It is January so it is time for journalists to look up how many games were released last year and say “Too many games. Steam is oversaturated, <x> number of games released this year and most games got 0 visibility.”

Here are the examples from last year

So this is my attempt to give a bit more context and also maybe have some alternative takes that aren’t so clear cut as “too many games.”

First of how many games were released in 2025?

It depends on which tracker you check. 

  • 20,558 – GameDiscoverCo
  • 20,353 – https://gamalytic.com/
  • 20,282 – VGInsights.
  • 20008 – SteamDb

The discrepancy among them is because sometimes Valve pulls games (like this one). Each of these tracking platforms handle these edge games cases slightly differently. 

For the purposes of this blog I am using VGInsights. Assume all numbers come from there except when I cite total users because that comes from SteamDB.

Is 20,000 really too many?

Often these articles have an exasperated tone: “TOO MANY GAMES!”… But too many compared to what?

Here are stats from other mediums:

  • Amazon: 350,000 new books are uploaded Every MONTH (Source – also, interesting font choice)
  • Roblox: Apparently there are 40,000,000 “experiences” in 2023 (Source)
  • Spotify: 112,000 new tracks are uploaded EVERY DAY (Source)

Independent musicians, authors, and Roblox developers look at us complaining about 20,000 games being “too many” and laugh and say “oh you sweet child.”

But 20,000 is bad right?

Let’s look at previous years

That line definitely goes up and to the right. 2025 had the most games released ever in a year. It also is the second greatest raw increase in numbers of games.

However the strange thing is 2025 was actually a down year for percentage growth. The growth was slower than the last 2 years.

Let’s look at the percentage growth. 

Ignore the numbers pre 2017 because that was when Steam Direct started where anyone with $100 could submit their game on Steam. So here is the graph zoomed into just the post Steam Direct period:

From 2023 and 2024 it looked like the number of Steam releases was accelerating. But then look at 2025’s 11% increase. It is the 3rd lowest percentage increase since Steam Direct started.

MAYBE just MAYBE we are entering a period of what they call “concave-down growth” also known as the number of titles every year is increasing, but at a decreasing rate. 

Hot take

For a minute, consider that we might be seeing the end of the endless growth of Steam. 2025 could just be a blip, or maybe the next couple years will see the beginning of the decline in the number of games uploaded to Steam. The number of new games will get lower and lower until fewer games are released one year.

If it does happen, here are some reasons why:

The end of free money

In the early 2020s big companies and investors were throwing money around. I even saw Hollywood companies that were so flush with cash that they were looking for new investment vehicles and picked games as a good target. For example Netflix bought its first game studio in 2021. It was a heady time.

A lot of games were getting funded and greenlit in the 2021, 2022 period. Developers that had little to no experience shipping games were getting millions of dollars to make a huge project. Their ideas were inflated beyond their initial scope because, why not, free money. Games with significant budgets take 3 to 4 years to make which could explain why 2023, and 2024 were the highest recent growth.

I also wonder if 2024 had that spike because the money ran out, layoffs started, and many teams panic-launched their game because there was no runway left. 2024 may have been the exfoliation of dead skin on the surface.

The end of weird money

The early 2020s also had investments from some weird ideas that didn’t go very far such as blockchain / NFT / Web 3 games. That money is now gone. I started seeing this last year when a lot of companies asked for advice on how they could reposition their game from an NFT game to a regular old Steam PC game. 

Also no more talk of Metaverse. And I am really sorry, and I know this is going to get me emails, but VR is also pretty much at a cul-de-sac. Yes, I know about the Steam Frame but I would be surprised if it will improve the chances of VR games. They just don’t sell unless it is a game about an animal playing tag.

So expect more boring games with regular old pay-up-front monetization where you just have to play your game with the typical keyboard / mouse / controller. 

AI is sucking the speculative money out of games

On the margins I am seeing some weird AI games getting funded. However, it seems to me that most of the AI speculation money is going towards AI-powered game development tools and business-to-business software; not so much funding actual games.

That means more tools, and more LinkedIn messages asking if you want to “jump on a call to hear how our AI startup can streamline your game dev pipeline.” But that doesn’t mean we will see more games on Steam.

I also think the investment class of entrepreneurs are so dazzled by the economic upside of AI applied to SaaS, data centers, and military applications that investing in games looks like small potatoes. Games are no longer the exciting speculative asset.

The AI-generated game flood didn’t happen

In January 2024 Valve opened the floodgates allowing games made with AI onto Steam. I was interviewed in this article with the scare-headline of “Steam Users Worry Good Next Fest Games Are Being Crowded Out By AI Slop, But The Real Problem Is More Complicated.” I think I provided the quotes (and graphs) that made it “more complicated.”

Devs were worried that there would be a “1500 slot machines problem” which is from this GDC talk about 2 devs who rigged it up so they could auto generate a game with a couple word prompt. This was all pre AI!

But, honestly, the flood hasn’t come. Sure people make some AI capsules, or insert some AI art here and there in the game. But if AI games were really a big problem, I would have expected 40,000 games to be released in 2025. 

On Steam nobody is spamming out a new AI-generated game every day. Honestly, there isn’t. I checked! 

Using VGInsights I scraped the list of releases to see which publisher put out the most games in 2025:

Here are the 3 publishers with the most games released in 2025

  1. Bully Revenge Studios [Warning! SUPER NSFW] published 138 adult visual novels in 2025. I don’t think a single game has more than 1 review. 
  2. EroticGamesClub is next with 82 games
  3. Me (that’s the actual publisher name) released 58 games and they all look like boring-old-pre-AI-asset-flips. Their best seller has 46 reviews and it is a vampire-survivor-like named Big Survivor 

There are a couple legit publishers that published a lot of games

Big Fish Games published 38 games 

PlayWay S. A. published 26 games

Really there isn’t any publisher that is auto generating 10 games a day. There are only 22 publishers who put out more than 2 games a month on Steam.

Here is a graph of every publisher and how many games they released in 2025. There is a very tiny population that publishes a game a month or more.

The Steam upload and approval process, the 10-review system, the 7000 wishlist popular upcoming threshold, and the discovery queue all basically makes spamming out games a non-viable strategy. 

There was no AI flood that was foretold by the sages on Reddit.

The end of Covid lock-down hobby projects

Ah the Covid years when we took up weird hobbies like making sourdough or learning piano, or finally making that dream game.

The second greatest percentage increase of new games (21%) came in the year 2020. Obviously the increase can be tied to quick hobby projects that were released within the year. But, as we know, for many people, what starts out as a quick project spirals out of control and turns into a “dream game” that takes 5-6 years to finish. 

I tried to filter the data for games that have a “Coming Soon” page that have not been released. I would be interested to see if there is a bubble of unreleased games that were created in 2020. But alas that data is not available on any public system.

So we just need to guess: we are now 6 years away from the Covid times and maybe we are finally seeing the last wave of lingering Covid hobby projects. 

Steam isn’t the get-rich-quick platform anymore

This happens all the time: A new platform launches, some guy makes “Fart Simulator” that earns $50,000 a day, the mainstream press writes an article about it, and then everyone jumps in trying to make a quick buck. 

It happened when the iPhone app store launched. Then that got over-saturated, so developers moved on to Steam. There were stories of developers who made quick games that sold half a million.

Now we have Roblox. The press is now writing articles like: This Teen Build a Game on Roblox — Then Sold It for Millions or Millions are flocking to viral Roblox game created by teenager.

Steam isn’t the get-rich-quick darling anymore. You actually have to make a fun game (damn!). So now the “get rich quick” crowd is flocking to Roblox. 

There are fewer 30-year olds 

This is a real thing: when you turn 30, you are about to get married, have a kid, and you think “I gotta release a game now or I never will.” 

It is astounding the number of freshly minted 30 year olds who contact me to tell me they are finally making their dream game that they promised themselves they would make when they were 14.

I can’t make fun of them because that is also exactly what I did. My first game was released when I was 31 when I had the same revelation.

Therefore all game market projections must account for the number of pending 30 year olds.

Looking at this worldwide beehive population chart I found from 2023, it shows that there is a slight crimp in the number of people aged 20-29.  

Prepare yourselves: over the next 10 years we will face a critical shortage of dream games from 30-year-olds. 

But don’t worry, in about 20 years, we will face a new golden age of dream-games as today’s five year olds enter their 30s.

The slowing game growth on Steam could be indicating that we are entering the dream-game-dark-ages. 

Counter argument: All these factors will be canceled-out by smaller games

So there are many forces that might be reducing the number of games released over the next few years. But what if developers get wise and start making smaller games?

My dream is that indie game developers will stop trying to make their DREAM-GAMES that take 5+ years. Leave that to AAA. Instead, indies should go back to making fun, scrappy, smaller, jankier games. It is much healthier and you are more likely to succeed if you can release 10 much smaller games in 5 years than 1 DREAM GAME in the same time period.

But my ideal scenario will increase the number of games released every year. I still think we will be better off in that world.

So I guess my final message is, don’t freak out about the number of games released. It really is about whether your game has *THE MAGIC*. And the best way to find *THE MAGIC* is to release a lot of smaller projects and getting feedback faster.

To read more about the benefits of smaller games check out this blog post