Sweden, 13 January 2026. Armed men broke into the home of a League of Legends streamer who goes by RATIRL, tied him up, and spent hours hurting him. They wanted cryptocurrency. He told them he didn't have any.
They kept going anyway, well past the point where anything they could plausibly have extracted would have been worth the sentence they were earning. Eventually they left. What they took with them was his Pokémon card collection. Two suspects were later arrested.
That's the whole incident. Before turning it into an argument about plausible deniability, I want to leave it sitting there a second, because the argument is about a practice and he is a person it happened to.
Why plausible deniability didn't stop them
The advice he'd have been given, if he'd asked anyone in this space, is the advice everyone gets. Keep quiet. If someone asks, say you don't have any. That last part has a name in the literature and a nicer one in marketing, and the nicer one is what sells: plausible deniability.
Work out what a denial is worth when the people you're denying to know it's the standard move.
If "I don't have any" is what a holder says, and it's also what a non-holder says, then hearing it tells an attacker nothing at all. He hasn't learned you're clean. He's learned that you either have nothing or have been reading the same forum posts he has. A denial that everyone is coached to produce is not evidence. It's noise with a script.
So what does he do next? He's already committed the crime. Walking away means the break-in, the restraints and the assault bought him nothing. The only instrument he has left for resolving the question is the one he's already using, and the obvious move is to use more of it.
Plausible deniability didn't fail in the sense of being ignored. It failed in the sense that it was heard, and understood, and correctly discounted.
Three readings, and we don't get to pick
There is a live person at the center of this, and the thing everyone wants to know is the one thing nobody outside that room can establish.
Did he have bitcoin?
I don't know. Nothing published settles it, and a denial given under torture settles it least of all, because a denial is exactly what a holder produces too. So the honest move is to take the question three ways and notice something uncomfortable: the advice loses on all three.
If the denial was a lie
Suppose there was a stash.
That splits again — they got it, or they didn't — and the split turns out not to matter. Either way the secrecy failed at the one job it had, which was to stop him being picked in the first place. It didn't, and that's a Kerckhoffs failure of the most ordinary kind: a defense that works only while the attacker doesn't know the game, up against an attacker who knows the game.
Either way, too, the same thing happened to the encounter itself. A robbery ends when the attackers have what they came for. This one had nothing to end it, because the thing they wanted was the thing they had no way to check, so it ran for hours instead.
Now ask what he does on Tuesday.
There's no second mechanism to fall back on, because what he was sold was the mechanism. His options are to keep saying he has nothing, or to say he was lying, which invites exactly the people who already found him once.
So he doubles down. And look at what he's betting against now. Not the ordinary background risk he thought he was managing, but the attention his own case attracted. He's a widely reported example of someone criminals decided was worth hours of their evening. The only move the strategy leaves him is to keep insisting, while the attack itself did the advertising.
There's no exit from the strategy that isn't the strategy, and the failure made the next attempt likelier rather than less.
If the denial was true
Suppose he had nothing, which is what the registry's own entry records.
Then he was tortured for hours over an asset he did not own, because a large number of people who do own it have been trained to sound exactly like a man who doesn't. The coaching is what erased the difference. Somebody who never bought the advice, never joined the community, never held a coin, paid the bill for a norm he had no part in creating.
That's not a side effect of obscurity. It's the cost of it, landing where costs usually land, on whoever is standing nearby.
If it was somewhere in between
Suppose there was something, but less than they'd decided there was. Small enough that the attack was never going to be worth what it cost them, and certainly not worth what it cost him.
Then both failures run at once. Other people's obscurity aimed the attack at him, and his own didn't stop it once it arrived.
That all three fit is the verdict
The reflex, reading those, is to want to know which one is true so the argument can be settled properly.
It doesn't need settling. Look at what the three readings have in common: in every one, plausible deniability produced a worse outcome than it promises, and in none of them is there a version where the denial did its job. The argument doesn't rest on picking a branch. It survives all of them, which is a stronger position than any single branch would give it.
And across a population, this stops being a fork at all. There are enough holders and enough lookalikes now that every branch is happening to somebody this year. Somewhere a holder is doubling down on a defense that already failed. Somewhere a non-holder is being taken apart for resembling one. The fork is only a fork for a single case.
The sign, not the size
One thing this case fixes that a hundred arguments couldn't.
Most of the time when a security measure doesn't work, it doesn't work in the boring direction: you spent money, you got nothing, you're where you started. Obscurity is not in that category. It took a robbery, which is a thing that ends when the attacker has what he came for, and turned it into an interrogation, which is a thing that ends when he believes you.
Belief takes as long as it takes. That's the mechanism, and it runs on the victim's body.
So the number attached to this isn't zero. It's negative. A defense that makes the encounter longer and more violent than no defense at all is worse than useless, and "worse than useless" is a measurable claim, not a rhetorical one.
The part I'd rather not write
I argue for self-custody. I've written a protocol for it, and the papers under this post exist because I think holding your own keys is worth defending. So the following is expensive for me to say and I'd be a fraud if I ducked it.
If the only available response to a defense failing is more of that defense, it was never a mechanism. It was a hope with a technical vocabulary attached.
Look at the first reading again. A man is tortured for hours, the secrecy that was meant to prevent it didn't, and the sole move left is to be more secret and wager against a repeat. No review, no adjustment, no second layer. That isn't a security architecture. It's a person alone with a habit, and the habit is the entire plan.
Self-custody as currently practiced, for most people, is that. A device, a seed phrase somewhere, and an unexamined assumption that nobody will ever ask you in person. It works right up until somebody does, and it has no answer for the morning after.
The fix is not more discretion, and it is not better plausible deniability. It's a design where the outcome doesn't depend on what the man in your living room believes about you, because you can't control that and he knows you're trying. What that looks like is the other post in this pair, and the paper it points at.
He got his Pokémon cards taken. That's the line everyone repeats, and it's funny for about a second, and then you notice it's the measure of how little was there to find.
The case, the three readings and the externality argument are in The Denial Spiral, open access, no signup. Its companion, The Deadly Race, covers what happens when a seizure leaves the attacker a path that isn't finished: you become a competing racer, and a racer within reach is priced.
Incident details from the registry entry in Jameson Lopp's public registry, a press-sourced sample. Nothing here goes beyond what was reported, and I have no information about what he did or didn't hold.
If this was worth your time. Send it to someone whose whole plan is not being asked. A ⭐ on Great Wall, the research or these posts costs nothing and makes the work easier to find. And if you want to fund it, support takes ⚡ Lightning and on-chain, no signup, no tiers, nothing expected back.