Outliers can ruin Metacritic lol, especially those that are intentionally outside the norm to be seen and get clicks.

It's still reliable as a ballpark judgement of games, though.

Isn't that just how math and averages work?  I mean metacritic uses a weighted average so this site noone knows about probably has less of an effect than a 10 from something higher, but this is just how averages work on a scale with a set maximum.  What would you suggest as a balanced system?

I personally think eliminating the 10% lowest/highest scores would give a more accurate result. Extremes are rarely a good representation of a sample.

The problem is combining different scores systems on the same table, some webs give 7 and 6 easily when a game is decent or good while other sites only give a 70 to an extremely mediocre game, most of the webs score between 70 - 100 and anything below that is utter shit, being 70 already mediocre, so when you combine those different systems the result is a mess, and also is funny that a 7 is more imprtant than a 10, since the 7 is drowing the score a whole point but the next 10 will not be able to raise it up. Maybe a good solution would be cutting some of the very best and very worst scores out of the table.

Agreed!

I also dont like that some reviews weigh more than other.

Why the hell a review from IGN or Gamespot should be more important than an indie site?

Thats like, their opinion man...

I would always prefer that we see every possible data point and not make any cutoffs.  I'm much more a fan of something like Gamerankings in which the average is unweighted and they include more sites.  If there are concerns about the spread than meta just needs to report more numbers such as the standard deviation to give you an idea on the range of scores that are being reported.  As an analyst I don't think Meta should concern itself with the idea that certain sites give out of the norm scores for hits, but simply report the numbers given and the relavent statistics.  

It's not unbalanced, it just is what it is. People need to stop focusing on the metascore and look at the individual reviews instead.

I think a good solution to the scoring system would be to have a set amount of reviews. Get the 20 most influential sites and create an unweighted average out of them.

Problem solved.

The fact that you can't link levels together to make an official Mario game killed a sale for me and I've been playing Mario since the first one launched. I just can't imagine playing in isolated levels long enough to justify that price.

An 88 for a 2D platformer is a 9000/100 for an open world game.

"Despite some lacking features when it comes to looking for other people's creations, making levels has never been so fun."

Gives it a 7.8 -,-"