Phoenixmgs said:
Regarding story and score weight I agree with you 100%. As far as stuff like Arkham or MSG4 having 90+s go, I think there are games out there that legitimately deserve such high scores (Doom 2 comes to mind), but I see where you're coming from what with a disproportionate amount being that high.
In that regard, I think it may have at least partially to do with games being innovative. Ocarina of Time is another game with a laundry list of issues and something I'd personally rate like a 7-7.5, but when it came out it was an absolutely revolutionizing moment for the industry, and as a result it's one of the highest-rated games on metacritic and often makes it to #1 on GOAT lists.
Arkham and MGS4 were both in turn highly innovative and/or ambitious games, so while we look at them through a more critical lense now, back when they came out they were a huge deal and the reviews at the time reflect that perception.
Haven't played W3 myself but from what I've been told from friends the flaws seem to add up to something like 8-8.5/10, so while DD or Bayo may have 9+ grade combat, which may make W3 seem like a 7 by comparison, not many reviewers are going to score it as such. Its the same idea behind why a schoolteacher won't mark the rest of the class down just because one student's 10/10 essay is miles better than the other ones that still met the criteria to be 10s.
I do think thats another thing worth noting, that while its possible for a game to have many flaws, if they're all comparitively small (example, Quake 1's two bosses are utter garbo, but they're just a teeny tiny part of the game as compared to the other 36 or so levels), then it may not have much of an impact on the score, especially if its something optional such as the Riddler trophys where the impact on the game's "core" experience can be easily ignored if not practically nonexistant.
With Yahtzee, I feel like the fact that he doesn't score the games leads to an inherently different review dynamic that leads to it being easier to point out more commonly-ignored/accepted flaws not only within a given game, but flaws more intrinsic to that game's genre.
As a personal example I could write entire essays about all the problems I have with JRPGs that use RPG Maker style gameplay, but if I was actually scoring a game from that family I'd likely be ignoring said issues as they're more inherent to the genre than to the individual games.
On a final note I think maninahat's post a couple posts above your response to me brings up a great point in his main paragraph, so def read it if you skipped it over.