Actually most of the gainers and the reductions, either way, are in a range of 0% to 10% and a few reductions are greater than more than a 10% reduction.
But the real outliers in my opinion are Minnesota, DC, Massachusetts, NY, NJ and New York City, which far exceeded the others, percentage wise in either direction.
When you compare the totals from all the states, the number of total deaths for 2020 is not greater than the total deaths for 2019 plus the number of reported Wuhan Virus deaths. The 2020 total is less than one would expect with all the 2020 “additional” Wuhan Virus deaths. Therefor, clearly and for reasons that could be looked into, many kinds of other deaths for 2019 must have fewer of them reported for 2020.
What about heart attacks, liver failures and other chronic things as cause of deaths, but not reported as cause of death just because the patient (or the body) tested positive for the Wuhan Virus, whether suffering due it or not at the time of death. But, even that being the case in some instances, there are not enough additional 2020 deaths, if we are to assume the Wuhan Virus deaths are all “additional” deaths that would not have otherwise occurred, statistically.
Maybe there is also just far too much inconsistency in reporting deaths.
I assume you meant Michigan, not Minnesota?
Fewer auto accidents due to vastly reduced miles driven per person probably contributes a good bit. A quick check of the last three years reveals about 3,000 deaths per month from car accidents. Cut that in half and you get 3,000 fewer car deaths over two months. Flu deaths vary a lot from year to year, but that could easily be a couple thousand more difference. Unintentional injury normally takes 170,000 lives a year in the US. With people stuck at home, that may have dropped a bit as well.
I don’t think we’ll have good data about exact causes of death for a year or two just because of how the CDC and others collect and review all the stats. But it makes sense that if you lock everyone in a padded cell for a time, fewer will die from “preventable” deaths.