As of July 17 I regret to announce that the CDC has ceased publication of its state-by-state dataset of total covid cases. At the same time, I have seen a graph on CNN substantially duplicating my once-exclusive state-by-state acceleration-reduction statistic. I conclude that my work here is done. Thanks for everybody who tuned in over the past 16 months and learned something new.

I published these statistics daily from March 22, 2020 through September 4, 2020 and on about half of the days from 9/4/20 to 6/21/21. Originally nobody at all was tracking infectuousness rates by state, so I started this page to keep individual states honest about their efforts. At this point, I believe that various statisticians of repute have finally caught up with me.

As of June 27 I regret to report that the CDC's web page where I download my data, https://covid.cdc.gov/covid-data-tracker/#cases_casesper100klast7days , has not loaded on my web browser for several days in a row now. If CDC has incompetent web page development right now, there's not much that I can do. Sorry.

All sorts of health departments are taking their weekends off this year. I'm moving my counts to the past seven days of new cases, from the past five days, as of May 19, 2021. My percentage of new cases per day is based on this count divided by 7 and divided by the state's population. I'm adapting my "turning the corner statistic" to look at the new cases in the last 7 days divided by the new cases five days back from this 7 day average. My unique "turning the corner statistic" is still remarkably sensitive to new trends in the data, while it still avoids some of the dangers of outlier data.

As of June 21, the great U.S. immunization plunge is progressing nationally . As of 6/21/21 2 states have case numbers that are accelerating and 49 states (including DC) have reducing case numbers by my statistic. The 6/21/21 median is a 21% reduction between periods spaced 5 days apart. As of 6/21/21 the median per capita daily new infection rate, averaged over a 7 day period at state #26, is .0026%. The value of R, the reproduction rate, seems to be .8 in the states with low vaccination rates and .6 in states with high vaccination rates. R may rise as the delta variant becomes predominant.

As of 6/9/21 I'm dropping the Northern Mariana Islands from my list, as they might have had the virus go extinct on their islands. The Marianas can still pick up the virus from occasional tourists, but until then there's no story.

I expect that infection rates will continue to reduce state by state as open-the-window weather and outdoor weather moves north with the spring. They might possibly rise state by state as the hot states' air conditioning units come on with the summer - I remember that last year the infection rate in Southern states picked up as people closed their windows and turned on their air conditioning. It's also possible that the country's vaccination rate, HEPA filter use and other measures seem to have dried up the virus's overall ability to keep moving through communities.

As numbers approach zero, new infection from reservoirs outside of completely cleared up regions will occasionally spike up a state's numbers.

One year ago South Korea reached a point where almost all of its cases were airport-related. They had one outbreak related to their bar scene. This summer, the U.S. may have large areas of the country that are covid-free, but with occasional outbreaks that are related to people bringing the covid in from other countries where it is endemic because of near-zero vaccination rates. The outbreaks may be more serious in the hot South where people live in air conditioned buildings in the summer, and the same may be true in the North next fall and winter.

Bad data reports:

As of April 28 New Jersey has had a strongly negative number of new cases recently, in the thousands, which is somewhat similar to negative numbers of votes in an election, in that the number is untrustworthy. As of April 19, 2021 Missouri new case numbers went wildly negative, about 10,000 below zero. Much as with negative numbers of votes counted in an election, this made Missouri case numbers unusable for two weeks. As of 3/9/21 Missouri found 80,000 misplaced covid cases. Missouri wasn't the first state to have sudden violent lurches in their covid case numbers. This might simply be a re-reckoning of what constitutes a covid positive case, but given a number of other states' recent history I also have to give consideration to the possibility that 80,000 cases were buried for embarassing political reasons. Texas had an outrageous spike in cases as of 12/19/20. My instinct is that the state's numbers were being manipulated. Oklahoma's case reporting system seems to be functioning again as of 12/20. It was consistently failing between 11/25/20 and 12/19/20, possibly for reasons of political shame or possibly it was an unfortunate side effect of the pandemic hitting their state. New York State numbers bounced around like crazy in June of 2020.

As always, New Zealand is happy to be mostly done with this pandemic. As of late January they had their first new case in months, and as of February 13 they had four more cases, all in one family. Their entire nation succeeded long before the vaccines were available.

The recent wave was partly caused by a politically driven need to ignore medical integrity.

I've seen U.S. medical reports that Vitamin D levels in people might be related to lower transmission rates and less severe cases. The Boston Herald reported on 9/21/20 that a Boston University doctor recommends giving Vitamin D a try. A high percentage of covid-19 positive blood samples also showed vitamin D deficiency. Vitamin D is safe enough to be on the shelves right now without a prescription, and it might give an enormous positive impact per dollar spent. So why not? Scotland is reportedly trying this, on the basis of it being better than trying nothing, it does no harm and it's relatively dirt cheap. New Zealand has long had a policy of protecting its citizens from Vitamin D deficiency. Read the medical literature. Read the freaking medical literature.

We eventually need our workplaces to become at least 99% less infectuous to minimize each state's total casualties, and 100% less infectuous would be nice.

As of 8/9/20 I raised my forecast to 400,000 extra U.S. deaths by the end of the year because of the pandemic. As of 11/27 I suddenly saw the numbers going down and I had to back off from 400,000 casualties on 11/27. I had to drop the 2020 number to 350,000 deaths for New Years Eve. Actual number on January 2, 2021: 346,925. Congratulations to the Washington State and to the Johns Hopkins forecasters for being a valiant competition. A bit of advice to the competition: watch U.S. politics if you want a more accurate forecast next time. I'm giving up on forecasting deaths for 2021 because the vagarities of vaccination rollout will make such a prediction so hard. My sense is that from February through spring, the new death rate will nose down and gradually head toward zero. The pandemic will have fewer and fewer vulnerable people that it can use to propagate itself, and it will slowly die back. There will be huge reservoirs of the disease in other nations for at least one year but the covid could have trouble re-establishing itself in the United States and in Europe next fall.

I can report two things that the CDC doesn't want to touch. First, is any particular state accelerating into another pandemic bout? Second, is the CDC data a bit funny today? In the long run, some people are wondering if the Trump administration will try to play with the data. My November 2 headline was that our national pandemic outlook appeared to be running away from the brink of disaster as fast as it possibly can, with an asterisk for suspicious data. 50 states (including DC) were improving on 11/2 and the other 1 state had an accelerating number of new cases. As of November 7, 43 states were accelerating and 8 states were improving. That's an unusual turnaround. It's quite possible that the data has been dangerously massaged by the CDC to coincide with Election Day.

The state median was .071% on 12/12/20. It was at .052% on 11/15/20. It was .012% on 9/26/20, so the pandemic has accelerated recently. On 7/26/20 it was .0136%. The median at #26 on May 1, 2020 was .0058%.

From a peak of 40 states making progress around 4/15/20, the pandemic reduced until the U.S. hit a low of 14 states on 4/27/20. The U.S. hit a second new case peak which was recorded around 5/1/20 and started back down a second time. The U.S. may have hit a second bottom around 5/17/20 except little further rise occurred after the bottom. As of 5/27/20 the nation went past a third micro-peak in new case infection percentages. As of 6/2/20 the overall numbers were starting to accelerate again. The nation had such a flat and short peak around 6/10/20 that it's almost better to say that no peak occurred at all, that the pandemic was flat and was accelerating weekly and monthly through 7/29/20. The national pandemic was pretty flat through September and accelerating in October, November and December. I suspect that as of December 12 we hit a peak. The holiday transmission bulge put a stop to the drop between December 12 and December 25, 2020. As of January of 2021 we were heading up toward a peak. That ultimate peak was probably reached on January 13. A rather dramatic covid-19 plunge bottomed out around March 15. A small acceleration took place up to perhaps April 4. The plunge resumed around April 18.

How Did Your State Do?

Without further ado, today's statistics, sorted first by state name:

Top Immediate Threats

Turning the corner statistic

Methodology

My goal is to provide various state governments, where the remaining political power resides, with statistics so that they can focus on replicating whatever the successful states are doing and not care that much about what the laggards do.

My threat index is a logarithm, base 10, like the Richter Scale for earthquakes, of the per capita new case percentage (the last five days only) multiplied by 10 million people. So, in a state with 10 million people with exactly one new case this week the threat is zero. With 10 new cases the threat is 1. With 100 new cases the threat is 2. With 1000 new cases the threat is 3. With one million people having the virus out of ten million, the threat would be up to 6.

Second, I want a statistic that can help us see when we have turned the corner. It's a ratio of new cases to older cases. Too many media outlets are giving us total cases since the beginning.

My exclusive turning-the-corner statistic subtracts (the last five days of new cases divided by the previous five days of new cases) from 100%. If the result is above zero percent, that's good because the number of new cases is exponentially going down with time. If the result is below zero then the number of new cases is still going up. Around -100% the number of new cases will have doubled in a five day period. In late March I regularly saw triplings and quadruplings in new case numbers in a five day period, but I'm only rarely seeing that lately in individual state breakouts.

My goal is to exclude the oldest and probably well-quarantined known cases. My goal isn't to measure any death rates. Rather, I try to measure the exponential growth rate or exponential shrinkage rate in transmission of the virus, in order to measure how well or how poorly testing, quarantines and social distanceing are taking hold today in individual states. Exponential growth and exponential shrinkage will ultimately be the critical factor between seeing millions of fatalities versus tens of thousands of fatalities nationally.

As of April 6, 2020 my turning-the-corner statistic was modified. On March 22 I was starting from scratch and had little or no older data to work with, so I compared all older cases to the last five days of information. During the coronavirus's rapid growth phase the difference between the last five days of data and the data back to the beginning wasn't overly significant. At this point the pandemic is really turning the corner and so accurate information is needed.

I estimate based on media reports that new cases are particularly virulent for their first five days of infection. If new cases over the past five days are triple the total number of cases 5 to 10 days old, then the "turn the corner" rate is -200%. (sorry for you.) If new cases over the past five days equals older cases then the "turn the corner" rate is down to 0% (don't let this go on forever). In other words, the virus is roughly breaking even, where new infections equal old cases that are no longer infectuous. If new cases are zero then the "turn the corner" rate is 100% and your state is basically cleaned out like China, thank you. I'm going to leave a minimum microscopic number, .01 case, in place for old cases to avoid a division by zero error. The goal here is to recognize certain states that are actually making long-term progress in winning the battle and other states that don't in practice seem to be trying that hard.

I don't want my "turn the corner" statistic to jump around too much from day to day. For this reason I take the average of any state's total cases reported from four days ago, from five days ago and from six days ago to get the state's average total cases from five days ago. I do the same for the state's average total cases from ten days ago. This minimizes unnecessary bumps in current statistics when a large number of reported cases falls over a certain five day old or ten day old precipice.

Population per square mile is noted to be fair to the more crowded states that might have a harder job controlling the virus. In fact, crowded large cities are a problem.

I don't agree with the CDC's implicit assertion in its state tables that New York City is a state. I'll consider their claim that the District of Columbia is already the 51st state. Early on the CDC also placed Puerto Rico with the states, while the Virgin Islands were one of the colonies, placed at the bottom of the list.

I'm hearing about R, the ratio of new cases to older cases. When R is above 1 the pandemic is accelerating. When R is exactly 1, that's still a problem. Picture a man who owes $10,000 to his bookie at 10% weekly interest, and he decides to pay off the debt at $1000 per week. The man's R ratio is 1, $1000 out every week matches $1000 in weekly interest so the debt never gets worse. The rest of us would say, what a lowlife! He needs to pay the debt down toward zero at some point.

I've heard of two relatively affordable but unused ways to lower the infection rate. First, plastic face shields and goggles seem to slightly inhibit transmission and they're affordable in bulk. Anything that inhibits transmission and is affordable should be examined as a tool and as a partial solution. Why shouldn't many essential workers be wearing face shields? Also, why shouldn't cental air HVAC units be using hospital-grade air filters?

Second, my friend at ConnieEash.com points to a good deal of solid evidence that increasing Vitamin D intake is cheap, does no harm for most people and has at least some effectiveness.

An Inventor's View of the Future

We may discover that human society has grown technologically wealthy in recent decades.  We can do many things that we couldn’t do.

I have been describing the exponential growth of the coronavirus and its exponential decay in nuclear reaction terms.  Remove the control rods in a nuclear reactor and the reaction grows exponentially until the top of the reactor pops off, as happened in Chernobyl and Fukushima.  Insert enough control rods and the reaction dampens down to almost zero, but not quite.  When the reaction has dampened down to about zero a few control rods can be raised and the reaction still won’t start growing, but pull too many control rods and the reaction starts up again.

 So, here’s the future as of April, 2020:  First, a large number of states will drive the pandemic down by about 99%.  We won’t be able to get absolute zero cases for a year because new cases will arrive at airports every day, that’s what South Korea discovered.  Certain allegedly pro-business states shall fail until they realize that they're failing and that failure is actually bad for business, and then most states will get back on the wagon after their benders. True pro-business states will get down to near zero and then will monitor their truck stops, airports and new immigrants to their states for any new breakouts.

Central air conditioning systems are being fingered for their ability to spread aerosolized viruses throughout the building. One partial solution is better filtering of the air, probably HEPA filtering or electrostatic filtering. I've heard a recommendation on NPR's "The Takeaway" that all central HVAC systems should upgrade to hospital-grade MERV 13 air filters in order to stop aerosolized virus particles from circulating throughout a building. Bringing in fresh air also helps. The direction of air current drift within any room is also important.

We need essential workers to wear cheap clear plastic face shields or goggles. They're more effective at protecting the worker's eyes than nothing at all. The rest of us might benefit also.

In perhaps one month from now I'd expect that new cases in the United States will be largely limited to institutional new cases in nursing homes, in prisons and in homeless shelters, plus overseas travelers bringing the virus in. Most businesses will reopen, although you may be handed a paper customer courtesy mask and plastic gloves as you enter the store. Business offices will be somewhat more compartmentalized but will be running. Large crowd events will still be problematic. Smaller schools and smaller class sizes will be better.

I usually get my raw data from https://covid.cdc.gov/covid-data-tracker/?CDC_AA_refVal=https%3A%2F%2Fwww.cdc.gov%2Fcoronavirus%2F2019-ncov%2Fcases-updates%2Fcases-in-us.html#cases_casesper100klast7days They tend to publish daily around 5:00 p.m., although they optimistically promise 4:00. For 5/8/20 they published quite late, after I went to bed and presumably before midnight.

Coronavirus reporting has quickly evolved from when I found my raw data at https://www.livescience.com/coronavirus-updates-united-states.html or from https://www.statista.com/statistics/1102807/coronavirus-covid19-cases-number-us-americans-by-state/. It turns out that these three different groups count differently, and then there are sometimes transcription errors in each reporter's raw data. That's why averaging over five days is important. I consider three day averages to be a bit sloppy. A five day average is a more likely representative measure of any particular state's probable community infection rate at any given time, although much can also be read into the current "turning the corner" trend.

The creator of this page, Paul Klinkman, has been a prolific solar and climate change inventor for over 20 years. This statistics page just needed doing daily since March 22 because this reporting tends to save many lives. That's also why I invent. Even a small climate change improvement might save millions of human lives. You can look at a wave of seriously new and great solar inventions, quite possibly the world's best collection of curated climate change inventions, seminal climate change books such as "Drawdown" notwithstanding, at klinkmansolar.com. You can reach the author at info (at) klinkmansolar . c o m

google-site-verification: google01103d457b16554c.html