Numbers Behaving Badly

I've been looking for good data on COVID for quite some time, and the more I look, the harder it is to find. For instance, I went looking for a rolling 7-day average of COVID deaths in the US nationally, and what I find in CDC data is at best confusing. For instance, the CDC appears to combine "deaths due to pneumonia, influenza and COVID-19 (PIC)" in its reports -- but naturally, that's not what I'm looking for. I want to know what the numbers are for COVID specifically. and I'm not sure if this is ever broken out.

A site called the COVID Tracking Project has charts that provide 7-day rollling averages for tests, cases, hospitalizations, and deaths. If I google "US daily COVID deaths", I get a google site that lists one total (3611 for December 16), while the COVID Tracking project site gives a different total (3400 for December 16). The last 7-day average on the COVID Tracking Project site was 2400 on December 15.

The last rolling 7-day average case total I could find on the COVID Tracking Project site was 209,307 for December 11, But dividing the December 15 rolling average deaths of 2400 by the December 11 rolling average cases gives a US case mortality rate of 1.1%, which is consistent with various calculations I've made from different sources over the past several days. This is down by about half from the 2% case mortality rate several months ago and still generally reported. And if we take the latest 5-day rolling death average of 2400 and multiply it by the remaining 14 days in December, we get 33,600 deaths remaining for the month, far below the 250,000 Joe Biden predicted the country would have by January.

While figures like Gov Newsom continue to insist their predictions are based on "models", nobody seems to give out hard numbers since the Imperial College model predicted 2 to 4 million US deaths. Instead, the closest we've had is Biden's "dark winter" 250,000 by January, while Dr Fauci is simply telling people not to celebrate the holidays -- but the fog of numbers prevents any real ability to calculate whether the holidays have led to "surges" in statisticcs or not. For instnace,

Data from the Transportation Security Administration (TSA) days prior to the holiday showed more than three million people traveled through airports despite guidance suggesting otherwise.

However, nearly three weeks later, there so far appears to be no post-holiday surge in numerous cities and states.

“We haven’t seen something significant to talk about now,” Illinois Health Director Dr. Ngozi Ezike said, according to the Chicago Tribune. “We’ll see for sure in this coming week … We’ll keep our fingers crossed that maybe we’re not going to see a big bump.”

Nevertheless Dr Fauci is warning us against Christmas:

". . . But that’s just one of the things you’re going to have to accept as we go through this unprecedented challenging time."

Fauci, noting that many Americans ignored health guidelines over Thanksgiving, warned that Christmas cannot be "business as usual."

The problem is that the data to support his position is hard to find and unreliable, and people's experience on the ground takes away from official credibility. For instance, thousands of people attend the non-conforming California megachurches each week without masks or social distance, but almost none even test positive, much less get sick. Individual states differ wildly in their COVID regimes, but there's little difference in their actual outcomes, with the states that have the strictest regimes experiencing the worst statistics.

This is one part of corporate media's ongoing failure.