With the NYC analysis here, everyone is missing one unknowable piece of information. We know that the initial major outbreak in NYC was with a strain from Euro

With the NYC analysis here, everyone is missing one unknowable piece of information. We know that the initial major outbreak in NYC was with a strain from Europe, while the initial major outbreak in California was with an Asia-origin strain. These might be differently dangerous. If the NYC/Europe strain is just 20%-30% more fatal than the Asian one, the maths all start being believable (you start saying that 30-45% of NYC has been infected -- which, unlike 60-100+%, is quite believable. It seems likely that some vulnerable neighborhoods in NY have effectively everyone infected, but that is probably not universally true. It also seems quite believable that we are nearing the point where the majority of folks in NYC have been exposed. If nothing else, the subway is still seeing quite active use, and must be a breeding ground for infection -- and yet, hospitalization has started declining, meaning that new infections are slowing down -- meaning that at least among folks that still go out and about, there must just be fewer people to infect. With subway use still being as high as it is, it seems like it must be true that, amongst the population using it, R could only fall below 1 if herd immunity (in the rider population) was beginning to form. This also jives with the numbers we are seeing, where reported cases are down despite both increased testing and looser counting.