DESIGN INTELLIGENCEJune 27, 2026·Mary · DEPIX Design Intelligence

Nintendo redesigned the Joy-Con from scratch and kept the drift.

There is a particular kind of design failure that is not an accident. It is a choice, made with full knowledge, defended for years, and then quietly carried forward into the next product. The Joy-Con thumbstick is that failure, and in June 2026 a regulator finally put a number on it: €35 million, levied on Nintendo in France over the drifting sticks that have plagued the Switch since 2017.

The technical story is almost boringly simple, which is what makes it damning. The Joy-Con stick is a potentiometer — a tiny strip of resistive material with a wiper dragging across it to read position. Drag anything across anything long enough and it wears. As the strip degrades, the controller starts reading movement that the player's thumb never made. The character drifts. The menu scrolls on its own. The hardware, in effect, begins hallucinating input. This is not a manufacturing defect in the usual sense; it is the predictable end state of a contact-based sensor asked to survive thousands of hours of a child's thumb.

A non-contact alternative has existed the entire time. Hall-effect sticks read a magnet's position with no physical wiper, so there is nothing to wear out. Third-party makers have shipped them in replacement Joy-Cons and rival controllers for years, at consumer prices. The engineering was never the obstacle. The cost-per-unit was, and so was the institutional reluctance to concede that the original part was wrong.

What turns this from an old grievance into a genuinely interesting design case is the Switch 2. Nintendo had the rarest thing in hardware: a clean sheet. A new console, a new Joy-Con, designed — in the company's own words — "from the ground up." This was the moment to fix it. Instead, teardowns by iFixit in June 2025 found a stick that looks structurally like the old one, with no Hall-effect or magnetoresistive sensor inside. Nintendo's own product developer confirmed the new sticks are "not Hall-effect."

The reason is the most instructive part of the whole saga. The Switch 2's headline feature is its new magnetic Joy-Con attachment — the controllers snap to the console with magnets instead of the old rail. But Hall-effect sticks read magnetic fields. Park a strong attachment magnet next to a magnetic sensor and you get interference. So the very feature meant to define the new product foreclosed the obvious fix for its worst-known flaw. Nintendo chose the magnet over the cure.

This is the decision a design chief should sit with, because it is not stupidity — it is a constraint cascade nobody mapped early enough. A single hero feature (the satisfying magnetic click) silently vetoed a reliability fix the company had been sued over for half a decade. There were paths out. TMR sensors — tunnelling magnetoresistance — are far less susceptible to stray fields and were available. Magnetic shielding was an option. Relocating the attachment magnets was an option. Each of those is a concept-phase question: a trade-off you can see on a whiteboard for the price of asking it, and a recall-grade liability if you ask it only after tooling is locked.

That is the quiet thesis underneath every one of these post-mortems. The Joy-Con drift was not discovered in the field; it was decided at the bench, years before any consumer touched it, by a feature priority that was never stress-tested against the product's own history. By the time the magnetic mount and the wear-prone stick were both in the bill of materials, the failure was already shipped. The expensive part is not the better sensor. It is the meeting that should have weighed "magnet snap" against "five years of lawsuits and a €35 million fine" and never happened.

At DEPIX we frame this as the cost of decisions made before anyone can see them. The concept phase is where a hero feature and a legacy defect quietly collide, and where — if you can visualise and interrogate the trade-off early — you still have the cheap option of choosing differently. Nintendo will sell tens of millions of Switch 2 units regardless. But it walked up to a clean sheet carrying a known, named, litigated flaw, and set it back down unchanged. That is not bad luck. It is a design decision, and it was reviewable.