Riot Games and 296,416 Boosted Accounts: Auditing the Trust Behind the Ranked Ladder

Riot Games and 296,416 Boosted Accounts: Auditing the Trust Behind the Ranked Ladder Last November, in a small PC bang on Gwangan Street in Busan, I sat beside a 19-year-old who had just finished watching the VALORANT Asia-Pacific regional final. He did not talk about the match. He told me the story that had haunted his whole season: more than 300 ranked games to reach Diamond, then a match against a player whose movement rhythm, crosshair placement and ability timing belonged to a completely different tier. He lost, dropped a division, and spent nearly a month climbing back. Three weeks later he read a notice from Riot Games: the account that had beaten him was on a list actioned for rank manipulation. He was not happy the way I expected. He asked me a question I carried with me for months: "So do my 300 games get counted again, hyung?" That question led me to the document Riot Games published about Anti-Boost, the automated system the publisher uses to detect and penalise boosting in VALORANT and League of Legends. Reading it closely, I realised the answer to the kid in Busan is far more complicated than a notification line. An Accumulated Total and Four Definitions of Violation Riot Games states that Anti-Boost has actioned 296,416 accounts engaged in rank manipulation across both titles. How you read that figure matters as much as the figure itself. 296,416 is a cumulative total published by Riot, not independently audited, not split by title, and not divided by region. It is a photograph, not a trend line. To claim "Riot is tightening its grip," you need at least two points in time to compare — and this document offers one. Four categories of behaviour sit in the crosshairs. Boosting: a high-skill player logs into someone else's account and plays ranked on their behalf, earning rank points for the owner. Buying, selling or transferring accounts. Intentional deranking — deliberately dropping your own rank, usually to enable boosting or to meet weaker opponents. And using a secondary account to assist climbing. The subtlest boundary is the one Riot draws explicitly: a self-created, self-operated alt account remains normal activity. Anti-Boost targets intent to manipulate rank, not the mere ownership of multiple accounts. This is a narrow, intent-based standard — a deliberate design choice that protects legitimate multi-account play while still chasing manipulation. From the perspective of someone who has spent many evenings in community chat rooms, I understand why this choice irritates both sides. Clean players want a bright line: own a smurf, get banned. Riot chose a blurrier line: ban only when intent to manipulate exists. A blurred line is always harder to enforce transparently than a bright one. To understand why this system matters, look at the economy running behind it. The boosting market is a black-market current flowing parallel to the official ladder. Buyers pay for rank; sellers trade time and skill for money. No contracts, no warranties, no dispute mechanism. Riot does not publish the size of this market, so every estimate of it is guesswork. The Four-Tier Penalty Ladder and the Logic Behind Each Tier Anti-Boost's penalty framework operates across four tiers, and each tier reveals a different logic. The first tier handles detected behaviour. Ranked points and rewards earned through cheating are cancelled. The account is returned to its pre-manipulation rank. A temporary suspension follows. What stands out is the verb Riot uses: "cancel" and "return." The system does not claim prevention from the start; it repairs the consequences after detection. This is a reactive-with-rollback model, and it implicitly concedes a lag between the manipulation and the remediation. The second tier is escalation. Repeat offences bring longer bans. The very existence of an escalation mechanism is a confession of sorts: Riot anticipates a non-trivial recidivism rate. If every ban deterred absolutely, the ladder of penalties would be unnecessary. The third tier covers the two most commercially driven behaviours: account buying and selling, and intentional deranking. Penalties can reach permanent bans. Placing these two at the heaviest level is an economic signal: the publisher reads them as organised, profit-motivated activity rather than a one-off prank by a solo player. It is also a direct strike at the supply side of the black market. When the expectation of detection rises, the expected cost rises for both buyer and seller, and service prices must adjust to cover the risk. The actual contraction of the market is unmeasured, so this remains inference rather than conclusion. The fourth tier is the most contentious: joint liability. A booster's main account may be actioned. Teammates who frequently queue with that person may also be actioned. This is a wide-net measure, and it opens an obvious risk zone. A pair of friends who play together every night, unaware that one of them is being paid to climb for someone else, could be swept into the same enforcement wave. The document states no tolerance threshold, no number of shared matches that counts as "frequent," and no appeals mechanism. I spent two weeks rereading the enforcement notices Riot published over the past three years, cross-referencing them against community forum responses. What I found: most controversy does not erupt at tier one or tier three, but at tier four. Players accept that cheaters get punished. What they do not accept is that innocent people get punished alongside them. The press room is never empty; sometimes it is simply full of things nobody can put into words. Pooled Data and the Price of a Pretty Statistic One technical detail in the document matters more to me than the 296,416 itself. Riot pools VALORANT and League of Legends into a single total. These two titles have very different boosting economies. VALORANT is a tactical shooter where individual skill shows clearly through reflexes and crosshair placement, so the value of a high-rank account is tightly bound to proven mechanical reputation. League of Legends is a MOBA where rank reflects both skill and time invested, and where seasonal rank inflation pressure behaves differently. Pooling the two titles yields a bigger, more impressive, more press-friendly total. It also erases every possibility of analysis. Looking at 296,416, a reader cannot tell which boosting market is swelling, which region is hot, or whether enforcement actually works in either title. Based on my experience following matches in both of these games, I see that boosting motivation in VALORANT tends to attach to competitive reputation — buyers want rank to prove they belong in community tournaments and amateur tryouts. In League of Legends, motivation tends to attach to end-of-season rewards and a sense of progression. Two different motivations need two different treatments. A pooled total helps no one do that. The Race Between Detection and Evasion Riot says it will keep expanding Anti-Boost and add match-level detection, based on behavioural signs rather than results alone. That is the right direction, and it is also a direction that exposes the limits of the present. Match-level detection means the system reads behavioural patterns: movement rhythm, decision-making under pressure, the correlation between demonstrated skill and account history. This approach is stronger than looking only at win rate, but it also depends on whether the detection model keeps pace with boosters' adaptation. The history of every anti-cheat system in esports is a race. The detection side improves its model; the evasion side changes its method. When boosting is squeezed on the direct channel, activity shifts to harder-to-detect channels: organised deranking rings, off-platform communication, tacit agreements inside same-hour ranked matches. Riot's document does not address this scenario. I do not treat that as a publisher failing. Quality control over a playground with tens of millions of accounts is a problem with no perfect solution. But readers need to understand that every claim about effectiveness is a provisional claim, and a cumulative total proves no trend at all. The Contrarian Angle: What a Clean Ladder Actually Nurtures Most commentary on Anti-Boost stops at the fairness question: are clean players protected? I want to push the story one step further, to a consequence rarely discussed. The ranked ladder is the raw material of the entire professional esports ecosystem. Academies, youth rosters and scouts all start here. A Challenger account in a region is the signal that prompts a team to send a tryout invitation. When the ladder is manipulated, that signal is corrupted, and the price is not paid by the rank buyer — it is paid by scouts and by genuine talent. When a new star blazes, an entire generation sees itself in that light. But the light only falls correctly if the ranking table beneath it is still trustworthy. There is another contrarian angle I consider necessary. Publishing enforcement totals is a communications act, and it works in two directions. It reassures clean players that the system is functioning. It also signals that the measurement of the system's success sits with the publisher itself. When the detecting body, the adjudicating body and the reporting body are one and the same, players have little choice but to believe. That is not necessarily bad. In traditional sport, federations operate similarly and publish their own figures. The difference is that football federations have investigative journalism, sports courts, and independent bodies cross-checking them. A game publisher has no one cross-checking. Miss one syllable and you remember it for life — and in a system with no second referee, a ban recorded under the wrong name can erase an entire season for one person. An Empty Stadium, and the Heartbeats Behind the Screen The stadium is empty, yet I hear the heartbeat of a whole community more clearly. On the ranked ladder there are no stands, no cameras, no applause. Only a silent queue and a scoreboard. Yet that is precisely where tens of millions of people place their sense of fairness into an algorithm. The 19-year-old in Busan asked me whether his 300 games get counted again. The honest answer, under the system Riot describes, is: rank points and rewards derived from cheating are cancelled, the account is reset to its original rank, and the effort of everyone who lost to a cheater during that window is not restored. The system repairs the scoreboard, not the experience. I told him that, and he nodded and went quiet. Then he said something I wrote in my notebook: "At least I know somebody is still holding that table." What I Think After reading the document in full and talking with players across both titles, I believe Anti-Boost's real value is not the 296,416. Its value lies in Riot daring to publicly define what counts as cheating, what counts as normal activity, and what penalty attaches to each behaviour. An imperfect system with clear definitions still beats a silent one. What is still missing is a mechanism for wrongly punished players to speak up, and concrete thresholds for broad-net clauses like joint liability. Both are technically small and enormously significant for trust. A playground does not lose its fairness because cheaters exist. It loses fairness when an innocent player cannot learn why they were punished, and cannot learn whom to ask. Anti-Boost can get your account permanently banned. It cannot make you believe in it. That belief comes only from watching, across many seasons, that clean players stay and cheaters leave — and that when the system is wrong, the system knows how to apologise. My position rests on the following data: Riot Games' official disclosure on the Anti-Boost system (the cumulative total of 296,416 accounts, four categories of violation, four penalty tiers, the joint-liability mechanism); my own direct observation over three years following the VALORANT and League of Legends ranked ladders in East Asia; and community feedback I collected from player forums. When the ladder is still held, an entire generation of young players still has somewhere to see themselves reflected. When it is not held, where do they look?