Last year, lamenting the fact that in today’s world there are no DATA (i.e., Science), rather only made-up numbers (i.e., Narrative), I counted my Day Lilies and fit the number of day lilies to a Gaussian, summarized in my Post, More on the Gaussian. This year (with the help of my wonderful neighbors, who also “pinched” my garlic plants while I was out of state), I counted Purple Iris Flowers and fit the number of Irises each day to a Gaussian as shown in Figure 1.
Figure 1. Number of Purple Iris Flowers vs. Day Number with a fit to a Gaussian. Day 0 is 16 May 2023.
The Gaussian fits my Purple Iris DATA with an R-Squared value of 91%; last year’s Gaussian fit to my Day Lily DATA had an R-Squared value of 93%. These are a pretty damn good fits. But I’ve seen better. Much better. (Hey, I’ve been around a bit and wasn’t born yesterday). See my previous Post, More on the Gaussian, to see a Gaussian fit to some real world DATA (Vostok Ice Core DATA) with an R-Squared value of over 98%.
Now, a really ambiguous number is, “How many people, per million, have died from CoVid?” I thought I would look at NOT-DATA, but rather Official-Numbers. (From the very beginning of the CoVid-1984 Fear-Pandemic, it was clear these “Official Numbers” were made-up bullshit numbers, not real data, as I briefly summarize in Footnote 11. Now, if we look at the death-rate of CoVid for all the 220 or so countries in the world, I’d expect to see a Gaussian distribution: an average number, but with some variation plus or minus, and probably a few outliers. For example (sarcasm alert), you’d expect the U.S. to have a relatively low number due to the U.S. not only having Freedom and Democracy, but the world’s very best (and most expensive) healthcare system, our mRNA Injections, all our NPI’s (non-pharmaceutical Interventions) and our spending almost 5 TRILLION dollars on the “pandemic”. (Spoiler alert: the U.S. had the most “CoVid Deaths” of any country, and 95% of the countries in the world had lower “CoVid Death Rates” [see my post, USA #1 !]).
OK, well maybe the U.S. is an outlier on the other side. Regardless, with all the variations and uncontrolled variables, I still expect to see a Gaussian due to the Central Limit Theorem2. What we see is NOT a Gaussian, as shown in Figure 2 below.
Figure 2. Histogram of the number of countries vs. death rate (blue), and integral of this curve (orange). The U.S. is in the bin denoted by the large Black Symbols. Half the countries have death rates below the bin marked with the large Purple Symbol. The large Red Symbol, in the first bin, or lowest death rate, is also the most probable death rate. The Integral (Orange curve) is merely the Percentage of all the countries that have a lower death rate.
Figure 2 is:
NOT a Gaussian! It is rather fit by a curve where the probability of a given death rate decreases as 1/(death rate) with an R-Squared of 83%. Of course, we knew from the beginning we were just playing with “numbers”, not DATA.
The most probable death rate is the lowest death rate bin (Red Symbol; 63 countries, almost 30% of the countries, had this death rate); half the countries had a “death rate” of less than 800/million (Purple Symbol), about a quarter of the U.S. “death rate”.
The U.S., had a higher “death rate” than 95% of the countries in the world.
I prefer playing with real DATA, but sometimes Physicists can have Phun playing with made-up numbers (as long as we don’t take this Phun too seriously).
DATA come from verifiable MEASUREMENTS; numbers are made-up, or come from made-up MODELS. From the beginning it was obvious that the “Official Numbers” were not DATA, but just bullshit made-up numbers, e.g.:
Defining a CoVid “Death” as anyone who dies within 30 days for any reason, 94% of them with an average of 5 other comorbidities, after receiving a positive test from a fraudulent PCR test.
It takes the CDC several months to determine how many people have died in a given week for any reason. But somehow there is an “Official Number” of “CoVid Deaths” at midnight everyday for every country in the world and every state in the U.S., which never gets updated or changed. Am I supposed to believe that the Gates-funded “World-O-Meter” used Elon Musk’s Neural Link and tapped into the Mind of GOD !?
Central limit theorem states that when we add large number of independent random variables, irrespective of the original distribution of these variables, their normalized sum tends towards a Gaussian distribution.
I miss working with statistics.