Note: iOS 9 + Facebook users w/ trouble scrolling: #super sorry# we hope to fix it asap. In the meantime Chrome Mobile is a reach around
hot  /  reviews  /  videos  /  cblogs  /  qposts


Nathan Evrard's blog

  Make changes   Set it live in the post manager. Need help? There are FAQs at the bottom of the editor.
Nathan Evrard avatar 1:43 PM on 09.11.2012  (server time)
A Statistical Analysis of Video Game Scores

Video games have long been reviewed and given scores (grades) to let people know how good they supposedly are. This should be an objective process but as many people know there is a definite subjective element at play. Over the course of the last several years the defacto scoring metric has become Metacritic.

The site aggregates critic reviews alongside user reviews for direct comparison. To get the final Metacritic score for critic reviews the site uses a special formula to weigh the reviews that are included in the metric. The end result is a score that unfortunately can affect the wages that game developers get paid.

Going into this particular analysis there were two theories that were going to be tested. The first is based on the average review score. On Metacritic games are scored on a 0-10 basis. Given these numbers the average score should theoretically fall at the 5 mark given that that is the mid point or average of the range. My theory going in is that this is in fact false and the average score will actually be much higher.

The second test revolves around user scores vs critic scores. It was my belief going in that user scores would be lower. This stems not necessarily from more stringent reviewers but rather from hatred towards publishers such as Electronic Arts.

The study used a simple random sample of twenty games from both the first person shooter (FPS) and role playing game (RPG) genre. The genres were graded separately from each other and broken down in various statistical forms.

Descriptive Data

Score Comparisons


The data above simply represents the breakdown of the numbers at our disposable. It does not however give us an answer to the questions that have been pondered up above. To do that we need to do something called a hypothesis test. The first test simply compares the average score to the theoretical average of 5.

The second test compares user reviews directly to Metacritic scores.


The formulas used above may look quite confusing to some and that is understandable. Ultimately, the tests we performed told us two things. In the first test the result simply states that yes, the average user score is significantly greater than 5. In the second test the results tell us that the average user score is less than the critic score. To be more specific however, in the world of statistics we don't say that we are correct but rather that we fail to prove that we are wrong.

CrossPost From Video Game Mojo

   Reply via cblogs

Get comment replies by email.     settings

Unsavory comments? Please report harassment, spam, and hate speech to our comment moderators

Can't see comments? Anti-virus apps like Avast or some browser extensions can cause this. Easy fix: Add   [*]   to your security software's whitelist.

Back to Top

We follow moms on   Facebook  and   Twitter
  Light Theme      Dark Theme
Pssst. Konami Code + Enter!
You may remix stuff our site under creative commons w/@
- Destructoid means family. Living the dream, since 2006 -