Thursday, March 19, 2020

like most people, my weight fluctuates mildly. 

over the last ten years or so, my bmi has fallen in the 18-22 range when it's been tested. i have gained a little bit of weight since going on hormones, but it's never been over 22.
"We're seeing younger patients, one a 29-year-old, some in their 40s, 50s. " Dr. Adam Jarrett, the center's chief medical officer, told NBC News on Monday, adding he was surprised patients' ages skewed younger than he'd expected.

All but one had underlying health conditions, such as heart or lung disease, diabetes and obesity.

"I am concerned about people who are obese," Jarrett said. "I wonder about lung capacity." 


so, perhaps we can add obese people to the list that currently includes the elderly and those with preexisting conditions.

do i care more about fat people than i do about the elderly? hardly.

we can create a total ordering here:

beer >> the elderly > obese people

i still want to go out and play. ugh.
33% of americans aged 45-64 and 29% of americans aged 30-45 are considered obese.

again: these numbers are different from everywhere else in the world.

so, there are claims that upwards of half of the patients being treated for covid-19 in the united states are under the age of 60. this is drastically different than data coming out of everywhere else in the world. how to make sense of that?

well, perhaps the first thing to point out is how unhealthy americans are - and this isn't simply another anti-yanqui screed, although it will no doubt become one if i end up ranting for long enough. you know this is mostly stream of consciousness, right? unless i pull out the math. there's really only any effort to order this if you see equations, that's the giveaway. 

but, it's true. americans have ridiculous levels of obesity, terrible blood pressure, off the chart rates of obesity, terrible heart health and even continue to smoke at high rates in some places.

the only data we have is their age. but, how many were obese? how many had heart disease? etc?

the second thing to point out is that if a lot of american cases are younger, that means they're at least doing a good job of staying away from the elderly. but, this is america. most people just stick their elderly relatives in a home and never call them ever again. whatever the reasoning, this is some cause for applause - they seem to be doing this right.

the third thing to point out is that it likely indicates a very, very high level of incidence in the general population. i don't have enough data to give you a rough calculation, but you're likely looking at hundreds of thousands or even millions of cases in the united states, alone. 

does it suggest that the virus is more deadly for young people than previously thought, or that young people should take extra precautions? well, we already knew it would be harmful for people suffering from obesity. so, i would not jump to that conclusion, no.
given the science here, and the mathematical likelihood of most people getting substantively sick, this stupid fiasco is already exceedingly difficult to justify - the vast majority of people are no doubt in a state of shock, and aren't thinking clearly. when they wake up...

if you can present an easy antibody test to demonstrate immunity and prove you're of no threat to the general public, it makes it that much harder to justify these draconian measures.

and, people pushing back against it should be denounced as anti-science luddites.
the readable version.

great.

so, test me. and, let me live again.

"but, i didn't get sick. that means i have to stay in? that's not fair, mom."

life's not fair, kid.

but, once these hit production, then you need to get the various local governments to fucking listen to the science, which has proven difficult up to this point. i may be given some bullshit collectivist line about doing my part and sacrificing for the team.

but, this is the problem with contract theory - i didn't sign up for this. i'm stuck in it. i don't want to do my part to help the community; i want out of the community. but, there's no escape - capitalism is ubiquitous. we're all stuck.

the ideal would be to get the government to produce immunity cards that you show at the border and at bars and venues. they already check ids, and they already kick out kids. so, it's not that onerous.

so this is what i wanted to see.

boris johnson - who i will repeat is surprisingly and refreshingly ahead of the curve on this - has labeled it a 'game changer'. and he's right.

i'd like to see some random sampling done.

but, i'd like to get tested for antibodies, myself - and i'd like to see the local authorities embrace this, and accept it as a get out of jail card. 

for example, i'd like to see them let me across the border if i test positive for antibodies.

if bernie's like your unstable uncle, biden's like your senile grandfather.
hey, kids, it's...

guantanamo joe!

"uhh. congress made me do it. and that one time back in 1873, i remember that..anyways."

(pause)

that's right, kids - guantanamo joe!
you know, i wasn't taking tulsi gabbard very seriously, but she was supposed to be the anti-war candidate, right?

it's a little odd to see her endorse guantanamo joe.
this is really the truth.

we're all modeling. we're all guessing.

and, these lawyers (at best.) in charge of the political system don't know how to do anything except what they're told. most of them couldn't even follow the calculation i just produced - which is roughly at a grade 12 level.

it's why it's so important to have the right people in the ministries; the executive power is clueless, across the board.

well...

how many people have the flu at any given time?

google's not helping.
2% of the people in the detro metro area would be 85,000 people infected.

there's roughly 100 known cases in all of michigan.

if 85% of the number of actual cases are unknown, a number used in models by epidemiologists, then the number of actual cases in all of michigan is pushing 700. 

(2^n)*700 = 85000 <---> 2^n = 121.43 <---->nln2 = ln(121.43) <--->
n = 6.92398108175

so, if the number of active cases doubles every week, you've got 7 weeks before you get to 2% - which doesn't include dead or recovered cases.

and, you should be able to crunch 5% or 10% or whatever else you want - but understand that this is concurrent and if you get to a number like 20% concurrent then you must be peaking. that number might actually be impossible.

but, is 5% possible? you figure that out.
it's not a question of "staring down death", it's a question of understanding the probability.

what is the probability of dying of coronavirus if you're under 50?

well, first you'd have to get it. what is the probability of that? the city that i live in has zero cases. there's some cases in michigan. if we assume there's even 1000 cases in detroit itself, and that it's evenly distributed amongst the population (a generous assumption), then there would be a 1/4200 (0.02%) chance of catching it, for every person you interact with. 

now, let's say you interact closely enough that transmission is possible with 5 people when you go out. then the probably of getting sick is equal to:

P(x1 or x2 or x3 or x4 or x5) = 
P(x1) + p(x2 or x3 or x4 or x5) - p(x1)*p(x2 or x3 or x4 or x5) =
.0002 + 0.9998[p(x2 or x3 or x4 or x5)] =
.0002 + 0.9998[0.0002 + 0.9998(p(x3 + x4 + x5)] =
.0002 + .000199996 + .9998^2[0.0002 + .9998[p(x4 + x5)]] =
.0002 + .000199996 + 0.00019992 + .998^3[0.0002 + .0002 - .0002^2] =
.0002 + .000199996 + 0.00019992 + 0.00039972007 =
0.00099960007 =
0.09996007%

that's not 1%. that's 0.01%. so, that's your chance of getting sick right now, in michigan, in the worst case.

now, if you're under 50, you have about a 0.1% chance of dying given that you have the disease, and you pretty much must already be sick.

so, if A is the probability of getting the disease,
and B is the probability of dying from the disease,
we have

P(B|A) = 0.001, P(A) = 0.00099960007.

as P(B|A) = P(B & A)/P(A),
P(B & A) = P(B|A)P(A) = P(A & B) = P(A|B)*P(B)

so, P(B) = P(B|A)P(A)/P(A|B) 

but, (A|B) = 1.

so, P(B) = P(B|A)*P(A)  = 0.00000099960007 =  0.000099960007%.

those are pretty good odds - if you're young, and the virus is not running rampant. 

what if you were over 70 today, in michigan?

then you're looking at adjusting P(B|A) to around 10%. and, then P(B) = 0.0099960007%. those are still pretty good odds, then, even if you're older.

now, what if 2% of the population were infected, instead of 0.02%?

P(x1 or x2 or x3 or x4 or x5) =
P(x1) + p(x2 or x3 or x4 or x5) - p(x1)*p(x2 or x3 or x4 or x5) =
.02 + 0.98[p(x2 or x3 or x4 or x5)] =
.02 + 0.98[0.02 + 0.98(p(x3 + x4 + x5)] =
.02 + .0196 + .98^2[0.02 + .98[p(x4 + x5)]] =
.02 + .0196 + 0.019208 + .98^3[0.02 + .02 - .02^2] =
.02 + .0196 + 0.019208 + 0.0372712032 =
0.0960792032 =
9.60792032%

now, that's actually 10%. or 1000x the previous amount, roughly.

now, if you're under 50,

P(B|A) = 0.001, P(A) = 0.0960792032
so, P(B) = P(B|A)*P(A) = 0.0000960792 = 0.00960792%

those are still pretty good odds - if you're young.

but, what if you were over 70 today, in michigan?

then you're looking at adjusting P(B|A) to around 10%. and, then P(B) = 0.00960792032 = 0.9%. and, we have our scary stat - that's almost 1%.

now, what about 20% of the people being actively infected, something that's only ever going to happen just right now?

P(x1 or x2 or x3 or x4 or x5) =
P(x1) + p(x2 or x3 or x4 or x5) - p(x1)*p(x2 or x3 or x4 or x5) =
.2 + 0.8[p(x2 or x3 or x4 or x5)] =
.2 + 0.8[0.2 + 0.8(p(x3 + x4 + x5)] =
.2 + .16 + .8^2[0.2 + .8[p(x4 + x5)]] =
.2 + .16 + 0.128 + .8^3[0.2 + .2 - .2^2] =
.2 + .16 + 0.128 +  0.18432=
0.67232

for healthy people, .001*.67232 = .00067232 = .067232%. but, this is an overestimate, because you probably don't have a preexisting condition. that is, you don't really have a .1% chance of dying if you're healthy...

but, the numbers for the elderly are actually pretty realistic.

.67232*.1 = .067232 = 6.7%. and, that's a very scary number.

so, let me tell you - you want the elderly to stay inside, pretty soon. as soon as immunity starts building, their chances of getting sick starts to skyrocket.

so, are you going to tell me i'm a daredevil for taking a chance that low? ok.

but, note that those are better odds than you get from most actual daredevil acts.

so, i don't really think i'm invincible. really. i'm just not as bad at math as the media is

https://www.cnbc.com/2020/03/18/psychological-reason-why-gen-x-is-taking-covid-19-pandemic-seriously.html
"So again, I'm going to call on that generation ... not only calling on you to heed what's in the guidance, but to really ensure that each and every one of you are protecting each other," she said. "We cannot have these large gatherings that continue to occur throughout the country for people who are off work to then be socializing in large groups and spreading the virus.

no, deborah.

the best way we protect each other is by building immunity. that's how this works; it's how it's worked for millions of years.

protecting the weak means getting them out of the way.

there isn't a choice. and you're just wasting time denying it.

is it just me, or does that fauci guy sound like a pack a day smoker?
well.

there's an incentive to get tested.

hey dougy, how's about sweetening the deal a little? 

some more actual science. and, this is a prominent publisher.

this one basically says that suppression is futile, because there's millions of undocumented cases and there's nothing you can do about it. 

listen - i'm not a doctor. but i get the math intuitively. and this was immediately obvious....

these hamfisted lockdowns are just needlessly harming people's livelihoods. 

you need to lock the old people up. anything else is just sacrificing them to baal.

i'm also going to ask another question.

hong kong is being presented as a model of success as a result of it's draconian response. "look", the fascists say, "hong kong is a totalitarian state, and they're not sick. they're keeping the trains running on time!".

well, there's some leaps of logic there. we don't know that this is the reason. because....

we just went over that one.

i have another suggestion. hong kong was also ground zero for the last sars outbreak, you'll recall. 

is there some immunity there from that? is that it?

who else got hit with sars in 2003?

taiwan & singapore, mostly.

hrmmn. they've been basically spared by this, haven't they? could it be that there's some immunity there, too?

you could test this - you'd have to do a random sample of people in hong king and test them for antibodies.

if that's the case, it's some good news for the future, because it means the next time this happens it won't be so bad.
i would predict that it is highly likely that you'll see some further hotspots develop in china fairly soon, perhaps on the other side of the country.
the reality is that there are likely millions of people already infected in europe. with 3,000 deaths in italy, and a fatality rate of 0.6%, a simple calculation suggests over 500,000 actual cases in italy alone.

there may be substantial levels of immunity building in northern italy right now, as well.

so, if they lock the country down right now and the numbers start coming down two weeks from now, do you jump to the conclusion that it was a lock down?

and, i'm going to leave you hanging with this for a few more sentences, because you had to know this was coming. i've been criticizing the scientific response by governments for a while now - basically calling them all a bunch of stupid thugs acting like goons in trying to strong-arm their way through this - and there's a few phrases that are inevitably going to end up in any such critique. really, it's sort of amazing that we haven't gotten here yet, although i foreshadowed it a little bit with the pissing in the flower bed analogy, and the suggestion that the efficacy of an overly broad measure doesn't justify the overly broad actions.

so, where are we going? say it out loud, kids. you know this one...

correlation does not imply causality!

you have to work that out. i'd be just as wrong to deduce it must be immunity. but, i have my views. and, what i've seen suggests that the studies being done right now are really leaning in my direction.

so, this might burn out in europe relatively soon.

but, don't deduce it's because of the nazi lockdown. it could very well be that the thing is actually out of control, and cases end up coming down due to immunity reaching an inflection point.
this is some actual science.

it suggests that the mortality rate in china is probably about the same as the published rate in south korea - somewhere around .08%.

it also suggests that roughly 20% of the population in the affected province may have contracted this disease by the end of january. 

the truth may be that china may be approaching herd immunity.