The Concern Trolling Problem... Why Nintendo Gets Criticized And Epic Gets Applause
Sorry, the substack edition for this Patch Notes was stalled in my drafts. For some this might be old news.. for others.. I posted this last week on LinkedIn.
To that end…
I’m back. And I’m about to be gone again.
Spring break starts this week. Next week’s Patch Notes isn’t happening... fair warning. Two kids, an intentional plan to unplug, and a to-do list that includes finally giving Crimson Desert the three uninterrupted hours it deserves before I form any real opinion on it. I’ve been putting it off.
The May calendar is already filling up fast. LEGO Batman: Legacy of the Dark Knight drops May 22nd, which is a game I’m playing with my kids and genuinely cannot wait for. 007 First Light drops May 27th, IO Interactive’s Bond origin story, which is on my personal list. Forza Horizon 6 launches May 19th... Japan setting, 550 cars, biggest Horizon map ever... and that one I’ll be watching on Twitch. Point being, Crimson Desert needs to happen in April before the wave hits. That’s the plan.
Speaking of Crimson Desert.
Last week I made my case that the Metacritic score didn’t matter. The real signal was going to be what gamers did with the game, not what journalists wrote about it. Looks like I was right. Steam concurrent players peaked at 248,000 on March 22nd. User reviews have climbed from just under 60% positive on launch day to 75% positive and still ticking up. The game is finding its audience, and honestly, that’s a pretty great thing to see.
Simon Pulman raised a fair follow-up question, which is whether IGN specifically still matters. I don’t want to belabor this too much longer, but there’s something worth saying here that goes beyond one review score.
Organizations built on advertising need people to click through to their content. The ads live there. Keeping the lights on depends on it. As those organizations compete against an ever-expanding universe of content... YouTube channels, Twitch streams, TikTok creators, Substack newsletters, Discord servers... the pressure to generate clicks only increases.
Headlines have to work harder. They get more reductive, more provocative. More clickbaity, for lack of a better word. Not editorial drift. CPM math. The goal is to get you to the site where the ads are, and the headline is the mechanism. I’m not calling out any outlet in particular... I think this is categorically true across the board, and it’s only going to continue as these sites work harder to maintain relevance in a world where the audience has endless alternatives.
Does that mean every spicy review score is manufactured for engagement? No. Maybe Western gaming journalists broadly just weren’t the audience for a dense Korean action RPG. Maybe it’s genuinely divisive. Probably some combination of all three.
Players found it anyway. And that’s the point.
Metacritic just published their 16th Annual Game Publisher Rankings, and it’s worth pausing on for a second. Square Enix took the top spot for 2025. No shots at Square Enix... they had a genuinely strong year. But let me ask a simple question. When you think about the games that dominated player time, cultural conversation, and commercial sales in 2025, is Square Enix the first name that comes to mind? Probably not.
This isn’t new. It’s the same dynamic Simon raised, which exists across every entertainment medium. Critical admiration and commercial success are related, but they are not the same thing. Film has always had this. Music has always had this. The most critically revered album of any given year is rarely the one that went diamond. The Palme d’Or winner at Cannes is rarely the highest-grossing film of the summer. Gaming is no different.
And within gaming specifically, the correlation gets even more complicated by IP weight. Does Metacritic make Zelda sell well, or is Zelda just one of the dominant IPs in the history of the medium and it would sell well regardless? My answer is the latter, every time. Breath of the Wild scored a 97 and sold tens of millions of copies. Tears of the Kingdom scored a 96 and also sold tens of millions of copies. The score didn’t move the needle. The IP did. The hardware install base did. The 30 years of generational loyalty Nintendo has built did.
Metacritic matters at the margins. For a new IP without brand equity, without a built-in audience, a strong score can open doors and drive discovery. For an established IP with a loyal fanbase... or for a game like Crimson Desert that targets a specific audience willing to find it... the score is closer to a press release than a purchase driver.
Crimson Desert peaked at 248,000 concurrent players on Steam. User reviews are at 75% positive and climbing. Commercial hit. Metacritic mixed bag. Doesn’t matter.
What I’d worry about more, if I were the “next” Pearl Abyss , isn’t the Metacritic score at all... it’s the Steam concurrent leaderboard. Gaming is unique as a medium in that we have a real time public scoreboard. Pull up Steam or Twitch right now and see exactly what’s popular, what’s holding, what’s fading. Music doesn’t have that. Film doesn’t either. Not a live counter like that. The CCUs are the truth, and at 248,000 peak concurrent, that truth is looking pretty good for Crimson Desert.
Alright. I was fully prepared to wrap this up, take my lap, and disappear into spring break.
Then the news had other plans.
Bloomberg ran a concern-troll piece about Nintendo despite Pokopia doing embarrassingly well. Epic Games announced layoffs north of 1,000 people. End of fiscal quarter, everything drops at once, and here I am with too many topics and no business writing a long newsletter the week before vacation. When it rains, it pours. Let’s get into it.
Bloomberg vs. Nintendo... Again
What’s Old Is New Again
Bloomberg published a piece this week with a headline that probably felt fresh in their editorial meeting.
“Nintendo Switch Loses Shine With Shipments Seen Missing Target... Price cuts, new models needed to reverse trend: analysts. Stock down about 33 percent from January on dearth of games.” — Bloomberg, November 2018
Alarming, right? Except that headline is from November 2018. That’s not a typo.
The current piece reads almost identically. Production cuts. Shipments missing targets. Demand softening in the US. Analysts wringing their hands. Stock under pressure. The framing is so familiar it’s almost impressive. Bloomberg has essentially published the same concern-troll piece about Nintendo twice, seven years apart, with different bylines.
Here’s what the 2018 piece actually said, for the record:
“With few attractive titles for the holiday shopping season and shipments on track to fall short of the company’s targets, doubts are growing whether Nintendo Co.’s Switch can ever become a mass-market product.” — Bloomberg, November 2018
And then, for emphasis, an analyst quoted in that same piece:
“The Switch excitement has rapidly declined. Unless there’s significant change or something else new, the Switch story has been exhausted.” — Cornelio Ash, Analyst, William O’Neil & Co., via Bloomberg, November 2018
The Switch story was exhausted. In 2018. That Switch went on to sell 150 million units, becoming one of the best-selling consoles in the history of gaming. The analyst was wrong. The framing was wrong. The narrative was wrong. And here we are again.
I don’t know what to call this pattern other than what it is. Concern trolling. A publication with a large audience finding a data point that fits a negative narrative, packaging it as breaking analysis, and moving on before the outcome proves them wrong. By the time Nintendo hits 150 million on Switch 2, nobody is going back to pull the receipts on this piece.
So let’s pull them now, while it’s fresh.
What The Data Actually Says
Here’s the thing about the Bloomberg piece. It’s not that the facts are wrong. Nintendo did trim production. The holiday quarter did come in softer than internal targets in the US. Those things are true.
Context is where the piece falls apart.
Let’s start with Nintendo’s own words, because the Q2 earnings deck from November 2025 tells a story that is almost the exact opposite of the Bloomberg framing. At that point in the cycle, Nintendo reported net sales up 110.1% year over year. Net profit up 83.1%. Switch 2 hardware sell-in surpassing 10 million units in its first four months, the highest global sales figure for any dedicated gaming platform in Nintendo’s history in that window. Demand was running so hot in some regions that Nintendo was apologizing for supply shortages. They didn’t cut production in Q2. They expanded it.
Full-year hardware guidance went from 15 million units to 19 million units mid-cycle. A 27% increase. Companies don’t raise guidance 27% on a console that’s losing its shine.
By the Q3 report in February 2026, Nintendo’s own sell-through data showed 15 million units globally through the fourth week of December 2025. Fastest-selling dedicated gaming platform in Nintendo history. Still running well ahead of the original Switch on a time-aligned basis.
Mat Piscatella at Circana put the US picture into even sharper relief, posting the installed base growth comparison across console history after nine months in market. Switch 2 is tracking ahead of PS5 and PS4 at the same point in their lifecycles. In tracked US history going back to 1995, only one console has sold faster than Switch 2 in its first nine months. The Game Boy Advance.
That is not a console losing its shine. That is one of the greatest hardware launches in the history of the medium.
Christopher Dring at The Game Business, one of the more balanced voices covering this story, flagged something important that the Bloomberg piece glossed over entirely. The Switch 1 vs. Switch 2 comparison isn’t apples to apples.
“Switch 2 was well supplied from day one, which led to a record-breaking first few months on the market. By comparison, Switch 1 was not well supplied at launch. As a result, Switch 1 sales were more spread out across that first year as more stock came to retail, while Switch 2 sales were more front-loaded.” — Christopher Dring, The Game Business, March 2026
There’s also a calendar mismatch that makes the comparison even harder to draw cleanly. Switch 1 launched in March, meaning its first nine months didn’t include December. Switch 2 launched in June, meaning its first nine months did include Christmas. The front-loading isn’t a warning sign. It’s structural.
So what is Bloomberg actually reporting? Nintendo trimmed Q1 2026 production from a planned 6 million units down to 4 million. After expanding production aggressively in Q2 because demand was exceeding supply, they right-sized output when the holiday surge normalized. That’s not a distress signal. Any supply chain analyst would recognize it immediately.
Framing that as “Nintendo Switch loses shine” is either intellectually dishonest or deeply lazy. I’ll let you pick.
The Software Attach Rate Argument Is Also Complicated
One of the other data points floating around this week is the software attach rate. Switch 2 is running at roughly 2.2 games sold per console versus Switch 1’s 3.7 at the same point. On the surface that looks concerning.
Dig one layer deeper and it evaporates pretty quickly.
As Nintendo’s own Q2 deck showed, 84% of Switch 2 buyers through September 2025 were existing Switch 1 owners. They already had libraries. Many of their Switch 1 titles had been updated and enhanced for Switch 2. The backwards compatibility story, which Nintendo built deliberately, is actively suppressing the attach rate metric in a way that doesn’t reflect demand weakness at all. It reflects the fact that most early adopters didn’t need to buy ten new games because they already owned forty that worked.
Dring put it plainly:
“84% of Switch 2 buyers (up until September 30, 2025) were also Switch 1 owners. This means that most early adopters already had a library of games to play, and some of them had been updated and improved for the new platform.” — Christopher Dring, The Game Business, March 2026
The data is more complicated than the headline suggests. It almost always is.
Software Moves Hardware. It Always Has.
Here’s the part of the Bloomberg narrative that I keep getting stuck on, because it’s the argument that everyone in the industry nods along to without really interrogating it. The idea that Switch 2 is struggling because it lacks system-selling software.
People keep saying Nintendo’s launch lineup was weak. Let’s actually look at what launched and what it did.
Mario Kart 8 Deluxe is the best-selling Nintendo Switch game of all time at 70.59 million units on Switch alone. Not Breath of the Wild at 33.64 million. Not Super Mario Odyssey at 29.04 million. Not Tears of the Kingdom. Mario Kart. The franchise the industry perpetually underestimates as a “casual” title is Nintendo’s single biggest commercial driver in the modern era, by a significant margin. Zelda and Mario, for all their critical prestige, aren’t even close.
So when Nintendo launched Switch 2 with Mario Kart World as its flagship day-one title, that wasn’t a soft lineup. That was Nintendo leading with their actual commercial weapon. The numbers are already reflecting that. Mario Kart World sits at 14.03 million units sold as of December 31, 2025, including bundle sales, far and away the best-selling software on Switch 2. Second place is Donkey Kong Bananza at 4.25 million. That gap tells you everything about what moves the needle. Mario Kart World is tracking toward becoming one of the fastest software launches in Nintendo history, the sequel to the best-selling Switch game ever, launched on day one with a near 1:1 hardware attach rate out of the gate. That’s not a weak opening act.
The argument that Switch 2 lacks software is partially true in the window between Mario Kart World and now. Donkey Kong Bananza was a critical darling and a legitimate Game of the Year contender. Great game. But it’s Donkey Kong... not Pokémon, not Mario, not Zelda. Kirby Air Riders launched into the same year as Mario Kart World, making it a niche title by default. Metroid Prime 4 launched with more baggage than momentum after years of delays. Its mindSHARE at launch reflected that, sitting at just 0.765%.
So yes... between Mario Kart World and Pokopia, there was a gap. A real one. And that gap is exactly what the Bloomberg piece is pointing at, even if the framing is wrong. The point isn’t that Nintendo has no challenges. After almost a year, they now have their second genuine system seller on the device. And it’s Pokémon.
Pokopia Is The System Seller Nintendo Needed
Let me put the mindGAME data on the table, because this is where the Bloomberg narrative completely falls apart.
Pokopia launched March 5th. Three weeks in, it currently sits at a mindSHARE of 0.776%, ranked #18 globally, #8 on Google with 7.8 million weekly search queries, #39 on YouTube, #23 on Twitch. For a paid Switch 2 exclusive three weeks post-launch, that is an exceptionally strong hold.
The more interesting story is the curve, not just the absolute number.
In week three, Pokopia’s week-over-week decline was 22%. Here’s what the same week-three decline looked like for other major Nintendo Switch 2 titles. Pokémon Legends: Z-A dropped 46% in week three. Mario Kart World dropped 41% in week three. Donkey Kong Bananza dropped 37% in week three. Kirby Air Riders dropped 46% in week three. Pokopia dropped 22%.
The retention curve is steeper, stronger, and more sustained than anything else Nintendo has put on Switch 2 to date. A word-of-mouth driven game finding new players continuously, not burning through its launch audience and fading.
The cumulative mindSHARE picture tells the whole story. At week three post-launch, Pokopia sits at 2.899%... essentially a dead heat with Mario Kart World at 2.901% at the same point in its lifecycle. Two of Nintendo’s biggest Switch 2 titles, neck and neck on cumulative attention three weeks in. That’s not a game fading. That’s a game holding at the level of one of Nintendo’s flagship franchises.
The Switch 2 exclusive comparison sharpens the picture further. Donkey Kong Bananza sat at 1.899% at the same period. Kirby Air Riders was at 0.977%. Metroid Prime 4 was at 0.765%. Pokopia at 2.899% is running more than 50% higher than Donkey Kong Bananza, and nearly four times higher than Kirby and Metroid.
The one comp that does outpace it is Pokémon Legends: Z-A, which sat at 6.175% cumulative at the same point. But Z-A launched simultaneously on both Switch and Switch 2, tapping into a combined install base of over 170 million devices. Pokopia is a Switch 2 exclusive, full stop, competing with a fraction of that addressable market. The fact that it’s tracking neck and neck with Mario Kart World’s cumulative score while locked to a single platform with roughly 17 million units in the wild is the story. When Pokémon Winds and Waves arrives next year into a dramatically larger Switch 2 install base, these numbers are going to look very different.
Nintendo’s stock had been in consistent decline since November 2025, down over 40% from peak. Then Pokopia launched. Per Bloomberg’s own reporting, Nintendo shares gained as much as 10.5% in their steepest single-day climb since April 2025, adding roughly $14 billion in market value over the course of the week, moving the stock from ¥8,503 to ¥9,932.
“The Pokémon game was a dark horse. It was totally off people’s radar, making its popularity a positive development.” — Hideki Yasuda, Senior Analyst, Toyo Securities, via Bloomberg
“Nintendo investors were yearning for system sellers to further maintain momentum for the hardware, and Pokopia is one, even though it’s not a 3D Mario or new Zelda title.” — Serkan Toto, CEO, Kantan Games, via CNBC
A Pokémon spinoff developed by the Dragon Quest Builders 2 team moved markets, changed the hardware narrative, and drove one of the biggest single-day stock rallies Nintendo has seen in over a year. Full-year guidance of 19 million units suddenly looks very achievable. And this is before Pokopia’s impact is fully measured.
The Nintendo Hardware Debate Nobody Wants To Have
There’s a framing I’ve been sitting with for a while that I think is underappreciated in how we talk about Nintendo’s market position.
People say Nintendo isn’t the “core” gaming platform. PlayStation is the core platform. Xbox and PC are the core platforms. Nintendo is the side piece, the second console you own, the device for Mario and Pokémon and nothing else. That’s the conventional framing, and I’ve heard it in almost every gaming industry conversation I’ve had for ten years.
Flip it around though.
If almost every household that owns a PlayStation or Xbox also owns a Nintendo, then by definition Nintendo is the universal platform. Everyone owns one. The PlayStation is the thing you buy on top of that. Xbox is the thing you buy on top of that. By that logic the device with the broadest penetration, the one that crosses demographics and age groups and casual and core audiences simultaneously, is Nintendo. Everything else is the specialty purchase.
That’s not me making the case for Nintendo over PlayStation. Both businesses are real and both serve different needs. The point is that the “Nintendo is the side piece” framing dramatically undersells what Nintendo has actually built. And it’s part of why the Bloomberg concern-troll cycle keeps repeating... the publication and the analysts it quotes are measuring Nintendo by metrics designed to evaluate a different kind of platform. Nintendo’s game isn’t to win the core gamer. Nintendo’s game is to be in every home. And they’re winning that game.
The flywheel underneath all of this, which I’ve written about in detail in both the Matthew Ball attention economy piece and the Adult Money piece, is operating across theme parks, film, trading cards, merchandise, and mobile simultaneously. The Super Mario Galaxy Movie opens April 1st and is conservatively estimated to deliver $160 million over its five-day opening weekend. Fox McCloud from Star Fox just got confirmed as a character in the film, fueling the Smash Bros. crossover speculation that has the internet fired up this week. Every one of those moments is a touch point that pulls people back toward the hardware and the software without Nintendo having to buy that attention outright.
That’s the machine Bloomberg doesn’t know how to quantify.
Nintendo Is Also Playing Offense
Here’s the part the doom narrative misses entirely. Nintendo isn’t sitting still. They’re actively working the levers.
This week Nintendo announced a new digital pricing strategy. Beginning in May 2026, starting with Yoshi and the Mysterious Book, new Nintendo published digital titles exclusive to Switch 2 will carry a lower MSRP than their physical counterparts. Digital is $59.99. Physical is $69.99. Nintendo’s statement on it was direct:
“Nintendo games offer the same experiences whether in packaged or digital format, and this change simply reflects the different costs associated with producing and distributing each format and offers players more choice in how they can buy and play Nintendo games.” — Nintendo, March 2026
This matters more than it looks on the surface. As Dring noted, Nintendo’s digital mix is significantly behind PlayStation and Xbox, where the majority of new game sales happen via digital stores. A new Switch title can still see over 80% of its sales come from physical retail. Closing that gap is a meaningful margin improvement without touching hardware pricing.
“If Nintendo can accelerate its digital business, it would offset many of the margin challenges it’s facing on hardware. It could also lead to a far more lucrative generation for the company, even if Switch 2 doesn’t match the lifetime sales of Switch 1.” — Christopher Dring, The Game Business, March 2026
The early signal is already there. Pokopia’s physical copies sold out at major retailers across the US and UK. Amazon briefly raised the price from $70 to $80 due to demand. The digital version picked up the slack. That’s the exact dynamic Nintendo is trying to structurally enable going forward.
So What’s Actually True?
Nintendo has real challenges. Dring cited Circana data showing 44% of consumers are cutting back on non-essential items due to rising prices, and 46% are putting off big purchases due to economic anxiety. Rising DRAM and storage costs are putting pressure on margins. The Western release slate in the near term is thinner than Nintendo would like, with the biggest confirmed Switch 2 titles not arriving until 2027.
Those things are real. The Bloomberg piece isn’t entirely wrong. The question is whether the framing matches the reality, and the framing doesn’t.
A console that is the second fastest-selling hardware in tracked US history. The fastest-selling Nintendo platform globally ever. A Pokémon spinoff holding a 22% week-three decline when every comparable Nintendo title dropped between 37% and 46% at the same point. A game tracking neck and neck with Mario Kart World on cumulative mindSHARE three weeks in, from an exclusive position with a fraction of the addressable install base. A digital pricing strategy that directly addresses the margin challenge. A film opening next week conservatively projected at $160 million in its first five days.
That’s not a company in crisis. That’s a company navigating a complicated moment with more tools than almost anyone else in this industry has access to.
Coming back to this in six months, Pokopia is going to look like a landmark title. The curve says so. The retention says so. The install base opportunity ahead of it says so. When Pokémon Winds and Waves lands in 2027 into a Switch 2 install base that will be dramatically larger than it is today, we’ll look back at this Bloomberg piece the same way we look back at the 2018 one.
Dring closed his piece this week with a quote from Take-Two CEO Strauss Zelnick that lands exactly right:
“You’re not going to be having a benefit dinner for Nintendo anytime soon.” — Strauss Zelnick, CEO, Take-Two Interactive, via The Game Business, March 2026
History repeated itself this week. Bloomberg wrote the same article they wrote in 2018. The outcome will probably be the same too.
Nintendo will be fine.
Fortnite, The Apple War, And The Cost Of Being Right
From King Of The World To Restructuring
Let me start with the mindGAME data, because it tells a story the industry hasn’t fully reckoned with.
In October 2019, Fortnite owned 11% of global gaming mindSHARE. Number one in the world. Number one on Google. Number two on Twitch. It wasn’t just the biggest game on the planet... it was lapping the field. Nothing else was close.
Fast forward to this week. Fortnite’s global mindSHARE sits at 2.208%. Still ranked #5 globally, #6 on YouTube, #1 on Twitch. Not a dead game by any imagination. But the distance traveled from 11% to 2.208% is not a natural lifecycle curve. That’s a collapse with a very specific cause, and the timeline tells you exactly what it is.
Fortnite was hovering around 6.5% to 7% global mindSHARE consistently through 2019 and into 2020. Then at the end of August 2020, Epic got kicked off the Apple App Store. The curve broke immediately. MindSHARE dropped from that 6.5% to 7% range down to roughly 3%, then kept declining through 2021 into the 2% to 2.5% range, where it has largely stayed ever since.
There was one significant spike. December 2023, when Epic launched Fortnite OG mode, pushed mindSHARE back up to 6.176%. A second spike in the arm. Proof that the IP still had pull when given the right reason. It didn’t hold. The decline resumed. And now we’re here.
“We’re spending significantly more than we’re making, and we have to make major cuts to keep the company funded. This layoff, together with over $500 million of identified cost savings in contracting, marketing, and closing some open roles puts us in a more stable place.” — Tim Sweeney, CEO, Epic Games, March 24, 2026
This is Epic’s second major round of layoffs in three years. The first was 830 jobs in September 2023, when Sweeney used almost identical language... the company was “spending way more money than we earn.” Two rounds, two years apart, two versions of the same sentence. Over 1,800 people gone across both, on the back of a game that by every external metric is still one of the most successful live service titles ever created.
A Billion iPhone Users And Five Years Gone
Here’s where I’m going to say something that will get me some pushback. Going to say it anyway.
Tim Sweeney went to war with Apple. The industry applauded him. Fortnite was the sacrifice.
To be precise about what I’m saying here... I’m not arguing the fight was wrong. The case that Apple’s App Store fees and restrictions represent anticompetitive behavior is legitimate, and Sweeney pursued it with real conviction and real resources. Developers cheered. The industry cheered. There was genuine altruism in the mission, at least partially.
The mechanics of how it started are worth revisiting. On August 13, 2020, Epic implemented its own in-app payment system inside Fortnite on iOS, bypassing Apple’s standard 30% fee and offering players cheaper V-Bucks as a result. Apple removed Fortnite from the App Store the same day. Epic had clearly anticipated this... within hours they filed an antitrust lawsuit and released a parody video mocking Apple’s iconic 1984 ad, calling on fans to #FreeFortnite. As The Verge reported at the time, this was a carefully calculated series of responses, not a reactive one. Epic picked this fight deliberately, on purpose, with a plan.
“Epic enabled a feature in its app which was not reviewed or approved by Apple, and they did so with the express intent of violating the App Store guidelines.” — Apple, statement to The Verge, August 13, 2020
The removal cut off Fortnite from access to more than a billion iPhone and iPad customers, per Bloomberg. It stayed that way for nearly five years.
The commercial context matters here. Fortnite earned Epic $2.4 billion in 2018 and $1.8 billion in 2019, driven significantly by its cross-platform popularity including iOS. That was the revenue baseline before the war started. Comparing that to where they are today... laying off staff twice in three years because costs exceed revenue on a game that is still number one in monthly active users... is the story nobody wants to tell plainly.
Fortnite didn’t return to the US App Store until May 20, 2025, and only after a court ordered Apple to comply. As Bloomberg reported, Epic had to ask a federal judge to force Apple to approve the app, which had been submitted for review weeks earlier and sat untouched far beyond Apple’s standard 24-hour turnaround. The game came back. But the world it returned to had fundamentally changed.
Apple’s App Tracking Transparency framework, which rolled out in 2021 while Epic was in court, had restructured mobile user acquisition economics entirely. The targeting that made mobile marketing efficient before ATT was gone. Re-acquiring a lapsed audience on mobile had become dramatically more expensive than retaining one. Epic walked back into the iOS ecosystem at exactly the wrong moment to try to rebuild what they’d lost. The audience had moved on. The economics of reaching them had gotten worse. Five years of momentum had been handed to someone else.
The Vacuum Epic Created And Roblox Filled
Here’s the part of this story that doesn’t get connected often enough.
In our mindGAME data, Roblox’s rise as a dominant global platform started in August 2020. The exact same moment Epic got kicked off the App Store.
Not a coincidence. A market.
Fortnite had built something extraordinary... a platform-scale game that captured the attention of young players across mobile, console, and PC simultaneously. When it vacated the mobile ecosystem, those players didn’t stop gaming. They found somewhere else to be. Roblox was available everywhere, free, and already building the kind of user-generated content engine that made it stickier with every passing month.
Roblox peaked last summer at a global mindSHARE score of 9.01%. From roughly the same starting point in 2020, it went from parity with a declining Fortnite to nearly five times Fortnite’s current mindSHARE. As I’ve written about in the Matthew Ball attention economy piece, the rise of Roblox and the decline of Fortnite as a cultural force are inherently linked in the data. One filled the vacuum the other created.
Epic eventually recognized this and started chasing Roblox directly... UEFN, the Fortnite creative tools, the metaverse positioning. All of it designed to recapture the platform-scale audience they’d ceded. Roblox had a five-year head start on that audience by the time Epic got serious about competing for it. The ship hadn’t just sailed. It was over the horizon.
The engagement picture Circana is painting right now reflects all of this. Fortnite is still #1 in monthly active users on both PlayStation and Xbox in the US. The average PlayStation player spent 16 hours on Fortnite in February 2026 though, down from 21 hours in February 2025. Xbox players averaged 15 hours versus 19 the prior year. Still number one. Losing hours. That’s a ceiling problem, not a collapse... and it is extraordinarily expensive to run a live service game at that scale when the engagement curve is moving the wrong direction.
Piscatella at Circana framed the broader stakes plainly:
“The big live-service behemoths (including Fortnite) take up significant player count and time-share on the consoles. The top 10 live-service games on PlayStation and Xbox comprise nearly half of all US gaming hours on the platforms every month. And of course, this raises the question of if Fortnite can’t make it, what chance do other games have? I do not have a good answer to that question. Heck, I don’t even have a bad answer.” — Mat Piscatella, Analyst, Circana, via Polygon, March 2026
That last line is the one that should be keeping people up at night. Not because Fortnite is dying. Because if the most successful live service game in history is structurally challenged at this scale, the math for everything below it gets very uncomfortable very fast.
The Epic Games Store: Seven Years In
While all of this was unfolding, Epic was also fighting a second front. The Epic Games Store, launched in December 2018, was supposed to challenge Steam’s dominance of PC digital distribution. After seven years the results are... complicated.
The store generated $1.16 billion in revenue in 2025. That sounds meaningful until you put it next to Steam, which generated roughly $10.8 billion in 2024. Epic holds somewhere between 3% and 7.5% of the PC digital distribution market depending on how you measure it. Steam holds 74-75%.
The more telling number is third-party game spending... the metric that tells you whether the store is working as a real marketplace or just a Fortnite launcher. Third-party spending fell 18% in 2024 to $250 million before bouncing back 57% in 2025 to $400 million. The overwhelming majority of Epic Games Store revenue still comes from Epic’s own titles. Fortnite. Rocket League. Fall Guys. Not the store working. The store surviving on the back of the same IP that’s now in structural decline.
After seven years, the Epic Games Store has not meaningfully challenged Steam’s position. It has offered developers a better revenue split, genuinely good for the industry, and something worth crediting. As a consumer platform though, competing for the player relationship against Steam, it has not worked the way Epic needed it to. Not a colossal failure. A middling result at enormous cost, funded primarily by the product that’s now hemorrhaging engagement.
Epic is not going to beat Steam. That window has closed.
The Iwata Contrast
Something has been sitting with me all week, because I think it reframes the whole conversation.
During Nintendo’s darkest period... the Wii U era, when the company was posting its first operating losses in decades... CEO Satoru Iwata made a deliberate choice. He refused to lay off staff. His reasoning was direct:
“If we reduce the number of employees for better short-term financial results, employee morale will decrease, and I sincerely doubt employees who fear that they may be laid off will be able to develop software titles that could impress people around the world.” — Satoru Iwata, CEO, Nintendo, 2013
Instead of cutting his way to stability, Iwata cut his own salary. And the salaries of his senior leadership team. Then he and his team built the Switch. Nintendo went from its worst hardware failure in decades to the best-selling console in the company’s history, without sacrificing the people who made it possible.
Epic has now laid off over 1,800 people in three years. Both times citing the same core problem. Both times on the back of a game that is still, by every external metric, one of the most successful live service titles ever made. And both times while the industry largely nodded along, focused on the nobility of the Apple fight rather than the consequences of how it was fought and what it cost.
There are choices and there are consequences to those choices. Sweeney made a calculated decision to go to war with Apple. The industry applauded him for it. Fair enough. But the consequences of that war landed on Fortnite’s engagement curve, on the mobile audience that migrated to Roblox, on the Epic Games Store that never became what it needed to be, and ultimately on 1,800 people who had nothing to do with making that call.
I’m not here to say the Apple fight was wrong. Whether it was the right call is genuinely an open question and reasonable people can disagree. What I am saying is that the industry has spent considerable energy concern-trolling Nintendo over a production trim on the second fastest-selling console in US tracked history... while largely giving Epic a standing ovation for a decade of strategic decisions whose costs keep getting passed to the people least responsible for making them.
Nintendo gets scrutinized for managing a complicated moment exceptionally well. Epic gets celebrated for a war whose casualties include its own employees, its own platform, and a generation of players it handed to a competitor.
Maybe we’ve got the concern trolling pointed in the wrong direction.