Riot Games and 296,416 Boosted Accounts: A Four-Tier Penalty Ladder Missing One Column

296,416 accounts. That is the figure Riot Games published when describing its Anti-Boost system — the number of accounts flagged for rank manipulation in VALORANT and League of Legends. Two titles pooled together. No regional split. No platform split. No baseline from a previous reporting period.

In the tracking sheet I have maintained across seven years of covering esports from South Korea, there is one column reserved for publishers' enforcement disclosures. That column now holds eleven rows. Not one of them carries a denominator. A total detached from a comparison base is a single data point, not a trend — and that is the anchor before anything else in this document.

Definitions and boundaries

Boosting is the act of a high-skill player logging into another person's account, playing ranked matches on that owner's behalf, and pushing the account toward a rank its owner could not reach alone. Anti-Boost is Riot's automated enforcement system, built to detect and penalise this behaviour along with related forms of rank manipulation.

The published violation taxonomy has four categories, and the boundaries between them matter more than the labels: boosting — playing ranked on someone else's account to earn points for them; buying, selling or transferring accounts; intentional deranking, meaning deliberately losing matches to lower one's own rank, usually to enable boosting or to find weaker opponents; and smurf-assisted climbing.

One point I marked in bold in my notes: Riot does not ban alt accounts; Riot bans the intent to manipulate rank. A self-created, self-operated alt account sits in the category of normal activity. This is a narrow, intent-based targeting standard rather than one based on the mere existence of an account — a deliberate design choice, and also the source of the transparency problem I will return to.

The four-tier penalty ladder

Reading the penalty structure, I found a clear escalation scale divided by how commercialised the behaviour is:

| Tier | Violation | Consequence | |---|---|---| | 1 | Detected manipulation | Cheat-derived ranked points and rewards cancelled; account returned to original rank; temporary suspension | | 2 | Repeat offence | Ban duration escalates with each offence | | 3 | Account buying/selling or intentional deranking | Possible permanent ban | | 4 | Associated parties | The booster's main account and frequently paired teammates may also be actioned |

Tier 4 is the part worth arguing about. Riot extends liability beyond the directly manipulated account: the booster's main account, and players who frequently queue with that person. The published text states no pairing threshold, no minimum match count for what counts as "frequent", and no appeal mechanism.

In the data I collect, this is the kind of rule whose false-positive probability is structurally non-zero. A legitimate duo, incidentally queuing alongside someone who happens to be boosting for a few matches, falls inside the clause's coverage. My sheet has no row recording such a case, because Riot does not publish one. The absence of data here does not equal the absence of the phenomenon — it is an observation gap, and I classify it exactly as that.

Reactive with rollback, not preventive

The most important technical detail in the entire description: Anti-Boost operates as a reactive system with rollback. Ranked points and rewards are cancelled only after the behaviour has been detected. That implies a lag between the moment manipulation occurs and the moment remediation happens. During that lag, match results have already shifted the ranks of other players, and the description names no mechanism to compensate that group.

Riot says the system will keep expanding, and mentions developing match-level detection of boosting "signs". That phrasing is itself an admission: current methods are not sufficient. The implication sits inside the publisher's own statement about needing further improvement; I do not need to infer it.

The asymmetry between detection and evasion

Every automated enforcement system in esports runs inside a race. Enforcement scales up; violators change methods. The enforcement-disclosure data I have logged over seven years shows a repeating pattern: after each major announcement, behaviour migrates to harder-to-detect channels — off-platform communication, organised deranking rings, intermediary accounts. This race has no finish line, only adjustment cycles.

Data never lies, but it keeps the questions nobody asked. The question here: if Riot is expanding enforcement while conceding that its methods need improvement, is the real detection rate rising or falling against total violation volume? Without a denominator, there is no answer. And I will not build a chart from a single point.

The contrarian read: "cracking down harder" is inference, not fact

Media coverage of this disclosure runs almost uniformly on one template: Riot is tightening the screws. I cannot find the data basis for that conclusion inside the disclosure itself.

296,416 is a cumulative total. To claim "tighter", you need at least two periods with two totals, or a ratio between actioned accounts and active accounts in the same window. The disclosure contains neither. It also does not separate VALORANT from League of Legends, despite the two titles having very different boosting economies — a tactical shooter and a MOBA do not carry the same rank-inflation pressure. Pooling both into one number discards precisely what the number should have delivered.

There is one more thing. The enforcement data is published by Riot itself, without independent audit. The figure may be correct; its verification status is unresolved. In my trade those are two different categories, and I do not merge them into one cell.

The question left unasked in the press room is the strongest signal I have ever recorded. Here, the unasked question is: where is the appeal mechanism for the people swept into tier 4? A penalty system built on behavioural signals, running automatically, without an independent adjudicating body, and carrying an expanded joint-liability clause, concentrates all governance authority in the publisher's hands. That is a deliberate design choice, and it should be read as one.

Signals for the next cycle

Three things I will watch across the next two disclosures. A second total, so my data column turns from eleven scattered points into a line. A clarification on the pairing threshold inside the joint-liability clause — the appearance of a threshold would tell me Riot has registered the false-positive risk. And the first public appeal case, if one comes, because that is the real-world test of an intent-based standard.

When the stands are empty, I hear the data's sigh more clearly. Ranked is infrastructure, not a stadium event, and infrastructure only gets attention when it breaks. Riot is spending money to keep it from breaking, and that deserves credit. But a four-tier penalty ladder with no appeals column, and a total with no denominator, are two gaps still open. They will stay open until someone is forced to publish the next row.