Not Your Demographic

Assertions so counter intuitive and nonsensical - they must be true!

Cogent insights by Benjamin Scherrey

Prev Next


MEDIA PROMOTES PANIC - NUMBERS IGNORE PRESS AND GO DOWN ANYWAY. Not surprisingly, just as states decide to start opening up the press and our "health organizations" start doubling down and revising their models UP UP UP! Well that's interesting except May looks to be going down down down. Today we hit records lows on all four core stats since Mar 19th when our rates of growth (ROG) peaked and since April 1st when absolute numbers really started rising. ROG of incidents, absolute incidents, ROG of deaths, and absolute deaths are their lowest yet since things really took off. Of course those are Sunday numbers and will be adjusted somewhat upwards the next two days but these Sunday numbers are also way below all the prior Sunday numbers on all counts. We anticipate the upward adjustments will be lower than the prior weeks adjustments as well. So even with all the caveats we are aware of - this looks like a banner day. Also - today is the first day since April 1st that we went below 1000 deaths!

Our new total for incidents is 1,173,257 (+21,251), an increase of 1.84% over the previous day (-0.49% from yesterday). Deaths have totaled 62,806 (+938) which is an increase of 1.52% (-0.39%) over the previous day. All new records! Since the peak in Mar 19, rates of growth for incidents has fallen by 49.79% and for deaths has fallen by 41.34%. There were 231,791 tests completed with a 9.17% positive rate - another two week record low. Being weekend numbers we'll need to wait til tomorrow or Wednesday to adjust the actual trend.

In more interesting news, Dr. Levitt is finally back in the news! If you don't remember him, he's the guy who was making really good observations and the ONLY ONE who was consistently making predictions based on the data (discover based) rather than completely flawed exponential models (prediction based) built on questionable assumptions. This article from March 23rd, just days after the peak ROG hit and the decline first became evident. I reported the same on that day. It was Dr. Levitt's clear and reasoned approach that was clearly non-political and focused on the data that made me start tracking Wuhan-virus stats because it was apparent just how bad the press was covering this in order to support political agendas. A Tale of Two Pandemics was my first post and how I introduced Dr. Levitt as contrasted against the "expert" that CBS doesn't even bother to identify.

In yesterday's article for the Blaze, Dr. Levitt explains that '...the damage done by lock down will exceed any saving of lives by a huge factor'. Now where else have you heard this? Well, here of course! And we knew this back in March!

Let's see how well we've done so far compared to the FUD merchants of our press and in our government:

March 16th : Acknowledging that data collection is still shaky, am hopeful that we may have peaked ROG on Mar 13 and, if so, that would limit us to 192 deaths by Mar 21. On Mar 21 there were 272 deaths. Clearly we had not peaked yet. I make the case that regardless of this outcome, outside of 10 states with potential hot spots "unless you are in a high-risk area or would be a high-risk patient due to a compromised immune system due to age or illness. You do not need to super isolate yourself. Just keep your personal space and hygiene procedures and you'll be fine."

March 20th : Over several prior days I talk about the % death rate approaching under 1%. This should happen within 4 days. I observe two caveats, however. Inadequate testing capacity growth would skew the numbers back upwards or people living longer in America (as opposed to Italy) might push that out. The former is what happened as I later identify that our new incidents is entirely bound by test numbers and no longer reflects actual incidents. I also introduce my projection that the moving averages for ROG will drop down in April and the pandemic will effectively end in June. This aligns with time frames seen in countries ahead of us.

March 20th again : I identify that the peak ROG for USA likely occurred between March 18th & 23rd (it actually happened on the 19th). I also make a more subjective prediction that things will " start loosening up by May and we should be in the clear for most of America by June." Remains to be seen but looking fairly accurate so far.

March 22nd : Based on the numbers thus far I project almost 500k incidents and nearly 3000 deaths by April 1. Instead we got 210,000 incidents and 4,700 deaths. Later discover the numbers were off because of being artificially bounded by lack of testing capacity. Not a coincidence that both are off by about half each direction. I also predict really bad news for Thailand due to some horrible policies. Thailand lucks out (as does most of SEASIA) likely due to BCG vaccine saving them from any effect at all.

March 31st : The government updates projections of 100k - 240k deaths over the next two weeks - with the lock down! Without they project in the millions. I describe how I can see where they got their numbers but that the presumptions the predictive model carries aren't supported by evidence and explain what must happen for this to occur - which is highly unlikely. I was correct. By April 14th there were only 25,668 total deaths. Government was off by an order of magnitude. Still have their jobs.

April 10th : I reiterate a prediction I had made (based on ROG peaking on Mar 19) that the total death toll for this pandemic would be between 20,000 and 100,000. Pretty wide range but still a LOT better numbers than what the government and media reported. Those are numbers for the earlier predicted end of June.

April 11th : I note that the ROG for deaths is going to hit under 10% the next day and decline going forward. Completely correct. I also state that should that occur, with ROG for incidents we could max out at 20,000 total deaths. That, unfortunately, didn't happen because I did not yet identify that new cases were being severely under reported due to lack of testing capacity. Testing capacity limits is something I complain about almost every post which is why I always argued that projections MUST be primarily based on death rates. In that it is quite clear I was correct.

April 13th : Due to the decline in Death ROG, I project deaths should stay under 34,000 for April. Instead we got 57,266. Because we were under counting new incidents due to lack of testing and the CDC revised their policy for reporting Wuhan-virus deaths, ROG for deaths spiked up for another week before it began to decline again. I have strong confidence that, if not for the CDC's inflation ok, this projection would be accurate. In fact when you look at the impact on all deaths it's pretty clear my numbers were more accurate than the CDC at that point. But this puts me off of making projections for a while because I know the data is wrong.

May 1st : Despite a month's worth of heavily influenced data, I feel able to isolate most of the garbage from the real trend re-iterate my prediction that May will be "a considerably mild month" compared to April. Government models all say upwards. We'll see.

How did I do? Well when the data wasn't manipulated I'd say quite well. In all cases, far better than the government models. So where do I stand today?

My projections from March 20th remain : The pandemic falls in May and effectively ends in June. It fit the data in March, before our data started getting messed with due to actual incidents overwhelming our test capacity and the CDC encouraging states to produce fraudulent death reports, and it fits the data now in spite of all that government inflation. We definitely should NOT be shut down.

It's nice to see Dr. Levitt come to the same conclusion. The political solutions are doing far more damage than the impact of the virus will ever have. Please watch the whole video.

Prev Next