A Used CPU, A Fresh Record: Vanguard and the Governance Gap Riot Has Not Patched

On August 12, a VALORANT player in Germany turned on his computer and found that his account could not enter the game. He had not cheated. He had not installed any third-party software. He had simply installed into his machine a CPU he had bought secondhand on Kleinanzeigen — a Ryzen 7 5800X3D that had passed through someone else's hands. According to his own account, posted on Reddit and later relayed by Kotaku, Vanguard had previously blacklisted the chip's serial number. Its former owner — a man he had never met, whose name he did not know — was suspected of cheating. The punishment meant for that cheater fell instead on the buyer.

A week later, the story still has no ending. Riot Games has confirmed nothing. No statement, no official response, no line explaining whether the chip really was blacklisted, and if so, on what criteria. The player has been left to tell his story on Reddit, hoping someone at Riot will read it.

To stop there would be to miss the most important part. This story is not about one unlucky player. It is about a gap in how an enforcement system is designed, operated and — above all — explained. And in my line of work, where I read the match report before I read the news, that is the part worth discussing.

Vanguard Is Not a Patch. It Is an Enforcement System

To understand the story, one must separate two things that media routinely blur together. VALORANT has balance patches — things that change weapon damage, smoke vision, ability cooldowns. And VALORANT has Vanguard — anti-cheat software running at the kernel level, the deepest layer of the operating system, where it holds near-total access to the player's machine. The two are unrelated, and treating them as one is a category error.

Vanguard always runs. It boots with the computer, monitors processes, and when needed, records the hardware fingerprint of the machine — CPU serial number, motherboard code, network card MAC address. This mechanism is what the technical world calls HWID, short for Hardware ID.

The logic behind an HWID ban is simple and, to some degree, entirely reasonable. When an account is locked for cheating, the cheater only needs to create a new account to return immediately. Banning an account is like ejecting a person from a building through the front door while he still holds the back-door key. An HWID ban closes the back door too: it does not ban only the person, it bans the machine. To return, the cheater must replace hardware — a far more costly and troublesome barrier than registering a new email.

That is why Vanguard uses HWID. Not because Riot enjoys tracking your hardware, but because it is the only tool still effective against the flood of cheap throwaway accounts.

A Short History of a War Without an Ending

The war on cheating in online games did not begin with Vanguard. It began long before, and each generation of tools bred a new generation of countermeasures.

In the early phase, anti-cheat was simply file checking — scanning whether a player had installed suspicious software. But cheaters learned to disguise files, rename them, encrypt them. The next generation was application-level anti-cheat, monitoring game memory to detect interference. Cheaters learned to hide again. At some point, only one way remained to win the race: descend to a deeper layer, down to the kernel — where control is greatest and concealment hardest.

Vanguard is the product of that escalation. It is not an arbitrary product; it is the logical answer to a problem every competitive game publisher faces.

But every time the system descends one layer deeper, it trades privacy for power. And every time it trades privacy, a new question arises about the limits of that power. This is the price the industry rarely wants to name aloud: to fight cheating more effectively, players must hand publishers more control over their personal computers.

I have followed this trajectory for years, and what I have learned is this: whenever a system is granted more power to solve a problem, it simultaneously expands the surface on which it can create new ones. The HWID ban is a textbook case. It solves the throwaway-account problem, but opens the collateral-liability problem.

The Marking Mechanism: When a Machine's Identity Replaces a Person's

The problem lies here: a chip has no will. It does not cheat. It knows nothing of the rules. It is merely a mass-produced block of silicon bearing a unique serial number, and that serial number exists independently of whoever owns it.

An HWID ban operates on the assumption that machine and user are one. That assumption holds in most cases: whoever owns the machine uses the machine. But it collapses the moment hardware changes hands. The secondhand PC market exists precisely because hardware changes hands constantly — that is its nature. A CPU may pass through five owners in ten years. With each transfer, the identity of the machine and the identity of the user drift a little further apart.

According to the German player's account, not only the CPU was flagged. He said other components in the system were affected too. This detail matters, because it suggests Vanguard does not merely remember a single serial number, but builds a composite fingerprint from multiple components. If so, the scope of impact is far wider than a single chip. A composite fingerprint could render an entire machine invisible to VALORANT, regardless of who sits in front of it.

I have seen something similar in another field. In 2026, when stadiums stood empty because of the pandemic, I chaired the drafting of a 38-point checklist for football referees — from how to react to artificial crowd noise to stoppage-time management. It was applied to 23 friendly matches in the Marseille region, and disputes fell 18 percent. But the biggest lesson was not the number. It was this: every added criterion creates a new blind spot. The more variables you feed into an automated system, the more likely it is to mishandle a case its designers never imagined.

The HWID ban is the same. Each hardware component added to the fingerprint is a new variable, and each new variable is a chance for an innocent person to be caught in the net.

Collateral Liability: When Punishment Lands on the Wrong Person

This is the point I want to dwell on longest, because it touches the foundational principle of every system of rules: punishment must target the right party.

In football, when a player is sent off, no one punishes the club in the next match by banning the entire stadium. In tennis, when a player is banned for doping, no one confiscates the racket from whoever bought it at a secondhand shop. The principle distinguishing the offender from the instrument of offense is foundational to law, from ancient times to modern.

The HWID ban violates that principle systematically, if unintentionally. It does not distinguish between the person who cheated and the person who bought that person's machine. It flags an object, and anyone who touches that object bears the consequence.

In legal terms, this is the problem of collateral liability. A third party, uninvolved in the original misconduct, is made to bear the punishment. In this case, the third party is the buyer of the CPU.

Some will object: a buyer of used goods must accept risk. But which risk? A buyer of a used car accepts risks about the engine, the brakes, the chassis. He does not accept the risk that the car was once used in a robbery and will now be seized from his hands by police. The difference is this: technical risk is risk the buyer can assess, whereas risk tied to an object's history of use is risk he cannot see at all.

And here is the most uncomfortable part: the buyer has no way to check in advance.

The Lemons Market: When the Buyer Is Blindfolded

Economics has a concept called information asymmetry, and a classic example is the secondhand market. George Akerlof, the Nobel-winning economist, once wrote a famous paper on the used-car market, showing that when buyers cannot tell good goods from bad, the market gradually degenerates — good goods are pushed out, bad goods remain. He called it the lemons market, borrowing American slang for a defective car.

The secondhand PC market now faces a new and more serious form of information asymmetry. Previously, a buyer of a used CPU needed to worry only about performance, temperature, and whether the chip had been overclocked too hard. All of that was checkable — with benchmark software, with heat readings, with the overclocking history the chip retains.

But a chip flagged by Vanguard shows no external sign. It runs perfectly. It passes every performance test. It does not run hotter than normal. It simply cannot be used to play VALORANT — and possibly a few other titles using similar anti-cheat mechanisms.

The seller may know, or may not. If the seller is a banned cheater, he has a clear incentive to sell that chip and conceal the truth. If the seller is an ordinary user who does not play VALORANT, he may sell a flagged chip without ever knowing. In both cases, the buyer is the losing party.

Kleinanzeigen — Germany's largest classifieds marketplace, where the transaction is said to have occurred — offers no tool to check whether a component sits on Vanguard's blacklist. Riot provides no public lookup tool either. AMD, the chip's manufacturer, has no obligation and no mechanism to warn that its chips have been flagged. The entire risk sits with the buyer, who has no means whatsoever to protect himself.

This is a market failure, not an individual one. And market failures must be fixed with rules, not with advice to be more careful.

One Logic, Two Wrappers: HWID and VAR

I have spent much of my career discussing VAR — the referee-assistance technology in football. And I recognize that the HWID ban and VAR share the same structural problem, though they belong to two entirely different worlds.

VAR is not wrong. Technology is not wrong. A camera system with multiple angles, software that draws the offside line, a process for referees to review a situation — all are good tools designed to reduce error. The problem was never the tool. The problem is the operator, and the process deciding who may review, how many times, and who holds final judgment.

The HWID ban is the same. The mechanism is not technically wrong. It does what it was designed to do: identify and block a cheater's machine. But like VAR, it lacks an accountability process. There is no mechanism for an innocent person to say, I am not the cheater, I merely bought his machine. There is no court, no arbitration panel, no published appeal deadline.

In 2026, at 25, I publicly remarked on Russian television that IFAB's Referee Review Area process in the France–Australia match at the World Cup was not properly followed. I was criticized as rigid. But what I learned from that experience was not to stay silent — it was that a process which is not published cannot be checked, and a process that cannot be checked cannot be trusted.

VAR has the advantage that its decisions are recorded, replayed, dissected in public. The HWID ban is the opposite. The decision sits inside a closed system, with no record, no log. Those affected do not even know exactly why they were banned, until they infer it from having just replaced a component.

In football, when a penalty is denied, we can correct the error by admitting it, by reforming the process, by retraining referees. An individual error can be fixed. But a gap in the law — a situation the law does not cover, or covers ambiguously — cannot be fixed with an apology. It can only be fixed by writing an additional clause.

The HWID ban sits precisely in that gap.

Vanguard and the Privacy Worry

One cannot discuss Vanguard without acknowledging its history of privacy controversy. Since launch, Vanguard has raised concern for running at the kernel level, booting with the system and operating continuously even when the user is not playing.

These concerns are not paranoid. Kernel-level software holds deep access to a system, and any software with such power deserves scrutiny. The question is not whether Riot has bad intentions, but how that power is controlled.

What matters here is that the community's emotional foundation was laid long in advance. Those who oppose Vanguard on privacy grounds have existed for years. When a story like the German one appears, it does not fall into neutral space. It falls into a community already primed with suspicion, and therefore can travel farther and faster than its factual basis alone would justify.

This is a feature I have learned over years of observation: referee controversies never begin from zero. They always inherit a history of grievance. And those who write about them must distinguish between pre-existing grievance and the new event actually unfolding.

Why Ranked Players Matter More Than We Think

There is a temptation to dismiss this story as trivial, because its protagonist is not a professional. He is an anonymous ranked player. No contract, no team, no tournament. So why does it matter?

The answer lies here: ranked players are not the fringe of the esports ecosystem. They are its foundation. Every professional was once a ranked player. Every tournament exists on the basis of a community large and committed enough to nurture talent.

If an ordinary player cannot trust that he will not be punished for another's crime, then trust in the entire system is weakened. And trust eroded at the base will spread to the top, slowly but surely.

Every publisher understands this in the abstract. But the difficulty is turning abstract understanding into concrete process. And that is precisely where cases like the German one become a test.

Two Markets, Two Readings of the Same Event

My life is split in two halves: born in China, working in France. That split taught me something I carry into every article: the same event can be read two ways by two legal systems, and it is the comparison that reveals the curvature of each.

In France, and more broadly the European Union, consumers are protected by a relatively tight legal framework. A product sold with an undisclosed defect can create legal liability for the seller. The right to information, the right to complain, the right to compensation — all are codified. If a flagged chip was known to the seller but not disclosed, the buyer has, in principle, grounds to complain. But those grounds only matter if the buyer knows he was deceived, and to know that, he needs a transparent system — one Vanguard does not provide.

In China, where I grew up, gaming and consumer cultures have different features. The secondhand PC market there is tied tightly to component shops in electronics malls, where buyers often have a direct relationship with sellers, and personal reputation matters more than paper contracts. There, a flagged chip can be pushed out of the formal transaction chain and flow into gray channels, where the buyer is even less protected.

Both systems have blind spots. The European system is strong on rights but weak on enforcement when the harming party is a foreign corporation running a closed system. The Chinese system is strong on relationships but weak on institutional transparency. And in both, the individual player remains the weakest party.

The offside line was never straight; it is only that today I can see its curve. The German case is one of the rare occasions when that curve becomes clear enough to see.

What Riot Has Not Said

Here I must be clear about the limits of what we know.

The entire story rests on a single source: the player's account on Reddit, relayed by Kotaku. Riot has not confirmed it. PCWorld itself, in its report on the incident, described such cases as rare.

That means we stand before an event not fully verified. We do not know for certain whether the chip was really flagged. We do not know whether the former owner really cheated. We do not know whether Vanguard actually banned for a hardware reason, or whether there is another cause the player has not recognized.

I am someone who always places the record before emotion. And in this case, the record does not yet exist. But the absence of a record does not mean the event is not worth discussing. It only means every conclusion must be stated with an explicit level of certainty.

Based on what has been verified, a provisional conclusion: the HWID ban mechanism is real, Vanguard is a real tool, and the possibility that a buyer of used hardware is swept into another's ban is a logically real risk. What remains unverified is whether this specific case is exactly as recounted.

That distinction matters. It helps us avoid two traps: one, denying the story entirely for lack of evidence; two, inflating it into a system-wide crisis on the basis of a single case.

A Counterintuitive Angle: Perhaps Riot Is Not the One to Blame

This is where I want to turn in a direction few writing about this event wish to go.

The community's natural reaction is to blame Riot. Riot makes the rules, Riot runs the system, Riot refuses to explain. Emotionally, that is understandable. Analytically, it is not enough.

Let us separate two questions. First: should the HWID ban exist? Second: should the HWID ban be operated the way it currently is?

On the first, I believe the answer is yes. A competitive game cannot survive if cheating goes unpunished. And account bans alone have been rendered ineffective by cheap throwaway accounts. Any publisher facing the choice between using HWID bans and letting cheating run rampant would choose HWID bans. It is a reasonable, even necessary decision.

On the second, the answer is no. Not because the HWID ban is wrong in principle, but because it is operated without any mechanism to protect the innocent. A system right in its objective but wrong in its process can still cause harm, and that harm cannot be justified by the benefit of the objective.

In other words, the problem is not whether Riot has an HWID ban. The problem is that Riot has not built a process good enough that the HWID ban does not catch the wrong person.

One more point must be acknowledged, and I want to say it before being read as taking a side. Fan emotion is valid data. When a player spends money on a chip, installs it, then finds himself barred from playing for a stranger's crime, his anger is an entirely legitimate reaction. I do not stand above that emotion to judge it. I only want to say that emotion, if guided well, can become the motive to fix a gap — rather than just to vent anger at a name.

The danger of this story is not a single chip. It is this: if Riot stays silent, the gap will be filled with speculation, and trust in the system's fairness will erode bit by bit. Competing teams need assurance that sanctions are applied consistently. Ranked players need assurance they will not be punished for another's crime. Both needs are legitimate, and both are left open.

Three Proposed Clauses

I am not one to call for tearing down the system. Tearing down is easy; building is hard. For every gap I find, I force myself to write a concrete clause — one whose cost, risk and roadmap can be measured. Here are three for this case.

First, make the appeal process transparent. Riot should publish a formal appeal channel dedicated to hardware-related bans, with a committed response deadline. A buyer of a flagged device needs to know what he can do, within what time, and with what evidence. Currently, a player can only submit a general support ticket and hope.

Second, build a pre-purchase lookup mechanism. Riot could provide a public lookup tool allowing buyers to check whether a component serial number is on the ban list — similar to how carmakers offer vehicle-history checks. This tool need not reveal the former owner's identity; it need only answer yes or no. This is the most concrete step to break the information asymmetry.

Third, build a de-flagging mechanism for innocent buyers. If a buyer can prove he had no part in the original cheating — via a bill of sale, via transaction history — Riot needs a process to remove the flag from that component, or grant some form of exemption to the buyer's new account. This requires some operational effort, but it is far cheaper than the price of lost trust.

None of these three clauses requires Riot to abandon the HWID ban. All can be implemented without weakening anti-cheat capability. The only question is whether Riot wants to.

Conclusion

A match does not end with the whistle; it ends when people finish reading the record. The German story is still waiting for that record.

A secondhand CPU is becoming a test for a question far larger than itself: whether an enforcement system designed to protect a game's fairness can protect itself from becoming a source of unfairness.

I read the match record before I read the news, because the record does not lie. And in this case, the record has not yet been written. Until it is, every player considering a secondhand component — and every publisher considering how to run its anti-cheat system — stands before a gap left unfilled.

The question is no longer whether the HWID ban should exist. The question is whether it can exist without catching the wrong person. And the answer does not lie in Vanguard's source code. It lies in the decisions of those who write the rules — who need someone standing outside the line to check their signature.