“Real games” (AAA, top tier indie, low tier indie) can easily be as low as 1-2% of “steam releases”. 12k games were published on Steam over half of this year. That would mean, if you want to care about this meaningless statistic of “x% of games drive y% of revenue”, that there’s room for over 240 “worthwhile” games to have released in 2026 without disrupting the scary 1% metric at all. Two hundred forty games this year. AAA releases are what, ten or twenty a year? Less? That’s not 1%, it’s 0.0004%. The other 99.9996% of the 1% are free to be indie games or GAAS or whatever else.

Slay the Spire and lower quality but still “real” indie games. That entire gamut. The best games ever made and “uhh it’s balatro but with throwing darts”. That’s all more numerous than AAA stuff and also dwarfed in volume by porn jigsaw puzzles, asset flips, and other actual zero effort slop “games”. AAA and “real indie games” are a tiny minority of the games on steam.

Twelve thousand “games” have been released on Steam this year as of July. That would mean the meaningless statistic in the article implies that 240 games from this year drive 85% of this year’s spending. 240 games in 2026 “worth spending money on” is not a dire portrait of AAA slurping up the market space. That “1% of games” is presented to look like “the top 1% fat cat games” when 1% absolutely includes not just the AAA and the top shelf indie (like Slay the Spire) but also the flashes in the pan and the trend chasers and the friendslop and the Balatro clones and the forever GAAS.

Those “not truly slop but” uninspired indie games like stardew clone #780 comfortably fit into the “top” 240 Steam games this year. Darkest Dungeon 2 is absolutely a “top percent” game when the overall pool includes so much actual nothing in it.

Well that’s what I mean tho. If DBFZ reaches a million players in year 1, Bamco gets sixty million dollars (pre taxes and fees and licensing yadda yadda). If 2XKO reaches a million players, Riot gets zero dollars unless you also got some of that million to buy hats. A guy who buys DBFZ and plays for a week before dropping it got you sixty bucks. Same dude in 2XKO that doesn’t buy a $20 hat gets you zilch.

This one was f2p, so it doesn’t matter how many copies it “sold”. It needed dedicated long term players that would shell out for hats.

The casual fighting games market does exist, but it’s going to be people that buy the fighting game for an IP they like (say, Dragon Ball Z) and play it for a couple months then put it on the shelf. Maybe they boot it back up (and pay $10 again) when the DLC adds Super Saiyan 4 Gogeta or whatever.

2XKO was never a game that could make money from casuals, because there was no reason for anyone except “fgc mechanics heads” to give a single shit about the game. And those fgc heads saw a 2v2 tag fighter with ten characters and responded with a resounding “lmao”. Doesn’t help that it’s a fighting game with kernel anticheat.

Apple’s customers don’t get a choice of stores in the first place.

And the smallest in volume. The number of games on Steam that cost a million dollars or more to make (owned by shareholder value generators) is much lower than other types. For every one Assassin’s Creed there are ten Peaks and Slay the Spires and lower quality “real, good indie games”. And for every real, good indie game there’s a hundred asset flips and ai slop “games”.

And what percent of the “games that barely make anything” are asset flips and zero-effort slop? Like, “percent of new games that are trash” is going to be very high when the barrier to entry is “pay Valve $100 once”. This is a total junk statistic.

Like, store A where you can buy 100 good games and 1000 asset flips will score 10x better on this un-metric than store B where you can buy 10000 good games and a million asset flips. And yet nobody would ever argue that store A is a better store.

Piracy is the most consistent and solid form of preservation, definitely. Well technically piracy is a subset of “DRM-free software” which is the true holy grail (all the advantages of piracy, with total legality and legitimacy!). To preserve software, it must be able to be copied instead of tied interminably to something perishable like a disc or single hard drive. Or, for that matter, a single online account that might suffer anything from a banning to deletion for inactivity to just “lost password and forgot the email and can’t get access anymore” to “the company that operates the servers the account uses no longer exists”. And piracy needs to be happening during the release windows of the software, not “be respectful and wait 5 years” - the only reason retro console libraries are as well-kept as they are is because people were dumping games to play for free back in the day.

Those Wii redownloads are tied to specific console IDs, and those console NANDs are just as vulnerable to damage or deletion or loss as normal physical media is. They’re even worse than account-based DRM’d media since account-based media at least tolerates hardware damage or loss (but not account bans or service EOL). Thankfully, every digital purchase on Nintendo platforms other than Switch 2 is fully preserved and made available. This would be impossible if Nintendo’s anti-piracy measures had ever worked for an entire console generation. Switch 2 games, on the other hand, are currently “loseable”. You might be reasonably sure that Nintendo will never delist DK Bananza until (insert threshold where you think it’d be acceptable here) but game preservationists also care about game versions. And it’s always possible that Nintendo finally managed to nail it and piracy for Switch 2 won’t happen ever (or for a sufficient definition of “ever”).

Game key cards (and discs for update-reliant or online-focused games) are in a weird place. The main reasons people claim to care about physical games are “preservation” and “used games” and those don’t actually overlap.

From the preservation standpoint, GCKs and discs that don’t have the whole game are both pretty shit. There’s an argument that buggy broken v1.0 is better than nothing and that’s objectively true, but neither of these things provide actual software preservation in the face of (for example) storefront closure or account banning. GCKs and all discs are useless for preservation of online games, of course: what are you doing with that Highguard or Concord disc? A further argument is that even a complete game burned to a disc that relies on DRM isn’t truly preserved as it won’t outlive the death of the DRM boxes themselves, when every Nintendo 3DS dies you’re not going to be able to play that pristine physical copy of Pokémon Moon. Used physical games themselves are also a physical product that can decay or be damaged or lost - you can’t say that a game with a million extant discs being traded around between various people is a “preserved game”.

A completely different appeal for physical games is simply the ability to resell. Your gaming budget goes a longer way if instead of spending $60 three times to play three games, you’re spending $45 three times and also selling the games for $30 after you’re done with them. You could play ten first-party single player Nintendo games legally on the Switch for $0 if you always bought used and sold the used copies for the same price you bought them for. Game-Key cards and discs for incomplete/online-only games do facilitate this perfectly, though GCKs lose all value when the CDN of the storefront closes and discs only lose most value in this situation.