It's relevant for two reasons. First, lay people presented with competing expert arguments are generally reliable. Not universally, of course, but often enough

It's relevant for two reasons. First, lay people presented with competing expert arguments are generally reliable. Not universally, of course, but often enough that it's probative. It is useful to get a "gut check" from someone who isn't close to an issue, the comparative increase in objectivity outweighs the comparative loss in expertise. Second, regardless of the abstract question of any particular hypothesis, the credibility of the scientific field in the eyes of the citizenry is important. Even if a scientist knew with certainty that the lab leak hypothesis were false, wasting credibility on that position without compelling evidence would be "damaging to science," in the sense that it decreases the influence of experts and, likely, their funding. TL;DR: It's important because, when choosing a hill to die on, you have to account for the strength of your evidence and whether people are inclined to believe you. The opinion polls are relevant to the latter question.