Nobody Is Signing It

Sit in a coffee shop on Sukhumvit Road in Bangkok for six hours and count the people wearing a new NBA pro model, or the latest performance running shoe from Nike or Adidas.

Now wait. I waited. My count was zero.

What I lost count of were Nike Dunks and Adidas Sambas. Two archive models, one from each brand, both older than nearly everyone wearing them.

That gap is the entire subject. The Dunk was a basketball shoe. The Samba was built for footballers training on frozen pitches, in 1950. Both survived because neither looks like what it was made for. They look like shoes. Unremarkable proportions, nothing announcing an athletic function to anyone who walks past. That is why they are in the canon and why a current pro model will not be.

The left handcuff

What Nike and Adidas do not say, but plainly act on, is that their largest source of marketing capital has not been a new signature athlete or a diffusion line in years. It is nostalgia, and the archive is where nostalgia is stored.

That makes the archive the most valuable thing either company owns. It is also completely immovable.

You can remix an Air Jordan 1, and people do it constantly. Colorways, materials, small alterations, and a handful of those have become canonical in their own right.

What almost nobody clears is the higher bar, which is changing the shoe structurally and having the result enter the archive beside the original rather than beneath it. That has essentially happened once, when Virgil Abloh took it apart for The Ten and handed it back with the construction exposed. Below that bar, the value lies in retaining familiarity, which is a polite way of saying do not touch it.

So the asset appreciates only as long as nobody meaningfully changes it. That is a strange thing to own. It generates enormous revenue and permits almost no development.

It is also depletable, which is what turns a constraint into a countdown.

Every revival saturates the model it revives, and the sequence is a matter of record. Stan Smith and Superstar, then the Dunk, then the Samba and the Gazelle, then the Club C. Each one flushed into the market to fuel a season. Each one reaching the point where the joke about it became more visible than the shoe. Each one then needing most of a decade of rest before it could be used again.

None of this carries clean dates, and that is not evasion. Fashion moves at different speeds in different places, and New York runs a year or two ahead of the Midwest on any given cycle. Which is also its own mechanism, because a model can be exhausted in one market while still producing in another, and a brand can keep extracting from it long after the people who started it have gone.

The Stan Smith is the precedent, because it completed the whole arc in public. Revived, saturated, memed, and then deliberately pulled by Adidas so it could recover. That is a company withdrawing its own best-selling shoe from a market that still wanted it, which only makes sense if you understand the asset is consumable.

The Dunk is the diagnostic case, and its mechanism is not the Stan Smith's.

The SB Dunk has had a following for twenty years and still has one. That following runs on thematic literacy, on knowing what a release refers to and why a colorway exists, the Koston Thai temple shoes, the What The Dunk and everything it was assembling. Every release was a reading test and the people who cared could pass it. I was one of them.

What Nike did was flood the market with general release Dunks alongside all of that. Same silhouette, enormous quantities, no thematic content in any of it. That did not saturate the SB following, which is largely intact. It saturated the shoe. The public Dunk became a shape with nothing inside it, available everywhere, and the model went from something you had to understand to something you simply saw. Nike has since listed it among its oversupplied franchises, which is the accounting term for what happened.

SB is still cool. It is also not the same, because the silhouette it works in now carries years of meaninglessness attached to it.

That is worse than ordinary depletion. The Stan Smith was exhausted by selling too much of the real thing. The Dunk was exhausted by selling a version of it that contained no knowledge, which does not merely consume demand. It degrades what the object means.

Saturation does not look like a shoe becoming popular with a group. It looks like a shoe becoming unobjectionable to everybody at once. By the time the Samba wave landed on fashion-conscious twenty-year-olds and on their fifty-year-old mothers in the same season, this was no longer a hypothesis about where the strategy leads. It was the fourth or fifth run of a visible playbook, and a model that has reached everyone at once has nowhere further to go and nobody left to convert.

Run that season after season, year over year, and two things go wrong at once.

The cultural cache thins. A revival works because it feels like a rediscovery, and the fourth consecutive one does not feel like anything, because the audience has learned the schedule. People can name the next model before it is announced. Once the mechanism is visible it stops producing the feeling it was extracting, which is the whole reason it was worth doing.

And the inventory shrinks. There is a finite number of silhouettes carrying enough stored meaning to be worth reviving, the rest period runs close to a decade, and the cycle has been accelerating. You can watch this in real time by noticing how far back and how far down the brands are now reaching, into terrace models and forgotten runners and shoes that were not significant the first time around. That is what running out looks like in practice. Not an empty vault, but progressively weaker withdrawals.

Then there is the part with no answer to it. You can spend an archive and you cannot deposit into one, because nothing becomes canonical by being designed to be. There is no meeting where Nike decides that a 2026 model will be archive material in 2050. That status is conferred by time and use, long after the fact, and the brand is not the party conferring it.

The archive is also, and this is the part people skip, a record of a specific window. Those shoes were designed when performance requirements were loose enough that a designer could satisfy them and still make something that read as ordinary footwear. The technology was air in a bag and a rubber outsole. You could hit the spec and still make a shoe that looked like a shoe.

That window closed. It cannot be reopened, only reprinted.

The right handcuff

Modern performance dictates form, and it dictates it hard.

Carbon plates need a geometry. Foam stacks need a height. Lockdown systems need structure in specific places, and the current understanding of energy return produces silhouettes determined by materials science rather than by anyone's eye. These are extraordinary machines. They are also shapes that arrived by calculation, and they read exclusively as equipment.

Which is why the pro model test works. Those shoes are aesthetically legible only in the context they were built for. Put one under a pair of jeans and it announces that you are cosplaying an athletic activity you are not currently doing, and everyone can see it.

Running shoes are the partial exception, and they are a genuine exception. But running crossed into casual wear a long time ago and on its own terms, and it left this conversation with it.

So the incumbents are pinned. On the left, an archive that pays well and cannot be developed. On the right, performance categories where aesthetics are a downstream consideration and the shape is decided before a designer gets an opinion.

The middle, which is casual sneaker design that is neither a reprint nor a machine, is largely vacated.

The incumbents have said so themselves

This is where the argument stops being an aesthetic complaint and becomes a matter of record.

Start with Adidas, because their numbers show the archive working before they show it failing. Analysts put the terrace franchises, meaning the Samba and Gazelle and Campus, at around 1.5 billion euros in annual sales, roughly seven percent of company revenue. For scale, that is close to what Yeezy produced at its peak. An entire growth story assembled from shoes designed between 1950 and 1991.

The Samba's trajectory is the part worth sitting with. By Bjorn Gulden's own account it went from a couple of hundred thousand pairs a month to millions of pairs a month. That is not a brand discovering demand. That is a company opening a valve.

And then closing it, deliberately, which is the evidence that matters most.

Gulden on the Superstar, the model queued up behind the Samba: "We will not push it globally in huge volumes. We will hold it back and let the consumer decide when he or she wants the Superstar." And on the sequencing: "We will maintain the current franchises, and then time the activation of Superstar as we and the retailers need it."

Read that once and it sounds like restraint. Read it again and notice who is deciding.

Let the consumer decide when he or she wants the Superstar. That is a company saying it no longer believes it can put a shoe into culture and have culture accept it. It can wait for signs that the market is ready and then supply carefully, and that is the whole of the plan, because the alternative is trotting out a classic and having the market answer try again, in public, with the inventory already made.

Brands used to create demand. That sentence describes detecting it.

One analyst put the risk in plainer terms, that distributing tens of millions of Samba in a single year would sell them and would also end the Samba dynamic. But the deeper problem is the one Gulden states himself. A company that can only detect demand cannot originate canon, because originating anything means putting out something nobody asked for and being right about it later.

They know because they have done it before. When the Stan Smith fell out of fashion in 2018, it took a meaningful piece of Adidas European sales down with it. That is the precedent, priced.

Nike is further along the same curve and did not manage it as carefully.

Their classic footwear declined by more than thirty percent in a single quarter last fiscal year, producing close to a billion dollars of revenue headwind across the year. The company called the reset painful in the short term, which it is, and necessary, which it also is.

Then the most recent quarter. Nike's revenue fell nine percent to $11.3 billion. Direct-to-consumer dropped twelve percent, digital fifteen. The stock hit a twelve-year low and roughly sixty billion dollars of market value went with it, with Jordan Brand the heaviest single drag.

The announced remedy is the interesting part. Nike is rightsizing Air Force 1, Air Jordan 1 and Dunk, cutting those three franchises by ten percentage points of footwear mix. The company has looked at the archive dependency, measured it, called it oversupply, and decided to reduce it. Oversupply is the polite word for burned.

Reduce it to make room for what, though.

The CEO said the beloved silhouettes with strong fandoms will stay, and that Nike is building a supporting cast around them. Read that phrase twice. It concedes that the archive cannot carry the business and simultaneously declines to replace it as the lead. A supporting cast implies the star remains, and the star is a shoe from 1985.

And the innovation announced alongside it is a partnership with Skims, which is Kim Kardashian's womenswear athleisure line. That is not an adjacent category to casual sneaker design. It is a different aisle of a different store. Whatever the commercial logic, it is not an attempt to make a new shoe that people will care about, and its presence in the same conversation tells you exactly where the company believes its problem lives.

The Vomero 18 belongs here, because it gets read as a counterexample and is closer to a confirmation. Nike put out a running shoe that reportedly cleared a hundred million dollars inside ninety days, and told everybody about it.

Sit with the fact that this is announceable. A hit at that scale used to be an ordinary quarter. Publicizing one is the tell that they have become rare enough to count as news.

It also landed in performance, the lane where shape follows a spec and the product is judged on whether it works. So demand still exists and the brands know it. What they have lost is the ability to produce it on purpose. Every swing is a gamble now in a way it was not, which is exactly why one connecting earns a press cycle.

The obvious objection

Someone will say On and Hoka, and they are right that those brands are taking share.

But look at what they are actually producing. Genuinely innovative footwear, iterated hard, replaced by next year's model, and then next year's after that. They are exceptionally good at occupying the present.

None of it is canon material and none of it is trying to be. These are shoes people buy because of what they do, and the moment a better-performing version exists the previous one is obsolete rather than historic. Performance products have a shelf life by construction. Nobody is going to be paying a premium for a 2026 Hoka in 2050, because the entire proposition will have been superseded twice over.

They are in consumers' minds. They are not in their hearts. Those are different organs and only one of them builds an archive.

Where the attempts are actually happening

I spent months walking around Bangkok and moving through Asia, and a good deal of that was fieldwork whether or not I called it that at the time.

What is visible there is a market far more fragmented than the view from a Western retail floor suggests, and a volume of footwear design experimentation happening at manufacturer level that has no equivalent anywhere in the branded world. Factories that spent decades making other companies' shoes are now making their own, and selling them directly through marketplaces.

The capability was never the missing piece. Those factories always could design; they were executing other people's briefs inside other people's brand guidelines. What changed is the channel. Previously the only unsanctioned use of that capability was counterfeiting. Now there is a route to a customer that does not require anyone's permission, and what is coming through it is not fakes. It is new, in the narrow sense that nobody has made it before.

Are the shoes good? Frequently not. That is not a concession and I want to be clear about why.

Canon is a survivorship phenomenon. It requires thousands of unconstrained attempts for a handful to still matter in fifty years, and it is never obvious in advance which ones those will be. Nobody sets out to design a canonical object. The Air Force 1 was a basketball shoe that got discontinued. Most of what came out of those Harajuku back streets in the nineties was forgettable and has been forgotten, and the survivors are the only part anyone can name now.

Junk at volume is what the front end of a canon looks like. It always has. So the relevant question is not whether the output is good. It is who is attempting at volume without a constraint, and right now the answer is a few thousand anonymous factories and essentially nobody else.

Except

Here is where my own argument stops working, and I would rather say so than route around it.

Temu is not a brand. It is a marketplace, and the manufacturers on it are faceless. There is no canon-level name anywhere in that system, and there is no mechanism inside a marketplace listing that would produce one.

Because canon does not run on objects alone. It runs on provenance. Who made this, when, in response to what. That is what allows meaning to accumulate on a physical thing across decades, and it is why the shoes people still care about are attached to identifiable people and specific places and datable decisions. An anonymous listing cannot hold a story. There is nothing for the story to be about.

So the factories cannot produce canon either. They have the volume and no name.

The actual diagnosis

Canon formation needs two inputs. Unconstrained attempts at scale, and attribution. Both, at once, in the same place.

Those two have come apart, and every current participant holds exactly one of them.

The incumbents have attribution and enormous cultural credit and have publicly stopped attempting, cutting their most attempted-at silhouettes by ten points of mix while announcing a womenswear collaboration as the growth story.

The performance brands have attribution and attempt relentlessly, inside a constraint that forecloses the outcome. They will be superseded on purpose, by themselves, on an annual cycle.

The factories attempt at enormous volume with no constraint whatsoever and no name to attach any of it to.

Nobody is holding both halves. That is not a market share problem or a taste problem or a story about a company having a bad few quarters. It is a structural condition, and it explains why the last decade of sneakers has felt like an archive being slowly reprinted while everyone waits for something.

The capability is sitting there completely unclaimed, available to anyone, at a cost that would have been unimaginable to the people who built the last canon.

What is missing is somebody willing to sign it.