sawesome's picture


MEMBER SINCE   October 27, 2010

Recent Activity

Comment 12 Jun 2020

Best part is that RichRod publicly said that Tressel never negatively recruited against him:

Some people would think Ohio State-Michigan is such a big rivalry, they probably did a lot of negative recruiting. They didn’t. There were other schools that did, but they didn’t, and that to me sends a message they’re trying to do things the right way.

Tatgate was absurd.

Comment 16 Apr 2020

I have a 6000 employee manufacturing company in my jurisdiction that has had only a single confirmed case with a lower than average number of employees off by illness. Statistically we should have hundreds and hundreds of  infected employees that are showing signs of SARS-Covid-2. Granted, they were proactive in work from home for administrative staff and separating  work stations when possible but they should have been missing a large portion of their workforce. 

Lyman Stone occasionally posts data from Hong Kong.  In general, when you're trying to avoid one communicable disease, it looks like you avoid a lot of others, too.

Comment 13 Apr 2020

The possibility of a major backlash against medicine is becoming more prevalent with every "reasonable reopening suggestion".

What form would this backlash take?  I'm genuinely curious, not trolling.  I'm thinking about what you mean here; I have ideas, but they're long (several paragraphs snipped) and rambly.  More importantly, they're not your thoughts.

Comment 11 Apr 2020

One disservice that has been done by the media is promulgate very misleading fatality rates.  We can not accurately calculate a fatality rate currently (until we have antibody testing).  The reputable sources are placing it at less than 1% but no one knows for sure.  Whenever you see statistics like "9%" think "Lies, Damn Lies, and statistics" (sorry not trying to impune you if you did the math yourself - just saying if you see sources saying that, just close the website it is junk!).

To be clear, those are Sweden's own numbers from their own reporting, but yes—the CFR is going to be tough to pin down without knowing how many people are actually infected.  Even calling it "the" CFR is probably a little difficult, since fatality rates vary on geography and demographics.

Using a per-capita comparison illustrates the difficulty well.  Per capita deaths in Sweden is 8.8/100k (88 per million)[1]; NYC reports a much, much higher rate of 68.37/100k (684 per million)[2].  But the overall US rate is 5.7/100k (56.73 per million), about 35% better than Sweden.[3]  (Interestingly, the simple CFR calculation for NYC is better than the one for Sweden by about 35%, too. NYC's is about 5.8%, while Sweden's is, as mentioned, about 9%.  This may just be coincidence—it surprised me a great deal, but it may simply have to do with underreporting in NYC?  Or the fact that how Sweden keeps records compared to the US and NYC in particular is different?)

A holistic picture is hard to come by, and we really do need to look at as many points as possible.  I postulated in one thread that the peak-cases-per-day chart might have been back calculated from guesstimated CFR and R0 numbers combined with hospital load.  If you don't have per-capita rates—or the data sources are fuzzy (Wuhan), or the culture is totally different (Hong Kong), well...a hard nut to crack.

But on the balance, no one will get an argument out of me that we should be locking it all down ad eternum.  It's counter-productive, expensive, and depressing.  I hope testing goes a long way to helping us open things up, and that we figure out how to wear masks to help prevent the spread.  And I sincerely hope that FINALLY someone is going to install a red light over the bathroom door at work that goes off when people don't wash their freaking hands.




Comment 11 Apr 2020

So Ohio had a choice.  They could have behaved like Japan or Sweden or Germany (who have behaved with more moderation and where more data driven).  But instead they chose a path that even the IHME had modeled as being sub optimal.  This (IMHE) is the model everyone is talking about currently and what the government is using to assess the likely outcome.  The IHME modelling showed that if you do what Ohio did (and is doing) you actually are not helping.  You are only suppressing it for the time being and you will have to pay the price later.  The optimal model (according to their data and models) showed was suppressing it only enough to keep hospitals from getting overwhelmed.  They labeled this as somewhere between social distancing and shutting schools down. Lock down was shown to cause massive issues later on if you didn't maintain it indefinitely (which you can not do). 

Yeah.  I think a lot of decisions were made while staring at Wuhan, Italy, and Spain.  Those were and are bona fide disasters.

Sweden has taken a somewhat more open approach to matters.  I'm unclear about how they're recording their data right now, but the percentages look really bad:  about 9% fatality.  Wikipedia says they are reporting mainly from hospitals, which means that people who have COVID-19 that aren't hospitalized are probably not being counted, and that would, I expect, reduce that number by a lot.  They've been less heavy-handed, but it seems like their policies have been relatively similar to those implemented here and elsewhere.

What various Asian countries (and European?) had that we do not is a robust infrastructure for testing.  Testing is the key to ending lockdowns:  make it fast, free, and convenient, and let us get back to work.

Comment 11 Apr 2020

And this is exactly why we the public are so skeptical of "epxerts" and their "modeling."

I guess I don't exactly understand this sentiment.  Aren't experts part of the public?  Whose advice should one follow during the course of a pandemic?  How do you make any kind of estimates about what's actually happening?

I get arguments about the tensions between the public good and individual liberties, and I share concerns that stay-at-home orders are really not super helpful (it seems like mandatory mask wearing and canceling school and establishing sane guidelines for distancing would be better?).  I also understand skepticism of government.  (Ironically, that same skepticism is what got Hong Kongers to wear masks:  they didn't have faith that the government would implement adequate protections for the common good.  It's a flip story to the ones here.)

I'm not quite sure I understand the antipathy towards qualified opinions, although they aren't all they're cracked up to be.[1]

[1] Anecdotally, my pediatricians' office has a "tier of reliability," a sort of food pyramid for what to trust.  Expert opinion is the least trustworthy on the list—they prefer double-blind placebo-controlled studies.  So far, there just isn't a lot of clinical data in that vein.  While we don't have just expert opinions to go on, it's probably going to be awhile before we get to something better.  Dr. Derek Lowe is a great read for clinical matters, and he has some good posts about current therapeutics.  If you visit his blog, make sure to check out Things I Won't Work With.  Go for the laughs, stay for the chemistry lessons.

Comment 11 Apr 2020

The bill of rights also applies to the states as the 14th amendment was ratified. So, although the tenth amendment does give most of the power to state governments (like that happens anymore), the 14th amendment makes sure they have to follow the bill of rights. 

I think this is a non sequitur?  Congress has passed zero pandemic laws that compel you or anyone else to do anything.  But let's go with this. The relevant portion of the 14th amendment seems to be:

[N]or shall any State deprive any person of life, liberty, or property, without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws[.]

It seems to me that declaring a state of emergency constitutes due process.  It's subject to legal review and can be challenged in court.  (Even the Constitution allows for Habeas Corpus to be suspended in cases of invasion or rebellion.)  State governors have a great deal of legal authority during states of emergency.  This includes orders that protect the common good at the expense of individual liberties.  David French's explainer is a good place to read about it

Whether the states should have this power or not is a great debate to have:  but the state governments definitely do have this authority.

Comment 11 Apr 2020

Feel free to get a job stocking food in supermarkets or cleaning bedpans in a hospital! Plenty of those positions are open I'm sure and you'll be making money. If you don't want to do that, stop advocating for putting thousands of people's lives on the line (which is what will happen if we go the "herd immunity" route), stay at home, and stop complaining.

This is singularly unhelpful, especially to people who really are hurting economically and who don't have the freedom to get different jobs.  Contraction in the economy affects everyone's ability to get a different job—and not all jobs are created equal.  People also have to parent children who would normally be in day care or in school (even allowing for revised childcare rules in force during the state of emergency).  There will also be economic casualties:  people who fall into depression or drugs because they cannot get work.

I think it's the Right Thing in general to flatten the curve.  I think it's great that we can argue about whether or not we're overreacting because we've made some real headway in flattening the curve.  I'm glad we're arguing about civil liberties again in a (hopefully) constructive manner, and I hope people pay attention to business and government intrusions into their privacy and rights.  I also really hope that these stay-at-home orders teach us to enjoy our home time, if we can (we started a garden!).

And while I don't think you think there are no economic consequences or hurts, they really are dangerous in the same way the pandemic is dangerous.  We have a terrible baby to split here, and we absolutely need to get people out of their houses and back to work as soon as possible—just not sooner.

Comment 11 Apr 2020

But blindly shutting down larger portions of a state instead of taking more measured responses until there was a clearer picture (like there is now) is irresponsible.

The problem is that exponential growth makes this very difficult:  if it turns out to be a nothing-burger, you over-reacted and potentially destroyed your economy for a couple years.  But if not, you accidentally killed a lot of people and destroyed your economy (because dead people don't contribute to the economy). It's a tough baby to split, and I don't envy anyone who had to make those choices.

I will bet anyone that more people die in Ohio from drug overdose this year than Covid-19.

Sadly this is probably true.  Drug overdose victims are making choices—not always the ones they want to make, of course, nor the ones they would make with a structurally different world.  But I would guess a smaller percentage of people make the choice to infect themselves with COVID-19.

This is not merely an issue of actuarial deaths.  Is it an actuarial issue?  Yes, of course.  We don't shut the economy down because of the flu, even though it also grows exponentially (albeit with different parameters).  Our culture has changed, too, since today it feels (wrongly) that death is somehow optional.  Maybe we also value our freedom less compared to earlier generations (some things are worth dying for), or we don't understand it as well (all freedom comes with duty).

How about dealing with that emergency instead of destroying countless lives with overreactions based on cya considerations.

Yeah, it would be nice.  Government isn't a substitute for the social fabric we need to hold a community together:  it's a product of that fabric (since, after all, we are the government).  Government can't replace family members or friends or churches or gyms or workplaces.  It can institute policies designed to promote those relationships (e.g., child tax credit, marriage benefits, social security, corporate tax benefits tied to employment, et c.), but those policies aren't culture.  I don't know that any culture can be fixed with money or programs:  you gotta change the people.

Comment 11 Apr 2020

I thought the original idea was that hospitals desperately were lacking in things like ventilators and beds and masks etc, so to avoid swamping them, we needed to initially slow the spread so they could catch up with all that. But now it’s like no one remembers that or knows where to go from here or where the goal posts are.

if we just continue to do this, the spread it just going to continuously happen slowly, we need herd immunity eventually to mostly make this thing disappear. Now it seems like no one wants to be the one to make that call, and I understand their reluctance, but I need to go back to work...

Yeah, this is the real big issue:  people need to work, both for money and for dignity.  I have friends and family who have small businesses, and the stimulus loans take too much effort to obtain.  Some of these same friends are also significantly immuno-compromised, so they don't have the luxury of taking the risk of catching COVID-19.  No doubt about it, this is terrible economically, and we need to get to a point where people can go back to their jobs pronto.

The sooner we get to mass testing—no-cost-to-the-consumer, drive-through tests for anyone who wants it—along with volunteer serology testing to assess how far the virus has penetrated into particular communities, the sooner I think we can get back up and running well ahead of any vaccines or herd immunity.  We can also build fences:  implement temporary travel restrictions into healthy areas with mandatory screenings/quarantine.  This would allow local communities to get back to work with a degree of confidence that the rates of infection will remain manageable.

Comment 10 Apr 2020

There is no doubt that if we band driving cars there would be fewer traffic fatalities.

Traffic fatalities don't scale exponentially.  More people died of influenza (~66,000) in 2017 than in traffic accidents (~37,000).

Banning driving seems way more critical than continuing the lock down as the lock down has showed very mixed results. Compare NY to Cali. They both did the same rules are the same time and completely different results.

Lock downs may not do a great job, it's true.  Social distancing and mask wearing seem like they're better at helping to contain the spread, and closing schools probably helps a lot in densely-populated areas since kids are germbags.

The difference is density and 20 million people live in the NYC MSA with 1/2 using public transportation. Basically like no place in Ohio.

Population density definitely has an impact, but it's also a matter of public awareness and public policy.  Hong Kong is more densely populated than NYC and has far fewer cases (990 confirmed vs. 94k!).  Hong Kong has a more authoritarian government and a culture more disposed to treat pandemics seriously.

Texas has fewer deaths per million people and way less rules than Ohio. Maybe we should re look at our assumptions and see if the data still fits the narrative we where told.

Numbers vary a lot geographically for a lot of reasons, so we definitely need to be careful.  Ohio will look different than New York and Texas.  All models are wrong, but some are useful.

Anyway, time for bed.  Stay healthy, all.

Comment 10 Apr 2020

Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances.

  1. Congress has yet to make any laws that restrict any of these rights.
  2. States have broad authority to make such restrictions in times of need.
Comment 10 Apr 2020

When Cuomo and the Mayor said it is like the flu, I understand what they meant, and I disagree with their politics. That doesn't mean it should be taken lightly. I think all they are getting at is hundreds of thousands die just in the US from the flu. Mostly those with weakened immune systems.

FWIW, CDC reported about 66,000 deaths from influenza for 2017, not hundreds of thousands.  And people rightly point out that the common cold and other types of influenza are also coronaviruses.  So are SARS and MERS and a bunch of others.  Vocabulary can be a little difficult to get right sometimes.

The response from elected officials runs the gamut, just like the response of ordinary folks trying to make some sense for how to respond.  (They're also basically ordinary people.)  I don't know much about Cuomo or DiBlasio or their politics, and I've tried to avoid running afoul of the politics rules here (specifically by not advocating for any of my particular hobby-horses or trying to dump on anyone else's).  It seems reasonable to argue about it in good faith while doing our best not to infect people if we're the lucky guys who get to be Typhoid Covids.

Comment 10 Apr 2020

New York is an outlier. Because of political decisions. Not that it take away from Covid 19 being different than the flu.

Yup—exactly.  Typically the point made about Covid being no different than the flu is that mitigation efforts aren't necessary.  (Of course, we can say, "It's just the flu" because we have vaccines for common forms of influenza and medicines indicated for it.  "It's just the flu" precisely because we have history with it, and absent those interventions it probably wouldn't be such a "boring" disease.)

Comment 10 Apr 2020

Also, I'm not sure that 62,000 deaths was ever projected:  the graphic making its way around the Internet was 62,000 peak cases per day, which is a much different thing.

An additional clarification:  the graph shows 62,000 peak cases per day if no mitigation efforts were taken. The initial projection was 9800 peak cases if measures were taken.  So the overall reduction is substantially smaller:  from 9800 to 1600, an 84% reduction.  9800 on a CFR of 1% is about 100 deaths per day; and then at 1600 you're looking at 16.

Pennsylvania reported 78, 69, and 29 deaths in the last few days (not including today) on the back of 1579, 1680, and 1989 new cases respectively.  There's a lot of variance.  So these numbers that you're seeing are really not at all unreasonable.

Comment 10 Apr 2020

And an upvote for you, too. sir. But I stand by my point that it makes no sense for DeWine to credit his policies for squashing a March 22/23 peak of 62,000 projected deaths. If he wants to clarify and say that he was using the model figuratively, rather than literally, then by all means, let him do that, and then we can pick this point back up again. 

Also, I'm not sure that 62,000 deaths was ever projected:  the graphic making its way around the Internet was 62,000 peak cases per day, which is a much different thing.  At a CFR of 1%, that suggests 620 people dying in a day (maybe at one month lag time?); at 1.2% that's about 775 people per day.  That's a high number, but it isn't outside the realm of reason:  New York City peaked at 406 people in one day (5 April); Italy and Spain both reported numbers in excess of 750 in late March.  Comparing numbers like this requires some care—cities don't necessarily scale to states and neither states to countries.

Another difficulty with the 62k number is that it implies we can actually test that many people.  Probably you'd arrive at this number statistically, not brute force:  random sampling the population to find infection rates, the estimated fatality rate and reported number of deaths, estimated reproductive number, various population-level effects, et c.  There's also a lag involved, since dying might take you a month, and the incubation period is unknown.

Can social distancing reduce that by 1.5 orders of magnitude?  I don't know; it seems relatively high to me.  It's probably a combination of a lot of factors that drives a number from 62k peak cases to 1600 peak cases.

Comment 10 Apr 2020

Okay, if you say so. It's amazing how quickly actions taken during the week of March 8-15 squashed the 62,000 deaths projected by their models to occur one week later! Moreover, the half-measures taken earlier in March were obviously deemed by the public health "experts" to have been insufficient - everything before March 23 was insufficient according to their eventual decrees.

Have an upvote.

Since the disease spreads exponentially, getting ahead of the curve reduces cases exponentially, too.  So, yes: beating it by a week can, in fact, result in significant improvements.  It's true that all models are wrong and projections made from them are all wrong.  No one knows the CFR or R0 rates with a high degree of confidence, and models usually account for this by providing a few different possible outcomes.  Models also undergo refinement as the facts change.  (I think they call that "science.")

We need to argue whether the cure is worse than the disease, whether lock downs have the intended effect compared to social distancing (because that's not the same thing!), and how to balance rights of the individual vs. the common good.

Even so, it's factually incorrect to state that DeWine and Acton began mitigation on March 23.  It started a lot sooner and progressed as facts on the ground changed.

Comment 10 Apr 2020

Correlation does not imply causation. We do not know that “mitigation efforts” lowered virus impacts. That’s unproven. The three countries doing the best at social distancing, according to Google, are Italy, Spain, and France. 

One way to evaluate this is to look at other communicable diseases and how they're reported.  Lyman Stone has an interesting discussion of this on Twitter.  He's based in Hong Kong and gets paid for his demographic work; in general, he notes that reports of all notifiable diseases are down, even norovirus, and not just due to bandwidth issues.

Comment 10 Apr 2020

Strangely, though, DeWine’s lockdown order was issued on March 23! So somehow the magic of social distancing and economic shutdown went back in time to squash the virus. 

Dewine issued an executive order declaring a state of emergency on March 9.  He banned gatherings of 100+ people on March 12 and continued issuing executive orders pertaining to Covid in days preceding the stay-at-home order.  On 14 March the OH dept. of Health director closed all schools and on 17 March closed all polling stations.  All of the orders are available for public viewing.  Naturally as the situation progressed different measures were taken.  (I live in PA, and here individual counties were placed on stay-at-home orders while cases rose to help mitigate the effects on the economy.  I don't know how well it worked or how it affected our cases and modeling.)

It's also highly unlikely that everyone waited for government order prior to taking their own measures.  Covid-19 achieved a pretty high level of media saturation earlier due to the situation in China and Italy.

Comment 08 Apr 2020

Emergencies have always been the pretext on which the safeguards of individual liberty have been eroded. That's not my quote, but it's on the wall in my office. I don't lean toward collectivism. I know the times are a changin', because I'm starting to be the oldest guy in the meeting, but I'm not comfortable with were this is headed when the smoke clears. I know our rights have erroded since 2001 and the war on terrorism, and probably Reagan's war on drugs in the 80's, but I was holding out hope on hanging on to what's left. I was hoping as the models were showing themselves to be false that we would turn the corner, but instead everybody doubled down. My sister has restaurants in New Orleans. She's really the only one in the family affected. It just amazes me how few question what is happening in our nation right now, and how many appear to be content with the new normal.

There's a lot of truth to this, and one of the reasons that we need people to argue for individual liberties.  (Although it seems more likely to me that the erosion of personal liberty will come from the confluence of big business allied to government rather than a naked power grab from Washington.)

I don't know that historically it's true that individual liberties were given a privileged status over the common good.  The writ of Habeas Corpus can be suspended, for example, in times of rebellion or for reasons of public safety (Ohio's constitution follows the US constitution in this regard).  The Ohio constitution furthermore holds:

sec 4. Private property ought and shall ever be held inviolate, but always subservient to the public welfare, provided a compensation in money be made to the owner.

You can find exceptions like this all the time that hold the common good in tension against personal individual freedom.  By all means, let's argue about whether a global pandemic merits the suspension of our economic activities and compulsory mask-wearing, but it seems to me like this is one of the actually rather few cases where the case for individual liberties really isn't a slam dunk.

Comment 08 Apr 2020

I'll take the upvote, but was actually hoping you'd value my life and safety enough to give up on driving. Surely one that leans toward collectivism would see that it's only for the common good. Have an upvote on me.

Well, let me know where and when you're driving and I'll do my best.

In all seriousness, I think we should have the debate about it.  But there are good-faith, well-reasoned arguments about the responsibilities that go along with freedom.

Stay healthy.

Comment 07 Apr 2020

I constantly travel for work and know for a fact that if more people would be willing to relinquish their driving privileges, I would be safer on the interstate.

Have an upvote. 

Taken as a bare actuarial fact, you might be less likely to die from Covid-19 than you are from driving your car (though more people died from the flu in 2017—about 66,000than from car accidents—about 37,000).  It's equally true that deaths from other causes like heart disease and cancers will far outnumber Covid-19 deaths (about 1.25M reported by CDC in 2017) unless something truly horrible happens.  (This isn't to minimize the fact that if the 100k number projection is accurate, it would still be the seventh-leading cause of death in 2017 numbers; if the 200k numbers turned out, it would be the third-leading cause.)

There are important differences between dying in a car accident and dying from Covid-19, actuarial tables aside.  There's a legal framework in place with driving, for example, to punish those who are driving and shouldn't be (e.g., DUI, fines for less serious moving violations).  Pandemics by their nature have more extraordinary legal provisions in place that aren't regularly tested or a part of common experience.  There is also an element of choice when driving, too, an assumed risk with fairly well-defined parameters.  Covid-19, not so much.

So one way to help balance competing claims of freedom—of assembly, movement, religion, et c. vs. the freedom to actually live—is to make reasonable accommodations like masks and social distancing while out and about.