Metacritic Game Data Analysis (too many pics?)

Avatar image for phrosnite
#1 Edited by phrosnite (3528 posts) -

Note: This is only based on Metacritic. Keep that in mind!

I was very curious about how gaming has changed throughout the years and wanted to see if my own thoughts about the subject were correct so I scraped over 16k game entries(as of end of May 2019) off Metacritic and this is what I learned. The average game score by Critics is 70.21 and by Users is 70.16. This pretty much confirmed my suspicions when it comes to what "an average" video game is. Some people are of the opinion that an "average game" is 50 though -_-

No Caption Provided

This graph is interesting but as you all know every year we get swamped by multiple platform releases of the same game, dlcs, constant re-releases of old games, all kinds of editions, collections, etc. and that makes the data above very noisy.

So I decided to go through all entries and add "Leading Version" and "Type of Product" categories. Leading Version is the game on the platform with most reviews and user ratings compared to the other versions of a particular game. So unless stated otherwise, all graphs going forward will be based only on Leading Versions of games and expansions. To show that I've done a good job of picking the leading versions, here is a link to a picture of Leading Platform by Year Graphs

Also, there are many entries(1311 overall, 810 of which are in the "leading version") on metacritic that have 7+ critic reviews but less than 4 user ratings which is not enough to form a "Users Rating" score, so to have a fair Critics vs Users comparison I have decided to exclude those entries as well. That leaves me with 8961 unique entries to use in the comparisons. Last thing, I've limited some graphs to show data between year 2000-2018 because metacritic doesn't have enough game data to show accurate average before 2000 and 2019 is only half the year so.

No Caption Provided

Let's see what the first graph looks like after all that trimming...

No Caption Provided

It appears that Gamers are becoming more and more unhappy every year(since 2005) but at least they are consistent. It's always a drop of a certain length, then plateau and considering the games that have come out this year and what's coming out in the next 6 months it's safe to say that the pattern will continue. By the way, Metacritic lets people give games a 0 out of 10 score. That definitely contributes to the lower scores in recent years.

Critics, on the other hand, have been harsher when it comes to rating games, being particularly negative in 2006, 2007 and 2008 but since 2012 they've changed their tune and might make you think games are becoming better and better every year. The average Critic score here becomes 69.98 and 70.85 by Users.

In the last 8 years we've been living in the "me-me-me" decade in which everybody loves double standards. People often talk ill of critics but will always mention the critics score of a game when they want to justify their own bias of how good a game is and completely ignore the much lower user score. The opposite is also common. People will use the lower user score to trash on a game.

In my opinion when you only mention either the Critic or Users score you are looking at half the picture. The true score is somewhere between the two, i.e. the average of both scores. Critics have always had the propensity to bandwagon and "overrate" hyped up, popular games. Just look at Dark Souls 2 and Mass Effect 3 which have 91 and 93 Critic scores respectively. Dark Souls 2 is universally considered to be the worst of the Souls games and yet it has the highest score because Critic jump on the bandwagon after the commercial and cultural success of Dark Souls 1. If you factor in the the Users score, that 91 goes down to 82(81.5) and 82 sounds about right for Dark Souls 2 imo. Same with Mass Effect 3 and completely ignoring major flaws in the game. Mass Effect 3 in particular is very flawed and I'm not talking about the endings. The whole design is bad but that only reveals itself on subsequent playthroughs. Factor in the User score and the 93 becomes a 76 which is exactly what ME3 deserves.

Some might say this approach will only benefit old games due to the disparity in Critic and Users scores. That is not necessarily true. Let's look at a recent game like Bloodstained: Ritual of the Night. The leading version(PS4) has a Critic score of 83 with 34 reviews and Users score of 89 with 255 ratings. The average score will be 86. This is the fairest approach in my opinion. What do you think?

The following graphs are based on the average of critics plus users score. They show the true state of gaming in the last 18 years imo but I'll include the individual Critics and User graphs as well.

First, let's check out the number of 84+ games throughout the years. Why 84? In honour of Obsidian not getting bonuses for Fallout New Vegas.

No Caption Provided

No Caption Provided

For those who might say, "I know many good games with 80 scores".

No Caption Provided
No Caption Provided

The number of good games since 2010 has dropped significantly, with an overall drop of 59% in the first 8 years of this decade compared to the first 8 years of the previous. And 54% if you compare the last 8 years to the 8 years before that. Now, some people might say, “There are many old games that have very few user ratings so obviously the average of the Users Score will be higher for that period”. Well, even filter out the entries that have less than 100 Users Ratings(which is pretty harsh imo), the trend remains.

No Caption Provided
No Caption Provided

Gamers seem particularly harsh, don't they? Well, the irony is that they have never been as harsh as the Critics have been in 2006-2008.

No Caption Provided

Are games really becoming worse? You decide.

No Caption Provided
No Caption Provided

What are the highest rated games according to metacritic? I've made a few lists based on the data. You can check out the Google Sheets for everything here(many pivot tables; don't open on ancient PCs): Metacritic Game Data

Top 200 Games of All Time Lists according to Critics, Users and Average of Critics plus Users

Links to pictures of the Top 50 for those afraid of Google Sheets: Ultimate List , Users , Critics

Here is a preview of the lists. Critics list is arranged based on critic score and number of reviews, Users - users score and number of user ratings and the Ultimate list - the average of critics plus users score(leading version) and difference between critics and users scores.

Keep in mind that this does not mean a game is better than the one below it. Scores can’t account for the impact on the game industry and/or game design a game like Dark Souls will have. It only shows how much the general public enjoyed a particular game. I’m sure people will still say this and that is better because it has a higher score. So in honour of that and Metacritic allowing people to rate games with a 0 out of 10 X_X I've added a Rank 0 game. And no, A Link to the Past is not my favourite game, it's not even on my list but imo if there was a game that deserved the title of “Greatest Game of All Time”, it's lttp.

Ultimate List
Ultimate List

Critics
Critics

Users
Users

The graphs below are based on the Average of Critics plus Users list. Keep in mind that everything is relative and the fact that games didn't have as many reviews or Users ratings as popular games today doesn't make their scores invalid. The only thing you could say is that some of those games haven't aged well but, in my opinion, that elevates the timeless games(past and future) to heights that will never be reached by games recent or future games.

No Caption Provided
No Caption Provided
No Caption Provided
No Caption Provided
No Caption Provided

Edit:

No Caption Provided

Avatar image for rkk667
#2 Posted by rkk667 (92 posts) -

Really good interesting stuff. Thanks.

Avatar image for majormitch
#3 Posted by MajorMitch (1170 posts) -

I love data. Great write-up though, it's always fascinating to me to look at these kinds of trends. You do a good job of slicing up and presenting the data, and also talking through a lot of the caveats. There's countless ways to do all of that. One thing I was curious about, did you filter out games for having too few critic and/or user reviews? Like, is 5 critic reviews too few? So many games come out nowadays, but reviewers can't review them all. So a high percentage of games end up with just a handful of critic reviews, where the big blockbusters can end up with hundreds. I wonder how all of these factors shift the average over time. Games have changed so much since 2000 haven't they? Kind of hard to even think about comparing "metascroes" of say, Fortnite vs Baldur's Gate 2.

Anyway, fun read!

Avatar image for nutter
#4 Posted by nutter (2394 posts) -
No Caption Provided

Avatar image for nutter
#5 Posted by nutter (2394 posts) -

What was not surprising was the high scores in the PS1-PS2 era.

That was the age of 10 page reviews at IGN, console wars forums, and everything really just being inflated and exploited.

It was also before the 360 era, which while great, saw a slew of shovelware like Carnival Games and Sneak King.

I’d be curious to consider the death of the B-tier game this generation, and how that impacted average scores, too.

Avatar image for phrosnite
#6 Posted by phrosnite (3528 posts) -

@majormitch: The fewest Critic reviews for a game are 7 and one of those games is Symphony of the Night so I decided not to filter based on few Critic reviews but I did do a graph with filtered User Ratings(because there were a ton of games with like 4-50 ratings) that are 100+ ratings just to show that the trend of game scores(Critics+Users) going down remained. So if game had less than 100 User Rating it got filtered despite the number of Critic reviews. There are a ton of game with like 20+ Critic reviews and either 0 or 4-10 User Ratings lol. There is a link to the Google Sheet in the OP. You can copy the page with all Metacritic Entries and arranged them however you want in your own sheet.

Avatar image for lolakarim2300
#7 Posted by lolakarim2300 (2 posts) -

@nutter said:

What was not surprising was the high scores in the PS1-PS2 era.

That was the age of 10 page reviews at IGN, console wars forums, and everything really just being inflated and exploited.

It was also before the 360 era, which while great, saw a slew of shovelware like Carnival Games and Sneak King.

I’d be curious to consider the death of the B-tier game this generation, and how that impacted average scores, too.

Really good interesting stuff. Thanks.