Again, like any scored system based on subjective criteria, it's a guide, not some sort of integer you plug into your brain and say, "I must act on this number."

These are reviews. Metacritic is actually a remarkable way to try and bring some semblance of order to the review process. It isn't perfect, but it's the best there is out there in the way of data reduction of reviews for games.

The problem is that metacritic is indeed starting to become poisonous. Publishers and Devs are starting to get forced to change their games in ways that are appealing to critics, even if it means screwing consumers over. Why? Because Metacritic slaps a giant round score in the face of the title, and that's all people tend to read, because they're lazy. Metacritic is bad, especially on the long run. Right now it may be only 1 or 2 devs doing it, but when people buy more of the other companies' games because they see a better score on there, the ones that don't do it will eventually start to have to, or sink because their game doesn't sell as much.

Basically, Metacritic is forcing developers to design games by ticking tickboxes (they are being standardized and streamlined, to the point of where most games are extremely similar) and designing them to appeal to focus groups (in this case, the critics).

I direct you to this video, as totalbicuit knows how to explain this better than I can: http://www.youtube.com/watch?v=Vfn1ZxE9zMU (relevant info starts at 44:30)

I am also firmly against the belief that students should be rated by numbers, but I'm not going to get into that, that's way off topic.

No need to apologize for the wall, I love discussing these kinds of things :P.

As you pointed out, the use of numbers in games journalism isn't doing anyone any favours. For as to why, I direct back to my second reply in this topic, right above.

Exactly as you said, the bad assigning of numbers is one of the biggest problems (I do want to point out, though, that I am referring only to critic reviews, not to user reviews, which I just find useless). However, in my opinion, the biggest problem is that we have been trained to find that numbers are always the answer, and that every time they are more precise than words. As such, when someone enters the metacritic pages, no one will bother to check the written reviews, because they have the number right there. There isn't even any need to check individual scores, just look at the meta one. And, why, imagine that the game has a metascore of 70. However, the reviews themselves pointed out that the game is good, and is worth playing, despite some control issues and dated graphics. The numbers don't tell you that. The devs may sell less games because of it, because people have grown to believe only games with 80+ metascore are worth playing.

Basically, numbers might be a valid, even if mediocre and archaic way of expressing opinions. However, I do not believe that they should be asigned to games, because people are lazy. If you put a number there, people will just scroll and see the score, and the number doesn't tell you everything that was written in the review.

The thing with the Obsidian bonus is that you don't know the way how metacritic gives their scores out. They give more relevance to certain reviewers than others, which makes no sense. To quote myself (sorry if that sounded pretentious) "I just want to point out that Metacritic does give more relevance to scores from certain sites when compared to others, which is basically saying that some people's opinions are better than others, which seems completely legit to me.". That should never be a thing, a metascore, if it absolutely had to exist (which I again don't think is the case) it should be just the average of every score, and it clearly isn't. That gives Metacritic space to manipualte scores at their will whithin a certain spectrum, which should never be allowed just for ethic reasons, but it is even worse when companies decide who stay and who leave based on metascore.

As any statistician would tell you a weighted average like presented by metacritic is worthless without a LOT of additional data (sample size, error margin, confidence interval, the weights themselves, the functions used to determine those weights, systemic bias). A single percentage on it's own is indeed unable to represent a vague opinion: 7.0 is simply more than 6.9 and less than 7.1.

Again. Not Metacritic's fault. This was Obsidian's poor policy. Disclaimer: Haven't watched videos. May have time this evening.

If Obsidian's management is that obtuse, then it's on them, not an amalgamation of game scores.

This same thing happens in social media hiring with places like Klout. It isn't Klout's fault that their proprietary, undisclosed system of social media ratings is relied on by hiring managers. That's a fundamental misunderstanding of variable weighting on the part of those managers.

Again. Education. Understanding.

edit: Also amusing that the opinion of one person in a Kotaku article is weighted more than Metacritics algorithms. ;)

well to be fair about obsidian's thing....all games vaguely associated with bethesda are buggy as shit on release and they never ever learn. Its like they don't want to betatest them. its always a short time with a very small pool can't catch everything bla bla excuses. why not a bigger pool then? how is their testing pool small enough to miss main story bugs that apparently made the game impossible to complete for some like skyrim had? people volunteer to do that sort of thing for free as an excuse to play early, people would even pay extra to do so. other companies do it(reach beta was in odst preorders or something as a selling point. games with early access buys ect.) but its always "we'll fix it later" and there are always glitches left when they stop supporting it to work on the next thing...glitches that some 13 yearold in the mod community fixed long ago.

that wasn't completely obsidian's fault(I'm pretty sure its bethesda since all their games have it not just the one obsidian did) but those things really do need to be taken into account in reviews or they'll never learn.

the critics didn't know about the bonus, and if they had should they have changed their reviews to overlook flaws? (imo the critics do more than enough of that already giving high ratings to all sorts of mediocre at best games rehashing the next installment an old franchise with nothing different but the year(not newvegas which was actually good though oddly).)

I can't speak to game reviewers, but in the case of most reviewers, they are literally experts in the field. Games journalism may be different. I don't know.

I work in automotive, and as with any other subjective process, there are certainly folks you will agree/disagree with. But most of the folks reviewing the cars have thousands of hours driving them. It's all they do. So they can get in a car for a shorter period of time than the average Joe and be able to discern some things right away and pass on that knowledge to the consumer. They also drive hundreds of vehicles per year, so they have many things to use as reference knowledge in comparing a vehicle. It's still a good idea to test drive a car, when possible, to confirm/disagree with reviews, but there is definitely good information in reviews, no matter how subjective. And there is definitely more weight in automotive journalism reviews than there is in the opinion of the majority of the general public, who usually drives one (maybe two) primary vehicles, and keeps them for a number of years.

I don't spend enough hours gaming to know if most games journalists have that same expertise. Certainly the price and time barriers are not as high as in automotive.

For my experience and knowledge, I know most game reviewers should most definitely have more heavily weighted opinions than my own. But perhaps the knowledge gap is not what it is in automotive.