I agree that reviewers aren't equipped to match the tastes of the market as a whole, but I think they do a good job of matching their readers' tastes.
First off, it's clear that they do a good job with traditional games. The score of Resistance is pretty much right on, and the divergence is about what we'd expect given the tendency of some people to vote zeros on great games. The complaint to be made here isn't that they prefer Resistance to Wii Sports, but rather that they just don't like Wii Sports.
Secondly, I think everything makes sense if we understand Wii Sports as a 'niche' game. It's not that it's really such a game, but that, among people likely to be on Gamespot, it's only going to really resonate with a few. I can't imagine its real target audience voting for it on websites. Look at the scores of other niche games - being a Mega Man fan, I looked at Gamespot's X8 review. There's almost a two point difference between reviewer and user scores.
While I'm inclined to agree to some extent, I don't think we can ignore the self-selection bias of relying on people who read a game's reviews to rate the game. You're going to either get people who are interested in the game or who really hate the game. For games that only appeal to a small section of a website's readers, 'people interested in a game' are also people that are far more likely to enjoy the game. I can't imagine that many people who don't like Mega Man read Gamespot's Mega Man X8 review.
That said, everyone knows about Wii Sports, and it is surprising that we see that divergence for this game in particular. However, I'm unconvinced that the site's readers' opinions are really reflected in the user scores.