A problem I see here is that it extrapolates the statistics of users users of the extension, onto all users of Youtube. Fortunately the users of the extensions

A problem I see here is that it extrapolates the statistics of users users of the extension, onto all users of Youtube. Fortunately the users of the extensions are most probably better informed than the average Youtube user. However, if the downvotes are more "accurate" (more deserved) because of that, it means they are also more accurate for new videos as opposed to old videos. So I suspect new Youtube videos might have more downvotes than old Youtube videos. Perhaps we could find some reuploads of all Youtube videos and compare the downvotes…