Even a challenge trial would need to be run properly and include enough people to generate useful data. But let's play the opposite. Let's be unscientific and a

Even a challenge trial would need to be run properly and include enough people to generate useful data. But let's play the opposite. Let's be unscientific and assume an untested vaccine on a novel platform would reduce transmission risk (Edit: if we assume this, why not just assume the candidate vaccine works and just skip the challenge trials too?). Further assume that recruiting enough volunteers for meaningful results would be easy in spring 2020, before the age cohort differences in severity were as established. Given those assumptions, there still needs to be a smaller safety trial before a large challenge trial. BioNTech started those in April, that ended in May. Deployment in the summer is already aggressive, assuming no challenge trials and we had just deployed vaccines shown to be safe in very small trials. But suppose a challenge trial was run rapidly over the next three months (to determine which of the BioNTech candidates was best). Assume this all goes optimally. This plan still assumes the existence of scale manufacturing facilities that likely did not exist last summer. Let's wish that away too, so we have scaled manufacturing capacities to June 2021 levels by May 2020. While we have our magic wand out, let's wave away the fact that the vaccine initially took over 100 days to manufacture, so that in our world we can decide what to produce today and have it ready tomorrow instead of next quarter. So we've got positive challenge trial results in July 2020 with faster-than-2021 production capacity to go to scale. Now, perception-wise, we are asking young healthy people to take a vaccine designed on a novel platform that was rushed through a novel ethics-skirting approval process, for a virus that likely won't seriously harm them if they get it. I'm not sure that would be as effective as people seem to think it is.