There is a long known effect that open access articles get more citations compared to non open access articles (e.g. doi.org/10.1371/journal.pbio.0040157 ). So

There is a long known effect that open access articles get more citations compared to non open access articles (e.g. doi.org/10.1371/journal.pbio.0040157 ). So I do not see why this should be very different for Sci-Hub articles. However I agree that the claimed effect might be exaggerated by confounding. More interestingly proper deposition of the data corresponding with the article increases also the citation rate ( doi.org/10.1371/journal.pone.0230416 )