Started By
Message

re: IHME model is getting better for the US...stabilizing as we get more and more good data

Posted on 4/8/20 at 9:30 am to
Posted by brgfather129
Los Angeles, CA
Member since Jul 2009
17112 posts
Posted on 4/8/20 at 9:30 am to
quote:

You mean they are correcting it for the ridiculous over projections used to take the economy down to get Trump


Posted by ell_13
Member since Apr 2013
85215 posts
Posted on 4/8/20 at 9:30 am to
quote:

while this once is barely into its second month of spread.
Yikes.

How many times do we have to show data that proves that this thing was widespread in February? Even now the state is trying to point out that the “confirmed” cases coming through now were from tests and patients first seeing symptoms in mid-March. And those are just the CONFIRMED cases. Which is a small fraction of actual cases out there.
Posted by Salmon
On the trails
Member since Feb 2008
83653 posts
Posted on 4/8/20 at 9:30 am to
quote:

Models are so wrong (and therefore non-useful) that meteorologists are embarrassed for you.


Bad answers are not useful, I agree.

But one should always update with more information. And when one does so, we shouldn't hold their previous wrong answer against them simply because of lack of information.

This goes with individuals as well. People that refuse to update their opinions when better information is presented at the worst.
Posted by JohnnyKilroy
Cajun Navy Vice Admiral
Member since Oct 2012
35588 posts
Posted on 4/8/20 at 9:30 am to
Tigers247 predicted 50 total US deaths
Posted by uscpuke
Member since Jan 2004
5040 posts
Posted on 4/8/20 at 9:31 am to
quote:

Again, no dog in this fight. I don't really want to argue anymore, just sharing the data as we get it.

Yet you got your 100 screenshots ready to go...
Posted by Klark Kent
Houston via BR
Member since Jan 2008
67051 posts
Posted on 4/8/20 at 9:31 am to
quote:

Lots of modeling apologists. You cannot put out these ridiculously overstated numbers and expect people to take you seriously next time


Yes. Yes you can.

And I’d argue that you have to. For officials, governments, and people to take it seriously. Like meteorologists I doubt they enjoy scaring people, but as Salmon has said in this thread they can only model based on data given. It’s not pandemic modelers fault if they are given shite data. blame the media and the internet for the fear mongering aspect.
Posted by RB10
Member since Nov 2010
44246 posts
Posted on 4/8/20 at 9:31 am to
quote:

Apples to oranges.

Flu numbers aren't really models. They just take every pulmonary related death during a certain time of year and call it flu.


Both are estimations based on limited data. Seems similar enough to me.
Posted by TheCaterpillar
Member since Jan 2004
76774 posts
Posted on 4/8/20 at 9:31 am to
quote:

Yet you got your 100 screenshots ready to go...



Not what I'm talking about...

I didn't want to argue the models or about their accuracy. Just sharing them.

I do want to call out morons on the OT though, that is just good ole fashioned fun.
Posted by ell_13
Member since Apr 2013
85215 posts
Posted on 4/8/20 at 9:33 am to
quote:

I do want to call out morons on the OT though, that is just good ole fashioned fun.

Posted by Salmon
On the trails
Member since Feb 2008
83653 posts
Posted on 4/8/20 at 9:36 am to
quote:

Both are estimations based on limited data. Seems similar enough to me.



Ok

and if we got better flu data and it showed that only 1/10th of flu deaths are actually flu related?

all the "its just the flu" guys may not like the answer if we got better data on flu deaths
Posted by RB10
Member since Nov 2010
44246 posts
Posted on 4/8/20 at 9:39 am to
quote:

Ok

and if we got better flu data and it showed that only 1/10th of flu deaths are actually flu related?

all the "its just the flu" guys may not like the answer if we got better data on flu deaths


"If"

All the "it's 10x more deadly" people wouldn't like the answers if they were testing everyone who was showing symptoms for Covid-19 either. They aren't though. We already know that.
This post was edited on 4/8/20 at 9:42 am
Posted by Salmon
On the trails
Member since Feb 2008
83653 posts
Posted on 4/8/20 at 9:40 am to
It's almost certainty that flu deaths are over estimated based off limited information

its also funny how the flu range is 12k - 60k and y'all always seem to use the high range when making these arguments
This post was edited on 4/8/20 at 9:43 am
Posted by ell_13
Member since Apr 2013
85215 posts
Posted on 4/8/20 at 9:42 am to
You could also make the claim that the estimate of people who get the flu is also overestimated. Since the majority used in the calculation aren’t confirmed cases.
Posted by RB10
Member since Nov 2010
44246 posts
Posted on 4/8/20 at 9:42 am to
quote:

It's almost certainty that flu deaths are over estimated based off limited information


See edit above. Same logic applies to Covid-19.

This isn't the "gotcha" you want it to be.
Posted by TheCaterpillar
Member since Jan 2004
76774 posts
Posted on 4/8/20 at 9:43 am to
I'd compare flu numbers to the COVID numbers from late March. That's how accurate they likely are. Completely pulled out of their arse.

We're getting more and more real data from the mass testing for COVID, while flu is just going to be what it is. The COVID stuff isn't perfect obviously, but it is fun to watch it improve every couple of days.

Hopefully it encourages our government to open the economy back up on May 1st.
Posted by TheCaterpillar
Member since Jan 2004
76774 posts
Posted on 4/8/20 at 9:44 am to
quote:

All the "it's 10x more deadly" people wouldn't like the answers if they were testing everyone who was showing symptoms for Covid-19 either. They aren't though. We already know that.



We'll know the CFR pretty damn well soon enough with the antibody testing.

Will be interesting, that's for sure. I do wonder if hundreds of thousands of people had it and don't know. I assume so.
Posted by Salmon
On the trails
Member since Feb 2008
83653 posts
Posted on 4/8/20 at 9:44 am to
quote:

All the "it's 10x more deadly" people wouldn't like the answers if they were testing everyone who was showing symptoms for Covid-19 either. They aren't though. We already know that.


Yes. If everyone was tested for COVID the CFR would drop.

Posted by Salmon
On the trails
Member since Feb 2008
83653 posts
Posted on 4/8/20 at 9:44 am to
quote:

This isn't the "gotcha" you want it to be.


I'm not the one trying the "gotcha"
Posted by CarRamrod
Spurbury, VT
Member since Dec 2006
57491 posts
Posted on 4/8/20 at 9:46 am to
quote:

There is going to be a ton of this, even though most of those projections were probably based on lack of/bad data rather than anything else.
well any intelligent person that wants to be right would know, bad data isn't data. Therefore it wasn't just, oh shucks we had bad data.
Posted by Salmon
On the trails
Member since Feb 2008
83653 posts
Posted on 4/8/20 at 9:48 am to
quote:

well any intelligent person that wants to be right would know, bad data isn't data. Therefore it wasn't just, oh shucks we had bad data.


which is why you update as soon as you get better data

first pageprev pagePage 4 of 9Next pagelast page

Back to top
logoFollow TigerDroppings for LSU Football News
Follow us on Twitter, Facebook and Instagram to get the latest updates on LSU Football and Recruiting.

FacebookTwitterInstagram