I agree that if reviews are coming, the reviews should be able to be voted on. This seems to work fairly well for MAL reviews, which lets you mark reviews as "helpful". I think this makes the top 4 reviews for a work on MAL fairly representative of the full range of opinions on the work.
As YouTube found out years ago when they made the transition from the 5 stars rating system to the like-dislike rating system, people tend to only make a review if they really liked it or really hated it. So personally, I find the distribution of the review scores on MAL to be more informative than the score distribution histogram because the selection bias filters out the people who don't have a particularly strong opinion either way, and really brings out the emphasis on the strengths and weaknesses of the work.
For example, if the top four reviews have something like a 10-9-9-8 split, it's probably very good! If there are no or only one or two fairly non-extreme reviews, it is probably not particularly good nor particularly bad. If it's something like a 10-3-9-10 split, then it's probably pretty good for most of the audience, but a pretty significant portion of the audience had some serious issues with the work, so YMMV. If it's something like a 4-10-6-9 split, then it's a pretty split fanbase. If it's something like a 4-3-9-8 split, then more than half of the audience have some serious problems with it, but a smaller subset of the audience really love it. If it's something like 10-10-1-1, then it's probably either Mars of Destruction or Boku no Pico ?
Just some of my thoughts.