This week kept circling the same question from different directions: once a character is out in the world, who actually gets to shape what it becomes? A restructure at Microsoft moved development of the next Halo to a team at Activision and cost a lot of people at Halo Studios their jobs along the way. Sega's head of Sonic Team admitted the series came close to being cancelled before a movie brought it back. And a longtime games marketer made the case in Aftermath that Fortnite is doing real damage to the characters it borrows, which is a take I understand but, frankly, see almost the opposite way.

Today: what the Halo restructure puts at risk, why I think Aftermath has the Fortnite question backwards, and why the loudest F1 viewership number this year is the least useful one.

Microsoft's latest round of cuts reached Halo Studios, and as part of the restructure, the next mainline Halo is moving to a team at Activision.

Another 268 roles were cut across Xbox Game Studios this week, Halo Studios among them, on top of the up to 3,200 reductions Xbox CEO Asha Sharma announced in July. The wider reorganization folds Obsidian into Bethesda, merges Turn 10 with Playground Games, and reportedly puts Ninja Theory at risk of closure. Xbox says it's about 75% of the way through.

As part of it, a purpose-built team at Activision will develop the next mainline Halo, and a smaller group stays in Redmond to support the current games, the community, and esports. It's hard to read that as anything other than a decision driven by cost and consolidation, and on the ground it means a lot of the people who kept Halo going for years aren't there anymore.

The IP doesn't go anywhere in a restructure like this. The people, and everything they knew about the franchise and its community, are a different story, and the hardest to replace. I wrote a few weeks back about Hollywood licensing game IP from studios that had been rebuilt since the last game shipped, with the Halo show as a cautionary tale about what gets lost in the handoff. Something similar is happening here, just inside one company. Activision clearly has the scale to make a big game (Call of Duty has grossed north of $30 billion over its life). Whether the next Halo feels like Halo to people who've played it for 25 years is something nobody can answer for a few years, and I'd rather not pretend otherwise.

Receipt: The Verge / Xbox News / Kotaku

Sega almost cancelled Sonic until a movie brought the character back, and if you still judge game IP on the last game's sales, that should give you pause.

Sonic Team head Takashi Iizuka said around the series' 35th anniversary that Sega once seriously considered ending the franchise, and that the 2020 film is what changed the plan. That first movie did around $320 million worldwide on a reported $85 million budget. Sonic 2 did roughly $405 million, Sonic 3 went past that, Knuckles got his own Paramount+ series, and a fourth movie is set for March 2027. All told, the films have cleared well over $1 billion for a character Sega was ready to put on the shelf.

The Shadow story is my favorite part of this. Sega had planned to drop him from the series entirely, then reversed course after Keanu Reeves voiced him in Sonic 3 and he caught on with a whole audience that had never touched a Shadow game. The movie created the demand and the game side followed it, which is pretty much the reverse of how this used to work.

So if you're sitting on game IP and the only number anyone looks at is how the last release sold, you're measuring one channel. It doesn't tell you much about whether people still care about the character, and Sonic turned out to be an expensive lesson in the difference.

Receipt: Dexerto

Reebok is quietly playing the long game, sticking with a handful of silhouettes while most of the industry chases whatever's hot this month.

The revived G-Unit G6 and the BBC Ice Cream Board Flip both sold out, so nostalgia with the right co-sign clearly still moves product. The one I'd pay more attention to, though, is the Freestyle Lo with Celine. A luxury house picking Reebok as a silhouette partner puts the brand in a conversation about taste rather than performance, and that's a lane you can't really buy your way into.

Nike's the obvious comparison, and it's a tricky one to make fairly. Nike has deeper roots in basically every category, but it's also dealing with oversaturation, a revenue slide, and some rebuilding with wholesale partners (DSW and Macy's among them) after its direct-to-consumer push. Reebok doesn't have that reach, but it isn't carrying that baggage right now either, and there's something to being the brand people pick on purpose rather than by default.

I wouldn't overread it, though. Limited runs selling out is a small-batch signal, and a couple of sold-out drops aren't necessarily a turnaround. The real test is whether Reebok can keep turning down the volume long enough for these shoes to become the brand.

Receipt: Highsnobiety

Aftermath argues Fortnite is hollowing out the characters it borrows. I work in the same field as the writer, and I land almost the opposite way.

The piece is from Sam Greszes, who's spent more than 15 years in games marketing and branding, and it's a thoughtful read. His argument is that Fortnite's crossovers break the rules that make characters recognizable in the first place. He points to the old Simpsons style guides that kept every character on-model no matter who was drawing them, and says Fortnite throws all of that out: Ben 10 with a real gun, Kairi from Kingdom Hearts holding a shotgun, Roxas dabbing on the Organization. In his view, the characters get reduced to empty skins sold back to fans as an ad for spending V-Bucks, and control ends up with Epic instead of the people who created them.

I don't think Sam is wrong about everything. The value of a Fortnite collab really does depend on the brand, and Greszes says as much. Nintendo and Pokémon staying out makes complete sense for companies that guard their characters that closely, and I'd give them the same advice. Where we split is on the starting assumption. The style-guide model he's describing comes from an era when the company decided how a character showed up, full stop. That hasn't been true for a long time. Fans have been remixing these characters through memes, fan art, mods, and edits for years, whether anyone signed off or not. Fortnite didn't take control away from the creators. It gave fans an official, somewhat managed place to do what they were already doing, and gave the IP owners a cut of it.

The strongest evidence for that is sitting in his own examples. Sega didn't just hand Sonic over. The Sonic skins are canonically androids because, as Iizuka put it, "Sonic will never hold a gun." The Keyblade pickaxe swaps Mickey's charm for a llama head. That's the style guide still working, just adapted for a new kind of venue. The brands that care about canon are setting guardrails and then choosing to show up anyway, because trading a sliver of control for a spot in front of a huge, young audience is a deal they've decided is worth making.

There's also a version of this that's good for the characters, and this same issue has the proof. Shadow found an entirely new audience through a movie, and Sega changed its plans because of it. Plenty of kids are meeting Sora for the first time in Fortnite rather than on a PS2, and some of them will go looking for the games. Greszes sees a character getting hollowed out but I'd argue a lot of those kids are seeing the character for the first time, and that's how a franchise keeps going for another 20 years.

I'll give him this much: there's a real ceiling on it. If a collab is purely dropping a skin with no real thought behind it (he calls out the Screamer skin, and I'd have a hard time defending that one too), then it probably is just an ad for V-Bucks. The collabs that hold up are the ones that feel like they belong in the world. For anyone weighing a Fortnite partnership, that's the actual question: is there a real reason for your character to be there, and have you decided up front what they will and won't do? Get those two answers right and the canon mostly takes care of itself.

Receipt: Aftermath

PlayStation's collab with LISA is a bet on reaching people who don't think of themselves as gamers, and a limited-edition controller (as part of it) turns out to be a pretty smart way to do it.

The collab is a limited-edition DualSense plus a merch line, dropping in October. LISA has more than 100 million Instagram followers and a fanbase that over-indexes in Southeast Asia, East Asia, and the Americas, which also happen to be the regions where PlayStation is working hardest against mobile and the Switch 2.

Her fans have a history of showing up the minute something drops. When LALISA came out in 2021, the video pulled 73.6 million YouTube views in its first 24 hours. That's the kind of day-one energy PlayStation is borrowing here.

I like the format choice, too. A controller like this sits on a desk or a shelf and ends up as a collector’s item, so it keeps getting seen well after it sells out. That's a lot closer to how streetwear works than how gaming hardware usually does, and I'd expect more of these to come from music and fashion names rather than gaming personalities.

The part I'm most curious about is whether it actually sells consoles or even games. A LISA fan picking up a controller isn't automatically a PS5 buyer. If it mostly moves the collaboration items, it's still a great brand moment, just a smaller one than the "new audience" framing suggests.

Receipt: PlayStation Blog / Hypebeast / Polygon

Total hours watched for F1 on Apple TV in the US, year-over-year and season-to-date, according to Liberty Media president and CEO Derek Chang on the company's Q2 earnings call. He also said overall viewership is up and that the Apple audience skews younger and more female.

If you've followed F1 coverage this year, you've probably seen the other side of this. Samba, which estimates viewing through a smart TV panel and isn't an official ratings source for F1 or Apple, put household reach for the Miami Grand Prix down 68% from 2025 and Monaco down 66%. F1 pushes back hard on those numbers and says it has no relationship with Samba.

I think both numbers are probably pointing at something real. Moving races from ESPN and ABC to a paid subscription almost certainly cost some casual reach. Hours watched still went up, though, and for anyone buying into the sport that's the more interesting number, because it tells you something about who stuck around.

Somebody who signs up for Apple TV to follow a season of Grands Prix made a choice to be there. Drive to Survive built F1's American audience wide but pretty casual, and even that show is cooling off (total views were down more than 10% for its 2025 season). What's left once a sport goes behind a subscription is a smaller group but a much more committed one, and that's the viewer sponsors should want.

It's already starting to turn, too. Samba's own estimate for the Italian Grand Prix at Monza had reach up 15% on last year and average viewing time up 78%, and Miami drew a record 275,000 people in person.

I'd still be careful reading too much into one season of any rights move, and the people closest to it are saying the same. The better signal will be 2027 renewals, meaning whether the people who subscribed for F1 come back for year two. In the meantime, if you're planning an F1 sponsorship for next year, push for hours watched and time spent, not just reach. And keep an eye on price. Sponsorship revenue is on track to top $3 billion this year, and at least one sports marketer is already calling the market frothy, which is usually right about when brands start overpaying.

Receipt: Digiday

Forward this to whoever still has gaming down as a "niche vertical" in their Q4 plan. Sega came close to shelving a character that's since made over a billion dollars at the box office, which should be enough to reopen that conversation before the budget's locked.

Tapped in. In case you aren't.