- Why no report on recoveries?
- Herd immunity
- What’s the prevalence index all about?
- Fatality numbers for my city / zip code
- What localities do you cover and why?
- Why don’t you keep your politics out of it?
- When will cases rise after ____ event?
- Didn’t the BLM protests cause the spike in cases?
- Who pays for this site? How do you make your money?
Why don’t you report on the number of people recovered? Isn’t that a more optimistic metric?
Yes, it is more optimistic, but it’s not a standardized metric across localities. Also, there is no reliable way of measuring it. If a person is hospitalized, if they are released, that should increase the number by one. But if they have symptoms, get a test, and stay at home, no one knows whether they are actively ill or have recovered.
Why not just gut this out until we all have herd immunity?
The people who advocate this know little about what it really means and how herd immunity is achieved. Basically, herd immunity is a function of R0, the initial rate of contagion for the virus in a naive population, i.e., a population that has never been exposed to the virus and for which there is no vaccine. The higher R0, the higher the threshold to achieve herd immunity. For example, measles has an R0 of somewhere between 12 and 18. That means that herd immunity is only achieved when around 80% to 90% of the population has antibodies. The R0 of COVID-19 has been estimated by virologists to be around 2.0 to 3.0. That requires that at least 60% of the population have antibodies. In terms of the prevalence index, this would mean a prevalence index of around 2. Right now, in Imperial County, one of the hardest hit places in the US, the prevalence index is 30, or 15 times more than the number required for herd immunity. Finally, we just don’t know yet how long these antibodies stick around. It COVID like measles, where one vaccine is good for life? Or is it like the common cold, where antibodies last about 3 months? The virus doesn’t seem to mutate that much, so there’s hope that a vaccine will confer some type of lasting immunity, but we have no idea how long yet.
What is the Prevalence Index all about?
Initially, I called this a “People per Case” chart. If the number for a location was 100, that would mean statistically that out of every 100 people, there was one case. However, some people pointed out that this could be interpreted incorrectly, because the one case could represent an actively sick person, a recovered person, or a death. The number really was intended to give a sense of how prevalent coronavirus was in a locality based on the population of the locality divided into the cumulative case count.
For the sake of clarity, I renamed this to a “Prevalence Index.” Generally, human networks run to a maximum of 500 people. So if the prevalence index on this chart is below that number, it indicates that everyone in that locality (statistically, of course) has a personal experience with coronavirus – directly, or someone in the family, a business acquaintance, etc. Beyond that, and certainly beyond 1,000, the experience with coronavirus is somewhat abstract – gained primarily through reading and watching/listening to the news.
Where are fatality numbers for my city and/or zip code?
Counties in CA do not report fatalities by city or zip code, probably to avoid any chance of personal identification. That would be a violation of the HIPAA law.
What localities do you report on regularly and why?
For the U.S. as a whole, I report on cases, fatalities, and hospitalizations. I’m able to download this data from covidtracking.com, so no manual entry is necessary.
States: Arizona, California, Florida, Georgia, Nevada, Oklahoma, and Texas. AZ and NV because they’re right next to us. Florida, Georgia, and Texas because they were three of the states to open up before anyone else. OK because that’s where Trump held his first rally.
Counties: Imperial, Kern, Los Angeles, Orange, Riverside, San Bernardino, San Diego, San Luis Obispo, Santa Barbara, and Ventura. These are the ten counties that make up the entirety of Southern California, and there’s a lot if intra-county travel.
Cities: I’m only able to get downloadable, historical data from cities in San Diego County. I picked my own city, Encinitas, as well as the 18 other cities that had been the most hard hit at the beginning of July: Carlsbad, Chula Vista, El Cajon, Escondido, Fallbrook, Imperial Beach, La Mesa, Lakeside, Lemon Grove, National City, Oceanside, Poway, Ramona, San Diego City, San Marcos, Santee, Spring Valley, and Vista.
Zip Codes: I picked the 33 zip codes that were the most hard hit within San Diego County at the beginning of July.
Why don’t you keep your politics to yourself and just present the data?
I believe the response to a pandemic is the very definition of politics. If my politics irritate you, then stop reading my commentary and just look at the data.
I often hear the argument, “when you talk about your politics, I can’t take your data seriously.” There are a couple of points worth addressing here.
First, it’s not “my” data. The data comes from the sources listed on this site. It’s public data – not mine, and not yours.
Second, this statement implies that there are biased scientists and unbiased ones. I don’t buy that. For example, there is a tremendous amount of bias against women in science, ranging from the kind of research that’s done, to the subjects of the research, to who rises in the ranks. There is also a serious problem with racism in science.
I admit my bias. I own it. You can look at my data however you want, but you can’t accuse me of secretly using it to push a political agenda. For me, the data pushes a political agenda all by itself. But I want to explain exactly how it does that. You can ignore my commentary or embrace it – your choice.
For every person who has raised this argument with me, there have been a lot more people who are appreciative that I’m completely open about it.
There was a huge gathering of people without masks and close to each other. When will we see case counts rise?
We don’t know. Any group may consist of all people who don’t shed a lot of virus. Or they may have one “super-spreader” who gets dozens of people sick. There is no evenly grouped transmission of the virus. On top of that, there’s a whole lot that we still don’t know about it.
Aren’t the protests responsible for the big increase in cases?
Most of the evidence says no. There are tons of articles and research papers on this, and for the most part, this idea is generally pushed by people who want to discredit BLM. A group of Economists and health researchers published a paper on this exploring it in depth. Their conclusion: “Furthermore, we find no evidence that urban protests reignited COVID-19 case growth during the more than three weeks following protest onset. We conclude that predictions of broad negative public health consequences of Black Lives Matter protests were far too narrowly conceived.”
There are many, many articles backing this up. A detailed explanation of why this is so is provided in a YouTube video by Dr. Kim Prather, professor of chemistry and biochemistry at UCSD, and an expert on aerosols. The bottom line: when you’re outside, the virus dissipates extremely quickly, but inside, it can hang around for hours. This is why most case tracking, when it can be done, finds the cause of community outbreaks are inside social events at bars, restaurants, religious gatherings, parties, etc.
Who pays for this site? How do you make your money?
I pay for the site. I don’t accept any money or goods on this site or from people on social media platforms like Reddit. I wanted to do something positive when the pandemic forced us into lockdown in March, so I started posting about the data, since so many people were expressing opinions not based on facts. I’m retired and have enough for my needs, so I don’t need to make money from this site and I have no intention of commercializing it.
If you would like to express your appreciation with a monetary donation, please give it to the Equal Justice Initiative.