There's been a pretty big hubub over Metacritic and review scores lately. Many reviews seem 'off' compared to the game, and there are no shortages of people saying so in the comment section. Jim Sterling created two reviews for Final Fantasy 13, one which used his opinions, and the other which was an objective review (and clearly a joke). This showcases a debate: How should a review be conducted? Should a bad game that leaves a good impression get a higher score than a technically perfect masterpiece which played like crap? I will attempt to give an answer below...
The first thing I want to do is provide some background on the human mind. Many of you may be familiar with the Myers Briggs personality test (ESTJ/INFP). This describes how people function by identifying 4 key traits, and is generally more accurate than a Chinese Horoscope (Dragons rule). There are two things about this typing that are important: the first is each letter is not dominant, so someone could be 60% extroverted and 40% introverted, or closer to 90/10. The second thing that this shows is in the third pairing: Thinking or Feeling. In my mind thinking describes the 'math' group and feeling described the 'English' group. Math people tend to like that there is one universal truth as far as an answer goes, and English groups appreciate how a complex question can be argued from multiple angles. I am a math person myself, which is probably why I feel I write so very, very poorly.
CBlogs need pictures. I cant think of a relevant one. So here's a bunny
So that's all well and good, but what does it have to do with review scores? From the perspective of a reader the 'math' type is more likely to scroll down to find out the score, and mentally categorize the game based off that. The 'English' type will read through the whole review, and base their opinion off of what was written instead of going after some bottom line. As mentioned before, this split isn't an on/off kind of split, so math types may read the article, but typically would put more focus on the score and vice versa.
Review scores are written from the 'English' point of view, from what I can tell. A game is graded similarly to how an essay is graded, or more accurately how Art is graded. The reviewer plays a game and is looking for something that cannot be defined by numbers and scales: how much they enjoyed the game. Then they assign a score to the game based off the proverbial taste in their mouth. Do you see the problem? The scores are typically arbitrary, assigned based on whim. So what we get is someone trying to justify to the thinkers why they feel the way they did, which is a difficult task, so both parties will wind up frustrated, and will troll the shit out of the comments (I mean, let's be realistic here).
Mostly unrelated, but Maddox has mastered how to judge objectively. Also decided to stick with the bunny theme
There are two solutions to this problem. The first solution is the remove the scores from the review entirely, but as a math person, I can't endorse that. What I would suggest is a new reviewing system. Each review is given 50 percent to the old style, and 50 percent to my proposed style. What the reviews need is a universal gaming rubric. People could then figure out which side they stand on, or what weighting they would want to apply to the two sides. The way I would imagine the rubric going down is half of it would be dedicated to things that all games contain: gameplay, graphics, sound, etc. The second half would be genre specific, so for an RPG the character progression, battle system, item management. Each genre could also assign a weighting to certain aspects so Arcady style games rely less on graphics, for instance.
The only problem I have encountered with this system is that I am not careful enough to capture every possible metric for each genre, so I do not have an example of the rubric available. However, it is possible to share documents via Google documents, if there was a desire to create a review metric that measured how a game rates without fear for objectivity and bias (mostly). I suggest, but I am too lazy to build, a review rubric that can be used universally. And some people didn't think it was possible to review without feeling....