Wuhan Virus Model Sees Another Dramatic Downward Revision

[joepamjohn]

One of the big unknown in these models is not the virus, but how people behave. They had to guess about social distancing and what the compliance to the recommendations was going to be. Get it a little wrong and the results vary wildly.

There was zero way to measure this. They had to guess. They guessed wrong.

Do you really believe any news coming out of communist China? Gees.

[DrjonesUSA]

The models projected final totals based on a several different scenarios, the results of which have a wild degree of speculation driving them. Some were worst case models, which never actually happens since people do react to major epidemics, but such worst case scenarios are useful when discussing mitigation measures with people who believe doing nothing is actually the ideal response.

These worst case scenario numbers have been touted by people who don't understand the necessary limitations and limited value of those model forecasts to demonstrate how the experts must be wrong. Largely this seems to have been done out of partisan political argumentation but some have done it out of a general distrust for medical science. In both cases ignorance has become the basis for a greater 'understanding', this is always a house of cards and it will always collapse in due time.

So don't let those people set their hooks into your reality. They'll distort it entirely out of whack and you'll end up with a perspective so distant from objective you won't have any frame of reference with which to recalibrate.

[Transient]

The original model was 2.2m deaths in the USA.



Later downgraded to 100,000 to 240,000



Now downgraded AGAIN to between 26,000 and 155,000



AND FOR THIS WE CREATE A DEPRESSION.

Sounds like annual flu numbers

Sent from my SM-N960U using Tapatalk

[sd_shooter]

WuFlu virus model updated again

https://www.breitbart.com/health/2020/04/13/another-ihme-model-failure-hospitalization-projections-drop-by-34/

It now predicts TX deaths virtually identical with the seasonal Flu

(But to be safe - liberals please stay home!)

[HecklerNKoch]

I appreciate you putting this out there. Especially the helpful analogies.

I'm with you on this. The math has been overly amplified from the beginning. Will the headlines ever admit it in the end? Doubtful. The fear driven by the media has soaked in most people without a scientific background.

[sd_shooter]

I appreciate you putting this out there. Especially the helpful analogies.

I'm with you on this. The math has been overly amplified from the beginning. Will the headlines ever admit it in the end? Doubtful. The fear driven by the media has soaked in most people without a scientific background.

It's soaked into those with a scientific background as well. If the data comes from an "expert" then it must be correct

[Notpc]

Kinda reminds me of the conversation in the movie "The Shipping News".

Billy : It's finding the center of your story, the beating heart of it, that's what makes a reporter. You have to start by making up some headlines. You know: short, punchy, dramatic headlines. Now, have a look, what do you see?

[Points at dark clouds at the horizon]

Billy : Tell me the headline.

Quoyle : Horizon Fills With Dark Clouds?

Billy : Imminent Storm Threatens Village.

Quoyle : But what if no storm comes?

Billy : Village Spared From Deadly Storm.

[capo]

As you should. As all people should. A model is nothing more than a projection which tries to forecast what will happen along given trajectories given the facts known at the time of the model. The inherent problem here is that nobody knew all the facts at the time the models were created, hell they still don't know what they all are. As such every model I've seen failed to predict accurately a fundamental aspect of this: how many people would get the disease in the first place. When you can't even estimate that everything else becomes impposible too. A huge problem with this is the exponential nature of growth over time, even the tiniest of mistakes in initial assumptions will lead to gigantic differences in final results, like orders of magnitude wrong. One single week of 25% daily growth rate incorrectly accounted for will create a nearly 500% margin of error.

The problem is the general public doesn't treat models with the same degree of inherent uncertainty. A model is useful for understanding what 'may' 'could' 'might' happen if current conditions and assumptions prove out, this is useful for planning mitigation strategies but not useful for an accurate picture of where we will actually find ourselves in 3 months, 6 months or a year from now. This is especially true when your planning specifically seeks to avoid what most of the models seek to predict.

The experts often do themselves no favors by not pounding this home and of course the media seizes on dire numbers because that gets the audience's attention, 'if it bleeds it leads' is true whenever news becomes a ratings enterprise.

Now there were some experts who specifically said not to put too much faith in the models and they're mainly useful for helping plan responses but of course that has largely been drowned out.

This is where we part ways. Getting something wrong because of so many unknowns doesn't mean there's an agenda behind it, it just means people and their models have an inherent degree of uncertainty that when dealing with something that grows and decays exponentially means expected margins of error become gigantic.

Take for instance people like heads of the CDC, NIH or NIAID, they are physicians first and often experts in infectious disease and clinical research. They have easily forgotten more about pathogen spread and how to prevent the diseases they cause than anyone here will likely ever know, yet some people immediately try to politicize their medical expertise and concerns once it becomes expedient for trying to reframe the discussion. That doesn't mean that's right or proper, in fact I don't even think it makes sense on the most basic level. Why did any of these people stop being doctors and medical researchers when SARS-CoV2 showed up? All over the world their research and physician counterparts are trying to prevent serious disease and loss of life, they've dedicated their lives to this pursuit and they are almost all in agreement on how to best prevent both as this new virus comes calling in their communities. Yet somehow people leap to politics first and manage to discount all of that... doesn't that say more about them than the actual medical situation?

You can't stop people from jumping to unsupportable conclusions but you can reevaluate the situation for yourself.

[Transient]

It's soaked into those with a scientific background as well. If the data comes from an "expert" then it must be correct

Anybody can claim to be an expert. Just look at the anti-2A groups full of people who've never touched a firearm.

Trust me. I'm an expert.

Sent from my SM-N960U using Tapatalk

[as_rocketman]

WuFlu virus model updated again

https://www.breitbart.com/health/2020/04/13/another-ihme-model-failure-hospitalization-projections-drop-by-34/

It now predicts TX deaths virtually identical with the seasonal Flu

(But to be safe - liberals please stay home!)

Quoting this as an example of the equivocation logical fallacy.

Yes, now predicting deaths in Texas comparable to a typical flu season. Except:

1. The flu is endemic, caught from a million possible sources, whereas COVID-19 is still hosted by a small pool of carriers
2. Deaths in Texas date from 17 March through 14 April, as opposed to a full year of the flu
3. Deaths from COVID-19 have been suppressed by the most disruptive social measures since the Civil War, compared to "nothing" for the flu
4. Deaths in Texas are still on track to double by 26 April
5. Texas is well behind several other states

I know this is frustrating for everyone, but to quote Lao Tzu, don't underestimate your opponent.

[five.five-six]

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

Dramatic!

[sd_shooter]

Quoting this as an example of the equivocation logical fallacy.

Yes, now predicting deaths in Texas comparable to a typical flu season. Except:

1. The flu is endemic, caught from a million possible sources, whereas COVID-19 is still hosted by a small pool of carriers
2. Deaths in Texas date from 17 March through 14 April, as opposed to a full year of the flu
3. Deaths from COVID-19 have been suppressed by the most disruptive social measures since the Civil War, compared to "nothing" for the flu
4. Deaths in Texas are still on track to double by 26 April
5. Texas is well behind several other states

I know this is frustrating for everyone, but to quote Lao Tzu, don't underestimate your opponent.

You think so? Let me address those points:

Equivocation Logical Fallacy
- Example: "A warm beer is better than a cold beer. After all, nothing is better than a cold beer, and a warm beer is better than nothing"

I don't think so. I linked to BB article that contained a reference to the latest WuFlu estimate from the vaunted IHME model. I'm not mixing terminology. (I know some of you are frustrated, but let's not start applying labels... ) (BTW as_rocketman I do sincerely appreciate your input! You're level-headed)

1. Deaths comparable to typical flu season. This is merely comparing numbers, saying nothing about the origin or infectivity of each disease.

2. Deaths in TX from the flu are about 2500/season. So we're comparing a typical TX flu to the IHME forecast for Covid. Again, just numbers. Similar timeframe.

3. Yes, deaths for covid may have been suppressed. Or maybe not:
- TX may have already experienced the worst of covid during Nov/Dec/Jan.
- Since proper testing isn't available even the reported covid deaths may not be properly documented or diagnosed. One could also also say covid deaths have been inflated

4. Yes, they may double by 26 April. That will still fit within the IHME model we're talking about and also within the typical Flu season. Once we hit May it's going to be Summer around here and all bugs die (except grasshoppers, cicadas, scorpions etc. )

5. Yes, see [3]. CA is also behind other states. The IHME numbers from the article just happened to call out TX specifically. It doesn't matter how we're doing compared to the other states, just talking about the number from the model.

[as_rocketman]

You think so? Let me address those points:

Equivocation Logical Fallacy
- Example: "A warm beer is better than a cold beer. After all, nothing is better than a cold beer, and a warm beer is better than nothing"

I don't think so. I linked to BB article that contained a reference to the latest WuFlu estimate from the vaunted IHME model. I'm not mixing terminology.

No, you are mixing terminology because you're treating the "flu season" and the COVID-19 spread as comparable phenomena. This is the ambiguity at the heart of equivocation -- a linguistic sleight-of-hand that leads to an inappropriate comparison.

However, on review I will concede that a potentially better identification for the logical fallacy is the related one of false equivalence.

With regard to your updates, I will pick specifically on the magical thinking inherent in Texas "maybe" having peaked in Nov-Jan. There is no credible evidence to support this, only a claim made or amplified in the mass media. I will of course agree that the first incidence of the disease could have been here earlier, but it goes without saying that no huge growth pattern went unnoticed, not in Texas and not anywhere else. Look at how little it took for Taiwan to figure out something was up, and then figure out how we could have missed a contagion that actually provided some solid herd immunity. Don't bank on this fantasy, it's not so.

[viet4lifeOC]

I know this is frustrating for everyone, but to quote Lao Tzu, don't underestimate your opponent.

Lao Tzu was a philosopher and founder of Taoism.

Sun Tzu wrote "The Art of War"

I know...I know we all look alike

[as_rocketman]

Lao Tzu was a philosopher and founder of Taoism.

Sun Tzu wrote "The Art of War"

I know...I know we all look alike

Yes, I know. Check it out.