So you are assuming that my statistics are based on number of downloads?
One of the easy statistics is how much more voting happens on threads that get front-paged. This means that even if we restrict to monks who have votes (which excludes a large fraction of our audience), we have a pretty large bias towards use of (not downloading of) the front page. (Yes, you can argue that better nodes are more likely to get front-paged and also more likely to causing voting, but I'm pretty sure that many are quite convinced that such an explanation just doesn't cut it.)