I personally take my reviews from very few sources. Admittedly, most of them are from the Escapist. But most of my judgement on a game is based on the trailer and past experiences with the game. If the previous games are good and the trailer seems interesting, I'll go for it regardless, then decide wether I agree or disagree with the reviewers (I personally thought FFXIII was a GOOD game). If it doesn't seem all that great, I'll check the reviews first, and if they really say things like 'the box didn't actually contain a game, it literally contained boxed defecation', then I'll avoid it.
The only way in my opinion to be able to healthily criticque games it to play a wide variety of good and bad games. Some of the bad ones might have a lot of redeeming features that could be improved upon, and an overall good game might have some serious flaw. But in order to understand what makes a good or bad game, you have to play a little of both.
That way you don't turn into the 12-year old COD player reviewing Half-Life 2 on Metacritic going "no iron sights qq no perks qq no multiplayer qq i know it's 2012 and the game was made in 2004, but they should update the graphics in a game they made 8 years ago because qq and it must be bad because the developer's name isn't activision or treyarch qq"
And that was a copy-paste of an actual review I read on Metacritic of Half-life 2
