Inside Riot Games' Anti-Boost System: 296,416 Accounts and the Joint-Liability Problem

THE STADIUM WITH NO CROWD On the night of August 12, 2026, I reopened a VALORANT ranked match at Immortal 2 that I had captured on recording software — an old habit from my days as a commentator for an esports outlet in Busan. No crowd, no commentators, no applause. Just ten anonymous accounts and ten brains behind the screens. But after three years of rewatching this kind of footage, I learned one thing: in ranked ladder, the distance between a genuinely strong player and one who was carried upward sometimes isn't measured in K/D, but in heartbeat. The ranked ladder has no stadium. It only has rules, an owner, and an underground economy running in parallel. Then suddenly, it added a number that made me sit up straight: 296,416 accounts. Riot Games claims it has handled that many accounts for rank manipulation across both VALORANT and League of Legends. That figure is large enough to force a question: what is Riot's Anti-Boost system actually doing, how far does it go, and is the ranked ladder that millions trust every day really as clean as the publisher claims? BOOSTING: WHEN RANK BECOMES CURRENCY For outsiders to understand correctly, a few lines of definition are needed. Boosting — which Vietnamese players still call "cày thuê" — is the act of a high-skill player logging into someone else's account to play ranked matches on their behalf, helping the account owner climb ranks without playing themselves. Sounds simple, but the ecosystem surrounding it is anything but. In titles like VALORANT or League of Legends, rank is not just a decorative badge next to a player's name. It is social currency. It is a priority ticket to amateur tournaments, where teams select members based on rank. It is a resume for professional scouts hunting for young talent. And above all, it is something a gray market is willing to pay for — from a few hundred thousand to several million Vietnamese dong for a single climb from Gold to Diamond, depending on the title, region, and difficulty of the target tier. That is why Anti-Boost exists. To be clear from the start: this system is not a champion-balance patch, it is unrelated to any VCT or Worlds tournament, and it has nothing to do with professional teams. It operates at the account and behavioral layer, not the in-game power-adjustment layer. This is a governance war, not a tactical one. And it is a war Riot is escalating. Per the publisher's official communications, Anti-Boost will continue to be pushed further in the coming period, with one notable new direction: detecting boosting signals at the match level, not just at the account level. The implications are significant. If successful, Riot will no longer have to wait for an account to climb too high before spotting anomalies — it can identify cheating signals while a match is still underway. I once called a legend by the wrong name — and since then, I listen to the ball more than to the title. That principle applies verbatim to esports: don't listen to the label "Riot is tightening the screws," listen to the number and the mechanism behind it. FOUR PENALTY TIERS AND ONE DELICATE BOUNDARY Riot's four penalty tiers form a clear escalating structure. For me, that structure is the most analysis-worthy part, not the 296,416 figure. Tier one covers detected rank manipulation: ranked points and rewards gained from cheating are cancelled, the account is returned to its original rank before interference, plus a temporary suspension. This is a state-restoration penalty more than a purely punitive one — the goal is to erase cheating traces from the ladder. Tier two covers repeat offenses: ban duration escalates with each violation. This is the key point. If a system needs an escalation rule, it implicitly admits that recidivism is high enough to make that rule necessary. In other words, Riot is quietly revealing that banning one account is not enough to stop people from returning to boosting. They will open new accounts, or move to harder-to-detect channels. Tier three covers the most commercially obvious violations: account buying/selling or transfer, and intentional deranking. The maximum penalty is a permanent ban. The fact that Riot reserves a life sentence for exactly these two behaviors shows it reads them as economic acts before rule-breaking. Account buying/selling is a commercial transaction. Intentional deranking, in many cases, is preparation for a boosting job or to enable low-tier amateur play. Both are tied to money flows. Tier four — the most controversial part, and the one I think readers should watch closely in coming months: joint liability. Per Riot's description, the booster's main account and teammates who frequently queue with them may also be actioned. The publisher extends the penalty scope beyond the directly manipulated account. Parallel to these four tiers is the targeting principle I consider the most sophisticated in the entire system, and the one many commentaries I read overlooked: Riot does not ban alt accounts. It bans the intent to manipulate rank. This is a boundary to understand correctly. A self-created, self-operated alt account is considered normal activity. Anti-Boost targets manipulation intent, not the existence of alt accounts. If you play a lower-tier alt to practice with friends, Riot won't touch you. But if that alt serves a boosting job, or is used to derank teammates in a ranked match, the story flips instantly. A star doesn't shine on its own — is there a hand fanning the flames? That question, which I use for football players, turns out to apply to ranked accounts with abnormally spiking metrics after just a few days. WHAT THE 296,416 FIGURE SAYS AND DOESN'T SAY The 296,416 handled accounts is a figure Riot self-published, pooling both VALORANT and League of Legends, with no regional breakdown and no time-comparison benchmark. Let me be direct: this is self-reported data, not independently audited. No third party verified it. That doesn't mean it's wrong — but it does mean we are reading data supplied by the enforcing party itself, not neutral data. Analytically, pooling a tactical shooter like VALORANT with a MOBA like League of Legends creates a problem. The boosting economy's dynamics differ sharply between the two titles. Rank-inflation pressure, regional boosting demand, service pricing, how tiers are perceived by the community — all differ. Merging them into a single number benefits PR but is useless for analysis. If the number doesn't tell the whole story, what does? I argue the penalty structure and how Riot defines violations are the most valuable information. The way it defines violations reveals its governance philosophy. The violation taxonomy has four main behavior groups: boosting (using another's account to climb), account buying/selling and transfer, intentional deranking, and using alt accounts to assist climbing. The last group is notable because it sits on the thinnest line between valid behavior and cheating. Using an alt to play with friends is normal. Using it to carry a friend above their true skill is manipulation. Riot must draw the line between the two based on behavioral data, not on player claims. This is where the system becomes complex in governance ethics. When you ban based on intent, you must prove intent. Proving intent with an algorithm is a problem with no perfect solution. HOW THE BOOSTING ECONOMY ACTUALLY OPERATES To understand why Anti-Boost must exist, you must understand the economy behind it. The boosting market operates as a multi-tier intermediary model, and each tier has its own incentives. At the top is the player with demand. Motivations vary: some want a high rank badge to show off to friends, some need a rank threshold to join amateur tournaments, some are simply tired of climbing themselves. In the middle are brokers, who don't play but coordinate orders. At the bottom are high-skill players taking jobs — sometimes amateurs, sometimes students, sometimes even semi-pro players seeking income outside practice. This layering creates difficulty for enforcement. A boosting transaction doesn't happen on Riot's platform. It happens via private messages, forums, payment channels outside the system. What Riot sees isn't the transaction, but behavioral traces: an account suddenly changing playstyle, an account with an abnormal win streak in fixed time windows, an account queueing repeatedly with the same group at a lower tier. That is why the shift to match-level detection matters so much. Account-level detection catches consequences. Match-level detection catches process. And catching process is always faster, more accurate, and less miss-prone. At the gray-market layer, penalties for account buyers and sellers — permanent bans — strike directly at the supply chain. If sellers are banned and lose accounts, their opportunity cost spikes. If buyers are banned too, demand falls. This is a strike at both ends of the market. WHAT IS ACTUALLY BEING PROTECTED Viewed through the lens of industry economics, Riot's true motives become clearer. The ladder is esports' largest pipeline of amateur talent. Pro teams, academies, and scouts all rely on high rank to screen candidates. If the ladder is inflated by boosting, its signal value collapses. A Radiant player could be a real talent, or a carried account. If you can't tell them apart, you must pay extra to evaluate candidates through other means — trials, scrims, direct observation — and that costs far more. This is why keeping the ladder clean isn't just a casual player's problem. It is a problem for the entire talent supply chain. At the mass-media layer, Riot publishing enforcement figures functions as a reputational signal. It tells investors their asset is being managed. It tells players the ladder has value. And it tells competing titles that Riot values ranked-system integrity in a way not every publisher does. WHERE THE SYSTEM MIGHT BE WRONG I write to argue, but I read to understand — if you only want to hear what you like, this article isn't for you. And what I want to say here is: Anti-Boost, despite smart design, still has at least three holes I haven't seen Riot explain adequately. First, and most serious: the joint-liability clause for teammates who frequently queue with a booster. This is a powerful tool, but also a PR time bomb. Imagine two friends ranking together every night. One, unaware, in a match happens to be queued with a booster using someone else's account. The match runs normally. Weeks later, that friend gets a notice their account was actioned for "frequently queueing with a booster." No tolerance threshold was published. No clear appeal mechanism described. Just a penalty. This is a scenario I cannot stop thinking about, because any system based on repeated behavior can catch innocent players. Second, an intent-based standard is harder to enforce transparently than a bright-line rule. With a rule like "you may not use another's account," determining a violation is simple: are you the account owner, are you the person who logged in. But with an "intent to manipulate rank" standard, you must infer from behavioral data, and inference always carries an error rate. Riot controls both detection and adjudication, with no independent appeals body described. Governance power concentrates entirely in the publisher. That's efficient, but it also means when Riot is wrong, no one forces it to answer to players. Third, the asymmetry between detection and evasion. Riot scales enforcement, but boosters also learn to adapt. Riot itself admits it needs to keep improving detection at the match level. That means current methods aren't perfect. In such an arms race, the winner is usually not the side with the best tools, but the faster adapter. And the history of ranked titles shows cheaters often adapt a little faster. There is also a structural issue: penalties are reactive after the fact. Points and rewards are only cancelled after detection. That creates a lag between manipulation and remediation. During that lag, the ladder is already contaminated, and honest players have already borne losses — such as losing a promotion match to a booster they thought was a normal player. Don't misunderstand me. I'm not saying Anti-Boost is useless. I'm saying a system that is praised is a system rarely questioned, and a system rarely questioned is a system that can drift in silence. THE ARMS RACE: LESSONS FROM OTHER TITLES The game industry has spent over two decades fighting cheating, and every major publisher has its own lessons. Valve faced a similar problem in Dota 2 and Counter-Strike, and its approach to boosting differs from Riot at a core point: it relies more on identity verification tied to phone numbers and credit cards, rather than in-game behavioral analysis. Blizzard, with Overwatch and its competitive titles, tried both directions: identity verification and behavioral analysis. Neither produced fully convincing results. Boosting persisted, simply shifting to harder-to-detect forms. The lesson from all these efforts is: no single measure is enough. Identity verification can be bypassed with a relative's account. Behavioral analysis can be fooled by slowly changing play patterns. Account bans only create incentives to buy new accounts, and that feeds the very market you're trying to kill. This is a structural paradox: the harder you squeeze at the account layer, the more you enable the account-trading market, and the more you devalue the very act of banning. Riot seems to recognize this paradox, which is why it's pivoting to match-level detection. But even that approach has limits. Detecting boosting signals at the match level demands an enormous volume of data and a machine-learning model sophisticated enough to distinguish between a normally good player and a player playing for someone else. The difference between those two cases can be tiny, and errors can genuinely harm honest players. WHAT I WILL BE TRACKING I don't have enough data to say whether Anti-Boost succeeds or fails. But I have enough to make three testable predictions. One, within the next 12 months, Riot will publish a new enforcement figure. If it is significantly lower than 296,416, or if they publish nothing new, that signals internal doubt about system effectiveness. If it rises, that signals enforcement expansion, not necessarily success. Two, the first public complaint of wrongful punishment for "frequently queueing with a booster" will be the real test of the intent-based standard. If it appears within a year, Riot will have to publish a tolerance threshold or an appeal mechanism. If it doesn't, it may be because the system is more accurate than I think, or because players have grown too accustomed to not trusting appeals. Three, I want to see whether a rival title publishes comparable enforcement data. If so, we have a comparison point. If not, Riot's 296,416 just floats in a vacuum — impressive, but with nothing to measure against. From keyboard to pitch, the shortest distance is a single mispronounced name — and the longest is never daring to correct it. Here, the shortest distance is a locked account, and the longest is players' trust eroding unnoticed. The ladder is a stadium with no crowd, and the referees there are algorithms. I don't know if the referee is calling it right. But I know a match with no crowd does not mean no one is watching. If you're an honest player, someone who spent hundreds of hours just to climb one tier, the question worth asking isn't whether Riot is tightening the screws, but how tight is tight enough — and who is watching the watchers?