Steam Cleaned: I Catalogued (Almost) the Entire Steam Early Access Library

I have driven myself insane but I did it.

Gotta be completely honest with my audience; if you weren’t convinced of my insanity before this point you’re not going to be thinking well of it after this piece. For the past couple of weeks I have been cataloging every Steam Early Access game. Every single one from 2013 through 2018. Every. Single. One. I don’t think I can express that enough.

Right now the few people who already worry about my continued sanity are probably shaking their screen and asking “WHY WOULD YOU DO THAT?” That’s a great question, and I will let you know when I have an answer. Let’s move on.

While cataloging every single one of the games on Early Access, I was taking notes. How much did the game cost, what was the average review score, is the game still active, and when was the last update. My criteria for activity is whether or not the developer has posted an update in the last year (to avoid being unfair I included the entirety of June 2019). If that sounds unfair, don’t worry. I have more deets.

First I want to go over some parameters for games I did not include:

I only included games currently for sale.
I did not include free to play games in the list as they pose little risk to consumers.
By posting updates, I mean the developer posted an update in the game’s news section. And I did read them to ensure they are actually updates.
Active development means pushing updates. It does not include game sales, dev diaries with no actual updates, or the developer advertising their other titles.
This does not include games that have since left Early Access.
This list does not take into account how “feature complete” a game is at the point active development ceases. I’m not God.

Why is this important? Steam’s Early Access program is riddled with fraud and negligence. There are thousands of games on Steam currently advertising themselves as in-development promising features that will never be implemented by developers many of whom have outright admitted and announced that they are no longer working on the game, or who disappeared years ago either to pursue other interests or just went out of business.

Many of the titles I found do not have reviews that might warn potential customers, have no forum activity or have had that activity deleted to hide negative comments, and I came upon an atrocious number of games whose last post is a plea from the developer that “we’re not dead, we’re still working on updates, buy our game” that were posted after years of inactivity with no further action to show for it. Those posts were not factored into the “activity/inactivity” equation.

We often hear about how Steam’s Early Access program is a graveyard but I wanted to put dollars to dollars to see just how bad the situation is. Between planning, data collection, and writing this article I would say that the whole piece is the finished work of about 40 hours.

Games On Early Access

I don’t think anyone will be surprised by the proposal that the number of Steam games in early access has exploded from 2013 when the program first began. The chart below shows the total number of games still in Early Access as of June 21, 2020 on a per year basis. The number does not reflect the cumulative amount as of that year, but the total released in that year.

All in all 1925 titles make up this graph. We go from 52 titles still in early access from 2014 to 150 in 2015, 349 in 2016, 570 in 2017, and 798 in 2018. This does not include free to play games which would add noticeably to the list, and I should note that these are the games still on early access as of 2020.

If we balloon that to include cumulative numbers the graph looks more like this:

Pretty much the same actually.

How Far Did They Get?

One figure I wanted to suss out in this list was how long on average games were supported after their initial launch, because it is one thing to decry a game that came out in 2015 not being actively developed in 2020, but another to point out that a lot of them didn’t even make it past the first six months let alone the first year. So I calculated the number of games that came out each year and then took those titles and determined what years they ceased to be supported.

2014: 13% of titles were abandoned by the end of 2014, with development ceasing on 28.9% by the end of 2015. A surprising 34% were still being updated as of 2020.
2015: 21% of games had ceased development by the end of 2015, with 42% no longer being updated by the end of 2016. 22% are still being updated in 2020.
2016: 34% of games were no longer being updated by the end of 2016, while an astounding 63% had been abandoned by the end of 2017. 12% have been updated in 2020.
2017: 43% of 2017 released EA games were no longer being updated by the end of 2017, with a whopping 75% abandoned by January 1, 2019. 18% have been updated in 2020.
2018: 50% of all EA games launched in 2018 had no active development by the end of the year and that includes games released up to December 31. 76% were not active by the end of 2019. 23% are still being updated in 2020.

Let’s look at the rate of support at the end of each calendar year categorized into cumulative years of Steam EA releases. As you can see, every single year the number starts off lower than the one before it in terms of how many titles are still being developed by the end of their first active year. So for example in the list of games released in 2015, 86% were still supported by the end of their launch year. Contrast that to 2018 where 50% of games were still active by January 1, 2019.

If you look at the activity rate in the chart below, you’ll see that from 2013 to 2016 the number of presently active titles (updated since June 2019) has steadily decreased from 50% to 18%. You might think that the trend would get higher as the threshold gets smaller and as older games are more likely to be abandoned after so many years. That is not the case.

Show Me The Money, What’s The Big Picture?

Here is the big picture.

Combining every single game on this list into one pile, we find that only 28.8% of the titles have been active since June 2019. That makes 71.2% that have not seen a content update in over 18 months. 555 games, 278 of which were released in 2018.

Loose Lips Sink Ships, The Review Conundrum

I wanted to have a whole discussion on the correlation between price, date, and review score if any patterns existed. They don’t, and attempting to link a connection would be worthless because the data presented is basically garbage.

Why? 46% of the games on this list have no review score.

Yes, 46% of games do not meet the threshold ten review minimum in order for Steam to give the game a review score. So any data we could try to correlate between the score and other factors is mostly worthless because you are immediately losing half of the data. In a perfect world I could collate unreviewed games by looking at each one individually and calculating the review score.

I’m not going to do that because I’ve already titled this project “all work and no play make Connor something something.”

What I can say is that of all of the games that received review scores, the average rating is 66% or as my dad would say, “passing by the skin of your teeth.” There are 34 games on the list that inexplicably have 100% review ratings and might be worth checking out for manipulation. You know who I’m talking about.

One premise that was brought up to me while I was creating the table was whether or not price had any difference on review score. Were people more likely to be loyal and support the game if it cost them more? To justify the price? The answer is no. Not at all.

What’s Your Offer?

Like I said, I don’t think this data is going to kill anyone from shock. From 2014 onward you can see a marked flood of cheaper titles hitting Early Access and, unsurprisingly, not getting supported to a full launch. The average price for a 2014 EA game was $12.22 which steadily dropped down to $9.12 for games released in 2018.

Some facts:

126 games are 99 cents.
686 under $5
1277 under $10
1842 are $20 or below (one game costs $20 flat)
83 titles at least $24.99

Another thing to note is that time does dampen perspective, as the average review rating for a 2015 game sits at 58% while the titles in 2017 command a higher 71% average approval rating. I compared those two because they have similar pockets of unreviewed titles. It should be noted that in 2013 when Valve cared about curation the average review of games from that era is 66%. Fitting.

Let’s Pull 2018 From The List

You know 2018 is pretty easy to fulfill my criteria for; for most of the titles on the list it means supporting development for somewhere between six months to a year. I have nursed the grease in a cast iron skillet longer than that. Let’s see what the chart looks like for games in 2017 and earlier.

The reason I pull 2018 from the list is that 2017 was something of a massacre for Early Access titles. It’s kind of interesting to see because for the first four years the common denominator for abandoned games is 2017. It’s like all of the crappy devs got raptured over the course of a few months. Does that make this heaven? Are we the blessed ones?

As you can see from the pie chart above, out of all of the games released in 2017 and prior, one third were abandoned in 2017. 52% of all games still in Early Access as of 2020 had ceased development by the end of 2017. Of that list, 276 or 24.49% are actually still seeing active development releases (since June 2019).

What Is My Solution?

There are too many games that show up on Early Access where the developer never continues to support it, leaving the forums and news posts utterly empty. It sounds like a cash grab scam. In some cases it probably is a cash grab scam. Others are just crappy developers, people who were reliant on early sales that never came through and shut down, or people who are selling a product but treating it like free income.

I have set up a four step program to clean up the Early Access system. I assume most of this stuff can be automated with Valve’s internal statistics to monitor an app’s depot activity for inactivity. You definitely have the tools to not put as much work into this as I did.

Games that are no longer on sale should no longer have visible store pages cluttering search results.
Any title that is not actively supported in the first six months can be reported for review and potentially removed. Valve must actually read these reports. If you launch a game and don’t post updates or news announcements for six months, I’m hedging my bets that game won’t be seeing launch.
Selling a nonfunctional game should warrant a developer/publisher ban.
Games no longer in active development should be given the choice of launching or retiring. Launching will require them to remove all promised upcoming features from the store page.

Otherwise I have no opinion on the matter. I’m going to go lie down now. Have a nap or something.

Play video games.