It is certainly true that there is a skewness in the infection data, and that there are a relatively small minority fraction of infected people who are responsible for the majority of “superspreader” events where dozens or hundreds of people are infected; however, from what I’ve seen there is no good trending in the data to allow for predicting specific groups or individuals who are prone to this, and it isn’t clear how much of this is innate versus behavioral. At this point, I think we have to assume that individuals who are prone to much higher rates of transmitting the disease to others are more or less evenly distributed in the population such that inoculating a representative demographic cross section would still result in the same general replication number among uninoculated populations.
In the US, we really have no idea just how widespread new variants are. The lack of genomic testing and surveillance is a stunning failure to take action to try to prevent wider contagion, with at least a few state health officials expressing the obtuse opinion that it didn’t matter which variant was dominant. Even with health departments stretched and labs worked to capacity, it would make sense to at least do enough sample testing of infectious people to get a statistically meaningful estimate, and we do not have that in any state to any degree of statistical confidence.
I have to disagree with that assessment. We see cases starting to rise in early December, which correlates with November (American Thanksgiving holiday, 26 November), and then at least a local peak in reported infections in early-to-mid-January for most states, which corresponds with Christmas-time gatherings, with a falloff in new cases after that as people stopped holiday gatherings and some states got more serious about imposing restrictions. There are a few states that buck this trend but they tend to be rural states with lower population densities and also lower per capita rates of testing. There may have been better compliance in December indicated by the lower peaking in January than in December, but the trend of local peaking is still there.
Regardless, at this point there is no reason to believe that we will assuredly achieve a herd immunity threshold without vaccination; aside from the native infectiousness of this pathogen, it is clear that at least in some significant percentage of people immune response wanes in a few months after exposure, which is consistent with what is known about other common human-infecting coronaviruses. With an effective vaccine and a vigorous inoculation campaign it should be possible to push this contagion down below the epidemic threshold and make it no more of a threat than influenza or any other of a number of infectious pathogens (provided, of course, that it doesn’t mutate into a really aggressive strain that evades the current crop of vaccines entirely, and even then, it is probably possible to make a vaccine that targets other parts of the capsid that are not as prone to mutation as the S-protein) but it does not appear that we would be able to get there via naturally acquired immunity through infection.
Stranger