

 

[B00M 6725]

16 hours ago, bolverk said:

Texas reports over 1,300 coronavirus deaths in the last five days

AUSTIN (KXAN) — With active cases of COVID-19 in slight decline around Texas, a projection by modelers at the University of Texas at Austin raises new concerns. It suggests there will be 15,000 more deaths in Texas over the next three weeks.

The model by UT’s College of Natural Sciences predicts that more than 23,000 Texans will die as a result of COVID-19 by Aug. 31. Currently 8,459 Texans have died since the pandemic began. The predicted escalation would represent a 177% increase by Aug. 31.

 

 

[bolverk 30231]

My gawd, I hope they're wrong. Ugh.

[BrazilHorn 6792]

Really positive momentum on total cases with the past two days being in the low 100s (122). This is first time with back to back <200 days since 7/20. Having no cases in those >80 is good news in regards to forward looking mortality as this bucket has ~20% mortality rate.

Age Bracket	<1	1 to 9	10 to 19	20 to 29	30 to 39	40 to 49	50 to 59	60 to 69	70 to 79	>80	Total
New Cases 8/9 vs 8/8	1	2	23	30	18	20	13	11	4	0	122
% of Daily Change	0.82%	1.64%	18.85%	24.59%	14.75%	16.39%	10.66%	9.02%	3.28%	0.00%
% of Total Cases	0.51%	2.80%	8.01%	26.79%	21.53%	16.36%	11.69%	6.53%	3.28%	2.50%	22,724

[ChickenSandwich 9918]

12 minutes ago, B00M said:

AUSTIN (KXAN) — With active cases of COVID-19 in slight decline around Texas, a projection by modelers at the University of Texas at Austin raises new concerns. It suggests there will be 15,000 more deaths in Texas over the next three weeks.

The model by UT’s College of Natural Sciences predicts that more than 23,000 Texans will die as a result of COVID-19 by Aug. 31. Currently 8,459 Texans have died since the pandemic began. The predicted escalation would represent a 177% increase by Aug. 31.

 

 

Curious as to UT’s prediction record so far during Covid. 

[Pimphand 6266]

6 minutes ago, ChickenSandwich said:

Curious as to UT’s prediction record so far during Covid. 

So these fucksticks think we're going to lose 1.6 times as many people over the next 3 weeks as we have lost over the preceding 5 months?

[B00M 6725]

9 minutes ago, ChickenSandwich said:

Curious as to UT’s prediction record so far during Covid. 

 

Not sure but this is pretty cool

https://covid-19.tacc.utexas.edu/austin-projections/

 

[Captainant 28387]

11 minutes ago, ChickenSandwich said:

Curious as to UT’s prediction record so far during Covid. 

Here's some other data from the site that you were too lazy to click through

Or you could go look at the raw data yourself comparing their estimates to actuals in their github repo, along with their forecasting algo

But it's also worth noting that your predictions so far during COVID haven't been so great yourself lol. Summer has come and gone, but COVID hasn't 🤔

[Guest Lobo]

Well if the heat is gonna kill this thing, then Thursday is the big day to kiss Covid-19 goodbye.  106 with no clouds, 115 heat index.  What the fuck?!? 

 

[victory88 5755]

A family friend owns a restaurant in southeast Texas.  His wife and himself are in their mid to late 30s.  He had fever and felt under the weather.  He ended up testing positive. His wife also got it and so did a few of their siblings just from being around him.  Thank goodness their parents did not test positive as they see them weekly.  He was pretty sick for a week.  His lungs have taken beating.  He has recovered now but his breathing is still not back to normal.  His brother now has pneumonia but has tested negative.  

Had another moron friend in Minneapolis go to a bar/restaurant to chase tail and he got sick and tested positive as well.  Fucking moron.  

Also spoke to a pediatrician that works at the ICU at children's hospital in the Covid unit.  Said their beds are full with kids.  She's said opening schools is going to be fucking terrible.  

[ChickenSandwich 9918]

2 minutes ago, Captainant said:

Here's some other data from the site that you were too lazy to click through

Or you could go look at the raw data yourself comparing their estimates to actuals in their github repo, along with their forecasting algo

But it's also worth noting that your predictions so far during COVID haven't been so great yourself lol. Summer has come and gone, but COVID hasn't 🤔

Link to my “predictions” lol. 

[B00M 6725]

7 minutes ago, Pimphand said:

So these fucksticks think we're going to lose 1.6 times as many people over the next 3 weeks as we have lost over the preceding 5 months?

I hope you're right. I think this is the model being used to come up with that 15k TX number: 

https://covid-19.tacc.utexas.edu/projections/

[MNLonghornFUKM 8963]

AUSTIN (KXAN) — With active cases of COVID-19 in slight decline around Texas, a projection by modelers at the University of Texas at Austin raises new concerns. It suggests there will be 15,000 more deaths in Texas over the next three weeks.

The model by UT’s College of Natural Sciences predicts that more than 23,000 Texans will die as a result of COVID-19 by Aug. 31. Currently 8,459 Texans have died since the pandemic began. The predicted escalation would represent a 177% increase by Aug. 31.

 

 

Wow. I don’t believe it for a second but wow at making that public.

[Hefeweizen 5416]

9 minutes ago, Lobo said:

Well if the heat is gonna kill this thing, then Thursday is the big day to kiss Covid-19 goodbye.  106 with no clouds, 115 heat index.  What the fuck?!? 

 

lol the heat theory turned out great

[BrazilHorn 6792]

6 minutes ago, B00M said:

I hope you're right. I think this is the model being used to come up with that 15k TX number: 

https://covid-19.tacc.utexas.edu/projections/

Are they looking at the max end of range? Every single metro on their chart shows a decline from their 7 day moving average (except for El Paso). To be fair I didn't deep dive into it, but just looking at the projections it would seem deaths declining across state vs having this type of significant uptick.

[deter 314]

17 hours ago, bolverk said:

Texas reports over 1,300 coronavirus deaths in the last five days

From DSHS spreadsheet, seems to be a positive trend

3/6- 3/13             0

3/13-3/20         10

3/20-3/27         37

3/27-4/3          116

4/3-4/10          187

4/10-4/17        237

4/17-4/24       221

4/24-5/1         223

5/1-5/8           229

5/8-5/15         231

5/15-5/22      179

5/22-5/29      186

5/29-6/5        171

6/5-6/12        198

6/12-6/19      272

6/19-6/26     353

6/26-7/3       673

7/3-7/10      1055

7/10-7/17     1425

7/17-7/24     1317

7/24-7/31     828

7/31-8/7      195

[Pimphand 6266]

44 minutes ago, B00M said:

I hope you're right. I think this is the model being used to come up with that 15k TX number: 

https://covid-19.tacc.utexas.edu/projections/

Here is the screenshot capture so you can come back and revisit

Max projection is 1,691 LMFAO

[bolverk 30231]

13 minutes ago, deter said:

From DSHS spreadsheet, seems to be a positive trend

3/6- 3/13             0

3/13-3/20         10

3/20-3/27         37

3/27-4/3          116

4/3-4/10          187

4/10-4/17        237

4/17-4/24       221

4/24-5/1         223

5/1-5/8           229

5/8-5/15         231

5/15-5/22      179

5/22-5/29      186

5/29-6/5        171

6/5-6/12        198

6/12-6/19      272

6/19-6/26     353

6/26-7/3       673

7/3-7/10      1055

7/10-7/17     1425

7/17-7/24     1317

7/24-7/31     828

7/31-8/7      195

If you go to the DSHS Dashboard, they clearly indicate that the last several days are incomplete data. They even have it in a different color to point that out and a note that reads:

"Because DSHS is reporting fatalities by date of death, the number of fatalities for recent days will continue to grow as more death certificates are filed for those days."

[Anastasis 23699]

Put a gun to my head and I'm gonna take the under on those projections, but they should sober us all up a bit. 

[bolverk 30231]

46 minutes ago, BrazilHorn said:

Are they looking at the max end of range? Every single metro on their chart shows a decline from their 7 day moving average (except for El Paso). To be fair I didn't deep dive into it, but just looking at the projections it would seem deaths declining across state vs having this type of significant uptick.

Based on the data they're displaying, their projections must be pulling deaths from the smaller metros and rural areas. Yes, that's a "duh" statement I just made.

1 minute ago, Anastasis said:

Put a gun to my head and I'm gonna take the under on those projections, but they should sober us all up a bit. 

I agree on all three of your points.

[Anastasis 23699]

3 minutes ago, bolverk said:

I agree on all three of your points.

reported. 

[Captainant 28387]

24 minutes ago, Pimphand said:

Here is the screenshot capture so you can come back and revisit

Max projection is 1,691 LMFAO

That's max projected deaths in a day. You'll want to add up each of those daily projections to get to the 15k over the next three weeks number.

I looked through their older predictions and while it seems like they're off by +-20% they're on the trendline pretty good and within their error bars, and have even correctly predicted days with aberrant lower case count. Those are some WIDE error bands on the predictions, but they acknowledge it in their abstract and give some good advice on how to interpret their analysis:

Quote

I've noticed that some of the error bars on your forecasts are pretty wide. Why is that?

Mostly because future death rates really are uncertain in a lot of areas. The error bars from our model are going to be wide in any state or city where the available data on deaths, together with the observed timing and extent of social distancing patterns, cannot rule out future growth in the death rate.

There are a couple of other points to consider here. First, our model does err a bit on the side of conservatism: on historical back-tests, our model's 95% error bars tend to cover a little more than 95% of the notionally "future" data points.

Second, there are a lot of things our model doesn't "know" about. In particular, it doesn't use recent data on cases (because the data are difficult to interpret in light of variation in testing availability) or hospitalizations (because the data aren't widely available). So we'd encourage you to think about our projections as representing a range of plausible scenarios in the absence of any knowledge of recent case/hospitalization trends in your area. If you know, for example, that hospitalizations in your area have flattened off, then you're likely to see death rates more on the lower end of our error bars. If hospitalizations are still growing, you're more likely to see death rates on the upper end of (or even beyond) our error bars.

 

Edited August 10, 2020 by Captainant

[B00M 6725]

34 minutes ago, Pimphand said:

Here is the screenshot capture so you can come back and revisit

Max projection is 1,691 LMFAO

Captainant beat me to it. Sum up those daily projections and you probably get around 15k. Nothing here should be particularly shocking. We expect deaths to lag hospitalizations and we're on the back side of the hospitalization wave, so here's the death wave...

[UT_OB1 2260]

41 minutes ago, Pimphand said:

Here is the screenshot capture so you can come back and revisit

Max projection is 1,691 LMFAO

Is the green line 7 day average forward looking?...  I don't understand the step change late July.  After the step change, the average is higher than any day preceding it.  Actually it appears to be higher than any of the handful of actual days after it as well.  Are there multiple axis that just aren't showing up?

[DefinitelyNotHollywoodColt 12301]

1 hour ago, ChickenSandwich said:

Curious as to UT’s prediction record so far during Covid. 

would you describe yourself as bored with their predictions?

[BehoId, The Underminer! 11669]

On 8/8/2020 at 10:08 PM, Wulaw Horn said:

0.025% of the population so far. Serious question- anyone know how many total deaths (all causes) since March and how many we’d normally expect?  

 

On 8/8/2020 at 10:30 PM, Wulaw Horn said:

Sorry- I meant specifically texas. I’m just curious to see how many people would normally be expected to die. 
I do the calculations every time I drive by a road sign ( drive safe there have been X number of deaths in the year xxxx). Always try to Figure out- so- I have a 1/15000 chance of dying on the road) etc. 

I calculated my odds for dying from Covid at about 1/4000. Thats certainly not nothing. Of course I have no idea what my actual odds for getting infected or odds of dying if infected are, that’s back of envelope math. 

 

Texas deaths.  Blue is total observed deaths. Yellow is the 95% confidence interval for maximum deaths in that week.

[Burt Macklin 19211]

43 minutes ago, DefinitelyNotHollywoodColt said:

would you describe yourself as bored with their predictions?

Only if they prove to be accurate. 

[Pimphand 6266]

1 hour ago, B00M said:

Captainant beat me to it. Sum up those daily projections and you probably get around 15k. Nothing here should be particularly shocking. We expect deaths to lag hospitalizations and we're on the back side of the hospitalization wave, so here's the death wave...

Except that hospitalizations are already on the decline...

[DefinitelyNotHollywoodColt 12301]

2 minutes ago, Pimphand said:

Except that hospitalizations are already on the decline...

Do we really need to explain this again? Fuck.

[Armybrat 41319]

20 minutes ago, DefinitelyNotHollywoodColt said:

Do we really need to explain this again? Fuck.

What’s up with that gap?

[Texas Jeff 1998]

Man, who knows at this point...  Few would have predicted the sharp rise in June after a relatively flat May and I think few at the end of June were thinking we would have a peak in mid-July.  Personal behavior is driving this thing.

I predict we will have 100-200 deaths per day in Texas throughout August, which means 3000-6000 Texas deaths in August.  Probably more Texans dead this month from COVID than died in then Vietnam War.

The real ballgame is what happens from mid-August through early October.  How does behavior change when college kids are back on campus, some kids are back in school, and some colleges are trying to host tens of thousands of people at football games?   I think we have learned to gradually beat back the virus the way we are living this summer, but life is about to change because the calendar says it is time to change.  Our reaction to that will drive the numbers in October and November, which could either dwarf August numbers or decay to not much, based on behavior.

Edited August 10, 2020 by Texas Jeff

[Hefeweizen 5416]

1 hour ago, DefinitelyNotHollywoodColt said:

would you describe yourself as bored with their predictions?

Yawn.  Kung flu stats are boring unless they are near zero.

[Pig Bellmont 8867]

1 hour ago, B00M said:

we're on the back side of the hospitalization wave,

Well that’d be fine if schools weren’t starting in person instruction and with Labor Day around the corner.  I’m going to wager the hospitalization curve across TX won’t be a steady decline