Archive for February, 2009

Part 2: Metacritic Comparison - Applying the Sample

Now that we’ve spent some time explaining the sample, it’s time to challenge metacritic.  In this video, we’ll be comparing metacritic against gamerankings.com and gamestats.com.  Our goal here is to determine if metacritic is an effective scoring index.

Get the Flash Player to see this player.

Tuesday, February 10th, 2009 Metacritic Uncovered No Comments

Part 1: Metacritic comparison - Examining the Sample Data

This video is simply a brief overview of the sample data we’ll use for our metacritic challenge.  As mentioned in the video, we pulled this sample in July for 100 game titles in 2008.  If you would like to view the sample data, please right click here, and “save target as.” 

Also, there is a small error with the dataset, but it won’t effect the results.  We originally prepared the data with 100 random samples, but added a few more at the end to offset some of the missing data from gamestats.  The video only touches on a few results from quite a long handling process with the data.  If you have questions, you know where to find me.

Get the Flash Player to see this player.

Monday, February 9th, 2009 Metacritic Uncovered No Comments

And we hit the ground running…

I’m going to be spending a good amount of time talking about Video Game Scoring trends using the industry standard Metacritic.  Before I actually start slicing and dicing the data, I’d like to take a step back and sort of defend Metacritic as a scoring index source. 

I deal with this stuff all the time, and frankly, I can model whichever data source the industry has adopted.  Still, there are those who critique Metacritic and they deserve my time as well.  With that in mind, I’ll be presenting a few videos which challenge the integrity of Metacritic, as well as diving into how their weighting system really effects the scores.  We’ll also get some insight in regard to how the weighting has effected the scores over the years.

Most reports available only show trends as it relates to the basic game meta data, i.e. Metacritic score, genre, date released, ESRB, etc. which we refer to as a “shallow dive.”  This “shallow dive” can incorporate this meta data for up to the 7,900 game titles (divided by platform) on Metacritic.   Our modeling application can also show various trends as it relates to the contributing outlets as well, for example, determining possible media bias. 

One example of a “deep dive” will apply to our analysis of the weighting system.  To do this, we’ll have to compare the Metacritic score to the actual non-weighted average of all the contributing media outlets for each title.  So if we have ten titles, each with 35 media reviews, that’s 350 data points.  That’s right, we have to average the scores for each title and compare it against the metacritic score.  Now, consider that we’ll present trends and comparisons from 2000 through 2008, and you can imagine the scope of the task.

So, lets get started!

Tuesday, February 3rd, 2009 Uncategorized No Comments

Bad Behavior has blocked 25 access attempts in the last 7 days.