You quoted the worst of the four models that they evaluated ("This model was the least parsimonious model we evaluated"), and only part of that model even. They

You quoted the worst of the four models that they evaluated ("This model was the least parsimonious model we evaluated"), and only part of that model even. They're not pushing that sole hypothesis; they were testing to see if that could explain the current evidence. The other models they tested were better: > First, we modelled increased infectiousness as an increase in the risk of transmission of VOC 202012/01 per contact, relative to preexisting variants. This model was best able to capture the data parsimoniously among the four hypotheses tested...