Georgia on My Mind

Georgia ramped up restrictions over time. They then relaxed some on April 24. As a result, people are currently enjoying arguing about Georgia. Even though I think it’s a few weeks too early to see if the trend in deaths changed after their relaxation, we are entering the time period where we might start to associated changes in death rates with the policy change enacted by the Governor.

I modified my script so I can start generating some for different states adding some possibly informative dates. (Adding the dates results in clutter. But you can see dates when some things happened with solid vertical lines each with a corresponding dashed line 18 days later.)

Update May 19 In comments Tom suggested we should think how people’s behavior lags government actions. In particular, that fear or caution might prevent them from all rushing out (which is a good thing.) I’d thought of that and had started making halfway decent graphs of change in time spent places based on Google Mobility data. I passed some google mobility data through a 7 day smoothing filter.

Here’s time spent at home with the red line indicating the day the rules were first relaxed. I show GA as a whole and the 7 most populous counties. (There are actually 2 “relaxing” days. One is 4/24, another is 4/27:

Here’s time spent at work.

Here’s time spent recreating.

We can discuss in comments.


Update: This is Georgia’s case and death rates updated on June 14.

80 thoughts on “Georgia on My Mind”

  1. It looks from the graph that new severe cases correlate with deaths at close to 1:1 with the deaths happening about a week later.

  2. DeWitt,
    Yep. The correlation with Critical is even stronger. Those deemed critical pretty much die. It’s the same with the Illinois data.

    I plotted them because they are in thise “data_categories.csv” file.

    Using the very potentially error prone “eyeball” method, it looks like death’s flatten about a week after cases. But honestly, the data are sufficiently noisy that I might just be seeing what I want to see there.
    One way or another, all fur metrics were down 12% over the week. (I actually had to check if I had a bug in the code when I saw they were ALL the same.)

    (I’m trying to gear up to comparing lots of things, including, potentially, looking at mobility data. But I’ll eventually want to do more than one state. So the motivation of GA was to make sure my script is getting generic enough to be more than just IL.)

  3. lucia,

    I meant critical rather than severe, but the brain and the fingers on the keyboard weren’t correlated.

    There was an article in the WSJ about how the CDC should be looking for treatments that work and spreading the news to doctors on the front line. They don’t seem to be doing this. Pelosi was asked on one of the Sunday talking heads shows and, of course, blamed any failings at the CDC on funding cuts, i.e. Bad Orange Man.

    Here’s the article by Dr. Scott Gottlieb, former FDA commissioner:

    Cut Through the Fog of Coronavirus War
    The CDC needs to streamline and publish clinical data to help doctors on the front lines.

    But CDC and its highly capable career experts must be elevated to play their role in reporting on these findings in real time, so medical practice can be quickly informed of the latest information about Covid patients. Whatever the reasons, CDC has spoken infrequently and with more reticence than is customary in public-health crises. Policy makers may worry that prescriptive guidance and descriptive clinical findings will fuel public fears or constrain a reopening.

    The opposite is true. The more information about how to reduce the risk of spread and the severity of sickness, the more lives that can be saved, and the more comfortable Americans will feel about starting to resume normal life.

  4. Those who said the goveror would “have blood on his hands” in two or theee werks were utterly wrong. Will they acknowledge rhat? Hell no! Now it’s wait three more weeks for the Georgia catastrophe to happen. When that prediction fails in three more weeks, they will simply refuse to address their past incorrect predictions. Here is the fundamental problem: The thinking of most conservatives is bound and guided by factual reality reality. The thinking of most progressives is not. Progressive thinking is pretty much disconnected from reality, and certainly not guided by reality. It is more like religion than rational thought. And therein lies the potential damage.

  5. I suspect that behavioral changes are going to lag the reopening by some undefined amount. People had the crap scared out of them so are going to avoid contact with other people. This is actually a good thing as it allows reopening to be safer. A sharp increase of gatherings in enclosed places would lead to spikes in infections. Super-spreader events that then take it home. Small cramped offices with poor ventilation, no thanks.
    .
    People should be told some basic things like run the ventilation fans constantly and do not recirculate the air if possible.

  6. Some hilariously strained phrasing by the NYT:
    https://www.nytimes.com/2020/05/18/nyregion/coronavirus-deaths-nyc.html
    .
    “While the majority of the deaths across the city have been older residents, race and income have proven to be the largest factors in determining who lives and who dies.”
    “Although the area has the city’s highest concentration of people over age 65, it was unclear why its death rate is so high.”
    “While the vast majority of the city’s deaths have been people 65 and older, the overwhelming difference between the neighborhoods that suffered most and least has been race and income, not age.”
    .
    Too good not to be true I guess. Is it really that hard to say age is the predominate factor, and race/income are other correlations? Age actually has very intuitive causation, where race/income have no obvious virus causation, and the NYT makes ZERO effort to supply it. Also note complete silence on the gender gap because it is in the “right” direction.

  7. Tom,
    That’s more or less what I think. I coded up mobility data and I’ll be adding it on an update.

    There are some nuances.

  8. Tom,
    Well….. of course it’s “unclear”. But I would strongly suspect the factors (which are necessarily associated with low income levels) include:
    1) more occupants per square foot of living space.
    2) more use of public transportation.
    3) more multi-generational households.
    4) more people/apartment or dwelling.
    5) poorer ventilation. (This is partly due to less control of HVAC, window opening etc. My sisters apartment has it’s own AC and furnace. She can crack windows.)
    6) lower average education and so less understanding of public service announcements.
    7) more likely to be sharing elevators stair cases.
    8) lack of ability to prevent your neighbors from having a big party. ( Swanky places have building security.)
    9) Lack of ability to prevent guests from visiting. (I can’t visit my sister in her high rise. The building is not allowing anyone other than residents to come in and out. That’s not going to happen in low income areas. )

    We know there are also more co-morbidities like diabetes and obesity with the poor now a days. I’m sure I can think of more. But yes, wealth is protective in this particular contagion. I’m sure it is in most but in this one, it definitely is.

  9. I would agree with all the factors you cite except for education. (As a high school graduate I hope that’s forgivable.)

    In past crises the more poorly educated have split almost evenly in their reactions, with males being more foolhardy and females more conservative than the public at large. I would expect the same to hold true here.

  10. Thomas,
    At any level of education, there is a range of ability to understand public health anouncements. I stand by the notion that if we are talking about an entire neighborhood one with a generally lower level of education will have a higher proportion of people who will not understand (and possibly not even read or listen to) public service announcements.

    Saying the have lesser understanding is not an accusation of being “foolhardy” and so on.

  11. It’s interesting to me the extent to which income is treated as an inherent cause of so many bad outcomes. Certainly there is a logical consequence of being poor that limits one’s ability to buy their way out of those bad outcomes.

    There seems to be a split that falls along party ideology the extent to which people perceive that income is the result of choices and behaviors . One can further attempt to attribute underlying factors to those behaviors but the simple fact is that if you spend $100 a week on scratchers and lotto tickets you’re extremely likely to end up with a lot less wealth than if you put that same amount each week into a Roth IRA.

    Not sure where I’m going with this pondering. It’s just that so many treat this as if there is a deterministic calculation that can be made, if only we can deduce the relevant underlying contributing factors. And then wave their flag of social disapproval at the bad outcome.

  12. Earl,
    I really don’t think the distinction is between people who spent on scratch off tickets vs. those who put money into a Roth IRA!

    I do think, right now being poor right now puts one at more risk for this disease. So, to a large extent, being poor is sort of “causal” for “getting sick”.

    Whether things someone did in the past– perhaps in the long past contributed to their policy or whether what their parents or society did are useful questions. But I don’t think that’s especially relevant to the fact that in the situation right now poverty does put one at risk.

  13. @ Lucia,

    I agree with your statement regarding the risk for folks in New York city and other high population density environs. I suspect that in rural areas the differential rate of infection based on income is marginal, if even detectable. Just speculation on my part, but the factors you listed are greatly reduced if not non-existent.

  14. “more occupants per square foot of living space” et. al.
    .
    Yes. So why don’t they list those as the causation instead of using race/income as a proxy? They want to imply it is malicious racism and classism as the cause. They want to imply that the victims have * no agency * in their outcomes. It’s a lazy framing.
    .
    Certainly the poor have limited options in where they live, but they can control certain aspects of their health, certain social distancing aspects, and so on. What I object to is the framing that racism/classism is the direct cause and there is no hope for improving agency. Sometimes it is bad behavior, but not always. It’s an important distinction. Poor educational outcomes are treated the same in NYC.
    .
    The virus is not validating income levels. It is certainly easier to avoid the virus with wealth, but what are the poor actually doing besides being poor that puts them at risk?

  15. Tom Scharf,
    They probably don’t have all those things in the database they put their number cruncher to work on.

  16. Tom Scharf,
    “ They want to imply it is malicious racism and classism as the cause. They want to imply that the victims have * no agency * in their outcomes. It’s a lazy framing.”
    .
    I think the explanation goes like this: “If there were no malicious racism, then all races would have the same average income and there would be no classes.” Which is as air headed a take as I can imagine, but it is the (1619) NYT, where everyone knows the minutemen were fighting to protect slavery from the British. That slavery was not prohibited by England at that time makes no difference, of course. Idiots.

  17. Is is reasonable to send kids to summer camps, with a very high likelihood(close to 100%) of getting infected?

  18. It might be worth re-visiting Georgia’s data now… averaged daily deaths appear to have hit a low point right around the start of June and have been climbing since then.

  19. Skeptical… Yeah… Cases stayed flat until the riots and then ticked up. Deaths were flattish…. then ticked up a bit after cases ticked up. The timing is going to make it impossible to decree whether cases are the riots or the opening of the state. The deaths are just a bit too soon for me to think it’s the riots. I’ll post tomorrow.

  20. I see no patterns in the Georgia data other than slight downwards trend in new cases (since early April) and deaths (since mid-April); and a more distinct downward trend in people staying at home since early April.


  21. The most recent week has a distinct rise in deaths. Could be “noise”, but it’s bigger than most the wobbling around since April.
    Cases suggest an uptick too– in the sense that they are up two weeks in a row. That’s not much though. That’s why I say “ticked up”. I wouldn’t use any stronger term. (Not “spike” etc.)

    I think mostly, this is a “watch” situation rather than a “conclusion” situation. Ordinarily, I would wait a little longer to put the update, but skeptical asked. So, that’s my description of motion. The image is obviously better.

  22. Yeah…. I don’t know. I absolutely know we won’t be able to distinguish deaths due to moving into phase III from deaths due to protests/riots in Illinois. It went “most state opens”, “protests/riots”, “Chicago opens”, in fairly quick succession.

  23. Lucia,
    My guess is that the increase in cases will be way ahead of the increase in in deaths. Still, don’t even imagine that you can return to normal. You need to suffer much, much more,

  24. It seems that hospitalizations with the Wuhan virus are rising even as hospitalizations for the virus fall. People are returning to the medical system to get treatment for their heart attacks, cancers, and “elective” surgeries. They are tested for the virus and some of them have it. That causes an increase in cases, by detecting cases for which the patient would never have sought a test or treatment. It causes an increase in hospitalizations, even though the virus has nothing to do with the reason for hospitalization. And it will no doubt cause an increase in deaths with the virus.

  25. MikeM
    The ratio of positive tests to total tests is also rising in Florida. If your theory of why people are getting tested were true, I would expect it to drop. Sure, some asymptomatic people who arrive at hospitals for other reasons turn out to be infected. But you wouldn’t expect the rate at which they are infected to be higher than the rate for symptomatic people who got tested because they thought they had covid.

  26. lucia (Comment #186372): “The ratio of positive tests to total tests is also rising in Florida.”
    .
    I did not say anything about Florida, which does seem to have a strong recent rise. But that sounds like a good idea for checking whether a rise is due to more testing.

  27. MikeM,
    Sorry.. I think I got onto florida because I got diverted to an article about the woman who used to run the FL dashboard starting her own!

    I’m going to put some graphs up for positive/new tests in GA and tests for GA. The latter is just smoothed… (and weird to me!!)

  28. Mike M, This is supposedly the case/new test ratio in Georgia…. (Well… sort of..)

    I pulled data and computed the ratio of “positive” to “new tests” in GA and also the number of “new tests” from this source. The open circles are 7 day smooths. (I actually work from cummulative and difference for reasons I’ll explain tomorrow. I had a trap to avoid “resurrections” which suggested a few people occasionally came back from the dead….)

    Anyway.. my graphs look like this:

    Note: There is something very weird with the data….. (The weird thing is that it claims there were -74189 “new tests” on 5/27. I did something to deal with there *small* glitched in records in the past… but this is such a huge glitch that I’m not sure we can interpret much. )

    The “glitch” in the data is so overwhelming that I think we have no idea what’s happened to the new case/ new test ration on the GA.

  29. Lucia,
    I suspect there is an explanation for the negative number of GA tests, but it may be very difficult to find. Deaths are probably a more reliable measure. I sure wish FL Gov Desantis would order that every hospital entry require both a swab test for active infection and a blood test for antibodies. If that were done in a few other states, with both high and low death rates, then the true extent of infection and a better estimate of Rt would be available. Alas, we can spend trillions on all kinds of things but it seems not a penny on actually understanding and defining the spread of the illness.

  30. The ‘big three’ (NY, NJ, MA) all have dramatic drops in daily deaths from their peaks… as much as a 95% drop in NY. How much is masks and social distancing? How much is from growing herd immunity? How many people had existing cross-resistance from other coronavirus infections? We don’t know. Will we ever know? I am not sure we will. Governors in those states are happy the rates have fallen dramatically, but have zero motivation to know if there is already effective herd immunity, since that would raise questions about all the draconian rules.

  31. SteveF,
    Here’s my guess on why there are negative numbers of tests.

    Based on older news stories, I think lots of states were counting both serology and viral tests as “new tests”. Then they decided it should only be viral test. Someone went in one day and fixed the “cumulative” value for at that point in the tally. But they didn’t go back and change all of them. So, the cumulative number drops.

    That’s a theory. Might be right. It happened a few weeks ago. As a result, I can’t really say much about how “new cases/ new tests” varies because the denominator is wrong.

    I should probably just have a “before shift” “after shift”. But I only just noticed this in cases.

    Oddly, I sort of should have noticed earlier– but In ever really discuss these graphs. They were just for sort of seeing how tests were rising whine at home. I DO notice that LOTS of states have a sudden drop in the rate that new tests are performed! But none that I looked at as big a GA. So I had thought… “Hmmm… must have been pent up demand… then we got back to normal demand.” But with GA, it’s so severe no one could fail to notice that…… Well… bettter look at the actual numbers!

    There was always a slight issue in number of deaths. But those numbers were never more than ±2 and happened early in the record. So I just figured they were transcription errors of some sort. That’s not unusual in records complied by people who pass papers from office to office. I decided how to deal with that and things were fine for eyeballing stuff. But this… well…

  32. lucia (Comment #186380): “Someone went in one day and fixed the “cumulative” value for at that point in the tally. But they didn’t go back and change all of them.”
    .
    I think that is the right interpretation. But even if the state of Georgia went back and fixed the earlier numbers, the COVID tracking project did not. I have seen problems of that sort with their New Mexico data. So maybe the web site where the state reports their data has usable numbers.

  33. lucia (Comment #186366)
    June 14th, 2020 at 1:17 pm

    Ordinarily, I would wait a little longer to put the update, but skeptical asked.

    Thanks Lucia. To me the uptick looks a lot more than what would be expected from “noise”. I just wanted to see what other people’s thoughts were.
    .
    (Real) question for everyone: how long would an uptick have to be sustained before you would consider it to be a trend? Are there any rules for this, or is it purely subjective?

  34. Some states are now mixing positive antibody tests with positive PCRs.That’s just another noise factor clouding the usefulness of new cases as a metric since a positive antibody test isn’t necessarily tied a recent infection. AZ is one that I know of.

  35. Skeptical,
    If I were betting and using that as data to predict next week, I’m lean toward the uptick not being merely noise. But it might turn out to just be “noise” and not indicative of a general trend.
    .
    To decide my own behavior, I look at local data. I don’t do zip code because…. well… even if my grocery store was in my zip code (which it’s not) I wouldn’t expect all the shoppers, clerks and so on to necessarily be from the zip code. But I check my county and the adjacent ones.
    .
    So…. not looking into the GA issue to learn why there was a huge negative number of daily tests in the record! (But I’ll bear it in mind when interpreting Illinois cases/new.)
    .
    Having said that, this is my more detailed theory of the GA data


    Somewhere around day 65-70, a large proportion of “serology” case numbers started being added to the “cummulative cases” record. That’s why there is a trend discontinuity.

    Some day just before day “80”, they figured out something was odd. (There were various news reports and commentary on this issue.) Someone went back in and yanked out the identified “serology” numbers. After that the rate of increase is more or less correct. (But my method of dealing with it makes my plot look weirder after the correction.)

    Of course this is all speculative. So even if I fix based on this, we still don’t know. But it would suggest that my ratios bewteen day 65-8? are not worth interpreting.

  36. From what states I’m tracking I don’t get excited about changes until well over a week has passed and that’s even with using the 7 day averages. I’d want two weeks to pass before calling it a trend. There’s too much of a 7 day cycle in the data and changes in testing policies have caused huge spikes over a few days.

  37. Luica, Mike M, Andrew P,
    If the states are mixing the two tests and calling them all new cases, then there is no possibility of figuring out the true rate of new cases. That anyone at the individual health departments would choose to combine them is as clear an indication of incompetence as I can imagine.
    .
    Better to just look at deaths. The NY, NJ, and MA death curves are exactly what you expect for a drop of Rt to well below 1.0. Too bad that why the Rt fell well below 1 can’t be determined.
    .
    Yes, even the number of deaths can be corrupted (fatal car accident victim or heart attack victim tests positive, someone found dead at home not tested), but deaths are far better than combining new cases and old cases and declaring them all “new”… as they say, death is almost as certain as taxes. ;-0

  38. A medical testing lab in my area is now offering serology test for Covid for only $10. The catch is you have to have a doctor’s prescription. The non-prescription tests remain over $130.

  39. Andrew P,

    I’m only looking at deaths… the case numbers seem meaningless to me as they are only as good as the testing, whereas the dead are all dead (provided they stay dead) and can be fully counted.

  40. SteveF,
    Heh. My guess is they can bill insurance if you have a prescription. (Total guess…. but who ever this is does need to make a living.)

    I’d wait until I went to my doctor for some other reason and ask for the prescription. 🙂

  41. skeptical,
    The overwhelming majority of the dead (in the record) stay dead. To the extent that one or two resurrect, they appear to die shortly there after. 🙂
    .
    The large range of issues with “testing” is why I’ve tended to take deaths as a more reliable indicator of the trajectory of the disease. It is a lagging indicator, but at… oh… well…
    .
    In principle new hospitalizations could be useful and less laggy. But that also depends on hospital bed capacity and other choices.

  42. AndrewP

    From what states I’m tracking I don’t get excited about changes until well over a week has passed and that’s even with using the 7 day averages.

    Yep. Though I do notice big changes. For local counties, I google to find news reports. It will often be a food processing plant, nursing home outbreak etc. These really make a small counties per-capita deaths spike. The nursing home spikes then drop quickly (measured in time scale of the disease.)

    SteveF

    From what states I’m tracking I don’t get excited about changes until well over a week has passed and that’s even with using the 7 day averages.

    Yep. That’s just the point made a few months ago. If I’m not mistaken the cdc was mixing!! Some states did too. Once discussed publicly, I think everyone understood “the issue”. But the data sets I’m drawing just have columns for “new cases” and “new tests”. They don’t have separate ones for “serology” vs “virus”. There’s no note to allow me to distinguish (and i suspect most the state pages don’t say either.)
    .

  43. Lucia,
    From the “The price is low, but we don’t have any” department:
    .
    One drive-up testing sites advertises free (or nearly free) test for both swab and serology tests. You just have to fill in an on-line form to get a scheduled drive-up time. At the bottom of the form (in fine print) “Due to the high demand and limited availability of testing slots, we will only do swab tests for active illness, not antibody tests for past illness.”
    .
    Gotta love it.

  44. I would guess that the number of contacts between people in reopening the economy dwarfs the number of contacts during the protests. Also the protest group was younger. One could in theory determine how many cases were tied to protests by asking some questions but I think the establishment will be unwilling to look under that rock for fear of what they might find.
    .
    After giving explicit permission from public health “experts” for protests, they are now shamelessly and predictably opining on Trump campaign events. I think it is crazy to go to a campaign event, but the double standard just makes me ignore the entirety of public health expertise filtered through the media.

  45. I think the FL death count is going to rise in the next few weeks. The active case count sure looks to be heading the wrong direction. There is a possibility that the most infirm in old folks homes have already been taken out so there is less opportunity. If they could that under control then the death count could easily go down while the cases go up.

  46. AZ doesn’t separate them in the new case count. They do list the daily number of tests and positive rates for each on the dashboard which is why I know they are combining them.
    Deaths is the most accurate obviously, but it is also the most random. Even the 7 day average in Ohio swings up and down by 50%. I had been using hospitalization rate as a leading indicator, which works ok in OH, but in AZ that data has issues. https://imgur.com/a/61bkj8k

  47. Florida released results from a state-wide serology study (details are sketchy). The results suggest two to three weeks ago 4.4% of residents of Florida carried antibodies. That means that by 3 to 4 weeks ago, just over 1 million people in Florida had contracted and recovered from the illness. Three weeks ago the number of confirmed cases was about 55,000. So the number of asymptomatic (or mildly symptomatic) untested cases was approximately 18 times greater than the number of confirmed cases. Two weeks ago there were ~2,700 total deaths in Florida. So, a reasonable estimate of the average infection fatality rate is about 0.27%.

  48. As of May 27, GA started separating serology tests from PCR. This caused a the number of tests to decline on that day from the combined number.

    https://covidtracking.com/data#state-ga
    So lucia’s theory is confirmed.

    They have a link to the Georgia health department web site, but I am not seeing coronavirus data there.
    .

    As of May 15, Florida is reporting both specimens (PCR and antibody) and people tested. We report positives and negatives based on the number of people tested.

    https://covidtracking.com/data#state-fl
    So it seems like the Florida data is contaminated by mixing unlike things. Maybe that has something to do with the recent rise.
    ——–

    Addition.
    SteveF (Comment #186400): “Florida released results from a state-wide serology study (details are sketchy). The results suggest two to three weeks ago 4.4% of residents of Florida carried antibodies.”
    .
    The recent rise in positive test rate in Florida has been in the direction of 4.4%. So that could be due to mixing in an increasing fraction of serology tests.

  49. Mike M,

    That explains the bonkers divergence between “cases” and deaths. The stupid, it burns.
    .
    Since we know there are about 18 seropositives for every positive swab test, there is no real limit to the number of positive cases that will be reported. Lots of people who had a mild fever or sore throat in the last 12 weeks want to know if they carry antibody. As those tests become more common, the trend in Florida case data will cease to have meaning. The only data for Florida that will be meaningful is the trend in deaths.
    .
    BTW, the serology study in Florida concluded that ~60% of seropositives were asymptomatic, and ~40% had some type of mild symptoms.

  50. SteveF

    Do you think Florida daily deaths are going to jump (say 70 or 80 deaths per day) within 2-3 weeks? I doubt it.

    When I first made the graph yesterday (after reading an article) and glanced quickly, I thought there would be a pretty good chance deaths would rise after cases rose.
    Lucky for me I screwed up, responded to MikeM’s discussion of GA with somewhat irrenevant observations of FL…. So then of course I had to look at GA. The effect of whatever correction was done on the GA data was so obvious, and I thought back to other states case numbers that had made me think “hmmm” (but go no further.) So now…. I think the case info may be so subject to changes in counting that I have very little confidence that changes in death rates can be predicted based on the case data I have access to right now.
    .
    So… dunno.

  51. Tom Scharf

    Also the protest group was younger.

    Yeah. Given what we know about the virus having large numbers of almost universally young protestors create #CHAZ or #CHOP groups in every major city and stay there for a month might be an efficient way to approach, reach or even surpass herd immunity.

    But of course, not all of them are destined to have mild cases. So no one would have enforced that sort of experiment on purpose!!

    Outside Seattle, most those protestors are subsequently circulating. What happens to cases will partly depend on whether the older more illness prone people self-isolate. (I bet many will.)

    Anyway: We’ll see.

    One could in theory determine how many cases were tied to protests by asking some questions but I think the establishment will be unwilling to look under that rock for fear of what they might find.

    I’m sure we’ll read all sorts of data sifting for years. I’m probably going to be dubious of all of it.

  52. Am I a little slow (don’t answer that) or is the silence = violence meme an apologia, in the sense of justification, for the riots and looting? I.e. the looting and riots aren’t really violence compared to being silent about white privilege, etc.

  53. I just checked, and as of June 12, TN is not including antibody test results in its daily update of new cases of COVID-19. They are, however, including probably cases according to the CDC criteria:

    ∙ Meets clinical criteria AND epidemiologic evidence with no confirmatory laboratory testing performed for COVID-19.
    ∙ Meets presumptive laboratory evidence AND either clinical criteria OR epidemiologic evidence.
    ∙ Meets vital records criteria with no confirmatory laboratory testing performed for COVID-19.

    So the increase in new cases in TN is real. Chattanooga and the surrounding area, part of which is in GA, is a major problem at the moment. TN still has a relatively low fatality rate, though.

  54. I think silence is violence means that if you are not outspoken in your anti-racism, it means you are a racist. The dreams of SJW struggle sessions …

  55. Combining anti-body and virus tests is insanity. At the very least provide the separated numbers.

  56. Tom Scharf,
    “…you are not outspoken in your anti-racism, it means you are a racist.”
    .
    But I AM outspoken in my anti-racism: I think it is utterly immoral to give anyone special privilege or for anyone to suffer special penalty because of their race. But somehow I doubt that would protect me from the SJ mob.

  57. I have scheduled a mandatory struggle session for you. Please report to your local social science department for re-education.

  58. “I see you are not doing as well as some of your peers in seeking out and eradicating racism. If you don’t do better, we will have to consider you a sympathizer.””

  59. I had another talk with my daughter that relates to this somewhat. She’d posted a third party account of somebody else complaining that having another female (white, of course) touch her hair without permission was violence. I won’t go into the story; suffice it to say that – perhaps it was too familiar or a presumption on the boundaries of friendship; a personal space faux pas, but alas. It was not violence.
    DeWitt, I think there’s something to your point above. When everything is violence, nothing is violence. This serves to elevate trivia and excuse real wrong simultaneously.

  60. mark bofill,

    When everything is violence, nothing is violence. This serves to elevate trivia and excuse real wrong simultaneously.

    To that point, the picture above the Babylon Bee article you linked is a young woman holding up a sign saying Whiteness is Violence.

  61. I like it.
    https://www.ajc.com/news/state–regional-govt–politics/democrat-who-endorsed-trump-wants-party-affiliation-added-proposed-hate-crimes-law/7StaAsDYjh0MHwNa7RobAO/
    Let me rephrase. I don’t much care for the idea of hate crime laws in the first place. But if we’re going to have them, is there a downside to adding political party affiliation protection to them? (real question – I’d like to hear why I shouldn’t think this is a good idea, if it’s actually not a good idea.)

  62. I’ve heard the argument that party is a choice, so shouldn’t receive protection, but then Ive also heard that gender is a choice as well so…

  63. mark bofill,
    “But if we’re going to have them, is there a downside to adding political party affiliation protection to them?”
    .
    You can’t keep people from hating. You can keep them from behaving illegally, or at least suffer the consequences if they do. I am sure that Nancy Pelosi doesn’t think she is a hater, but her politics and her actions betray that delusion. She absolutely loaths most of the people in ‘flyover country’… folks like you. Everything she does is designed to punish her political opponents. But so far at least, I don’t see that she has been acting illegally, just hatefully. Better to outlaw specific actions, not hatred.

  64. Steve,
    So your thinking is such a law would have little utility? That could be so. I wasn’t really thinking of the Pelosi’s though, but instead more along the lines of not so peaceful protesters. Antifa for example.
    [Edit: I agree with the sentiment that hate laws are sort of stupid that way in the first place. I don’t personally see why it matters if I get assaulted by a cold blooded hitman who’s strictly in it for the money vs somebody who deeply and personally hates me; what difference does it make? But given that we already have to suffer hate laws…]

  65. Thanks Lucia. I did consider commenting with a right wing example as well, but I couldn’t think of one beside the Proud Boys, and I wasn’t sure they were a good example.
    Boog sounds like a good example. I learned something today. Right wing Antifa, who knew.
    [Edit: Sound worse, honestly. Antifa doesn’t have a reputation for murder, just mayhem so far.]

  66. mark,
    I’d just learned about the boogaloo movement. I saw they wear hawaiian shirts. It may be the most trivial thing to be annoyed with about them, but I’m not sure I can forgive them for that. I mean… what are we supposed to wear to Jimmy Buffet concerts now?!

  67. mark,
    I had to google to find where the “I became unreasonable” came from.

    https://www.nbcnews.com/news/us-news/man-charged-deputy-ambush-scrawled-extremist-boogaloo-phrases-blood-n1230321

    “I became unreasonable” is a reference to a quote written by Marvin Heemeyer, an anti-government extremist who bulldozed 13 buildings in Granby, Colorado, in retribution for a zoning dispute. Heemeyer killed himself after the rampage, which occurred on June 4, 2004, almost 16 years to the day of Carrillo’s attack.

    https://en.wikipedia.org/wiki/Marvin_Heemeyer

  68. WOW.

    Over about eighteen months Heemeyer had secretly modified a Komatsu D355A bulldozer by adding layers of steel and concrete, intended to serve as armor.[1] He used this to demolish the Granby town hall, the former mayor’s house, and several other buildings.[1] Heemeyer’s rampage concluded with his suicide, after his bulldozer became trapped in the basement of a hardware store he had been in the process of destroying.

    Look at this thing!
    https://en.wikipedia.org/wiki/Marvin_Heemeyer#/media/File:Killdozer.jpg
    I live such a sheltered life. If I hadn’t checked I would’ve assumed it wasn’t real. In fact I’m going to do some more checking, I’m still having a hard time persuading myself this really happened.
    .
    [Edit: Yup. It really happened.]

  69. I was just reading accounts with images. I too would have assumed this was the plot of some sort of zany failed-crook comedy.

Comments are closed.