New research studied on 300 largest cities did not find Protests to be cause of new spikes
Florida, Texas, Nevada, Arizona laxed their laws, but also never gave mask guidelines. Now they are starting have mandatory mask guidelines.
Just read the "study". It's bullshit.
They sought to determined whether or not infections rose by studying the "social distancing impact" in the cities using phone tracking data. They use this to assert that overall social distancing increased during the protests. Nevermind that only in cities where protests persisted for over 3 consecutive days did you see people stay at home at the most elevated rate, and the increased rate of people in these cities staying home all day rose by a paltry
0.73 percent of the population (a ~2.0% increase relative to the mean). They fabricate a ridiculous pair of equations to attempt to contort this into proof that social distancing increased while ignoring the total violations of the 6ft rule and intercommunication between so many different individuals inherent to a protest versus their broad descriptions of daily commute (i.e. stayed at home, mostly stayed at home, went out):
SDcst = γ0 + ∑ γ 8 =0 ct * Tc + Xct β1 + Zst β2 + αc + τt + εsct (3a)
COVID Growthcst = γ0 + ∑ γ 21 =0 ct * Tc + Xct β1 + Zst β2 + αc + τt + εsct (3b)
Furthermore, not only is it drawing conclusions about social distancing from elaborate simulations using phone tracking sample data, but they can only hypothesize that social distancing may have decreased because those people staying home did so out of fear for their personal safety, and because businesses closed in anticipation of violence or destruction by the mobs. What a great positive for society.
But let's get down to brass tacks: new infections in those cities. These researchers assert they used CDC data via the Kaiser Foundation, and they offer maps showing the key cities of interests, but they simultaneously assert that daily new confirmed cases in these counties has
declined from May 15 to June 20th.
The average day-over-day COVID-19 case growth over the sample period was 215.2 cases (7.4 per 100,000 population). Our key variable measures the difference in the natural log in the cumulative number of COVID-19 cases on a given day and the natural log of the number of cases on the previous day. The mean of this variable is 0.019.
My fucking ass.
Here's the daily new cases log graph for the USA as a whole since May 26th when the protests started. Of course, as the researchers acknowledge, due to the average incubation period and onset of symptoms, resulting in more hospital visits and confirmation of infection, something Fauci pointed out in the past, it takes two weeks following a new measure or social behavior before you know what that measure wrought. They also acknowledged that mass protests didn't broadly intensity until May 28th, and became most populated after June 1st. That means you should be looking at the yellow line starting around June 11th below:
And no, that isn't because rural counties are driving the increases (though cases are also increasing in those places). They are still way, way, way below the per capita averages. Take a look at Los Angeles daily new cases where protests were most intense in my state-- paying close attention again to June:
Here's Hennepin County, the home of Minneapolis, where the protests originated. They didn't start to reopen until May 15th (11 days before protests began). So if you choose to look upon the flattening slope of new cases favorably, understand there's no case to be made against reopening, either, because the upslope preceded that by three weeks. If accounting for lag reopening corresponds to the onset of the daily downslope log change as the total cases rise:
Here's the state of Georgia where protests have been among the most intense (where Arbery was killed):
The protests created the surge. The most conservative estimate I've seen by an actual modeler was that they were increasing cases by 3%-6%. That's the lowest figure I find credible, but given the overall spike in new cases, nationwide, concentrated per capita in the cities, it's obviously much, much higher.
