How Riot Handles Boosting: 296,416 Accounts, a Four-Tier Penalty Ladder and a Joint-Liability Gap

One night in March, I rewound the VOD of a Diamond-ranked match on the Southeast Asian server and paused at the eighteenth minute. On the opposing side sat an account that played exactly two champions, queued exactly one role, had won eighty-seven percent of its forty games, and had been created ten days earlier. That player turned before the opponent stepped out from behind the wall fourteen times in eighteen fights. Reflex does not explain that. Luck explains it even less. Someone else was holding the mouse.

I am not telling this story to accuse an anonymous name. I am telling it because of that moment when you understand the match was sold before it began, and because Riot Games is trying to name that moment with a system: Anti-Boost.

Anti-Boost is the automated enforcement system Riot operates across both of its titles, VALORANT and League of Legends. The name sounds like a technical feature. In substance it is closer to a miniature criminal code, written for a ranked community of tens of millions, and enforced by the same party that writes the law.

Four categories of behaviour fall inside scope. Boosting, where a highly skilled player logs into someone else's account and climbs for them. Buying, selling or transferring accounts. Intentional deranking, losing on purpose to drop tiers, usually to make boosting easier or to meet weaker opponents. And smurf-assisted climbing.

Those four categories do not sit side by side as separate boxes. They form a supply chain. Intentional deranking produces low-ranked, high-skill accounts ready for boosting. Account trading supplies clean identities for the trade. Smurf-assisted climbing is the add-on service for paying customers who still want to play themselves. Riot does not list them as an ecosystem, but the structure of the penalty ladder shows they understand that hitting one link shifts pressure onto the rest.

Riot built a four-tier penalty ladder. The lowest tier: ranked points and rewards earned through cheating are cancelled, the account is reset to the rank it held before manipulation, and a temporary suspension is applied. The next tier: repeat offences increase ban length. The third tier: account buying and selling, or intentional deranking, can lead to a permanent ban. The final tier: joint liability, where the booster's main account and teammates who frequently queue with them can also be actioned.

Alongside that ladder sits a fairly clearly drawn scope boundary. Riot states plainly that alt accounts created and operated by the player themselves are normal activity. Anti-Boost targets the intent to manipulate rank, not the existence of alt accounts. This is a narrow standard, and it is narrow on purpose.

Riot has also said the system will keep expanding, with a new direction being detection of boosting signals at match level rather than only at account level. The end goal, in Riot's framing, is a fairer ranked environment.

Riot reports 296,416 accounts showing rank manipulation across VALORANT and League of Legends combined, with no regional or per-title breakdown.

For anyone used to industry enforcement reports, 296,416 is not shocking in a negative direction. It is shocking in another: how large the grey market is. Three hundred thousand accounts is large enough to constitute a market of its own, with supply, demand, intermediaries and a price. And every market responds to penalties the same way: by repricing.

Anti-Boost is not a purge. It is a mechanism for repricing risk.

A system like this operates much like a market does. Boosting services exist because people pay to close the gap between the rank they want and the skill they have. To make that service more expensive, you do not need to destroy it. You only need to raise the probability of detection high enough that expected cost exceeds expected benefit. Every penalty tier added is Riot raising that price.

And the escalating ladder is the quietest confession in the whole system. If a single punishment were enough to end the behaviour, there would be no reason to build a second, third or fourth tier. Designing an escalation path in advance means the publisher has already assumed offenders will return. Recidivism is not a hypothetical risk in this design. It is an assumption encoded directly into the law.

The intent-based standard is the most elegant part, and the blurriest. Riot deliberately does not ban alt accounts. It bans the intent to use them for rank manipulation. That protects most ordinary players, who create a second account merely to practise new champions or play with friends at a lower rank. At the same time, it places the entire burden on a system that can only read behaviour through technical signals. Inferring intent from data is a problem no studio in the world has fully solved, and Riot is no exception.

The joint-liability clause is where to stop and look longest. Extending penalties to the booster's main account is logically sound: if a violator uses an alt to ply the trade, the alt is not the only thing that needs hitting. Extending it to teammates who frequently queue with them is a different kind of step altogether. A player can queue ten straight games with a booster and know nothing about that person's behaviour. They signed no contract, cannot see login history, and have no way to verify the real identity behind the screen. In this system, being unaware can be treated as complicity.

Then there is the data problem. Pooling VALORANT and League of Legends into one figure looks tidy, but the two titles have structurally different boosting economies. VALORANT is a tactical shooter, where individual skill gaps show up clearly in statistics and can be caught through repeated behavioural signatures. League of Legends is a multiplayer arena where a standout player can still be dragged down by teammates, meaning boosting signals are weaker and noisier. Pooling the two loses the ability to compare what actually needs comparing.

Global pooled reporting also hides an operational reality. Boosting demand is not evenly distributed. It concentrates in regions where rank carries high social value, where account markets are active, and where the share of young players with money but no time is large. A global figure does not tell an analyst where the pressure sits, and therefore does not help allocate enforcement resources more effectively.

The figure is also self-reported by Riot. No independent auditor verifies that 296,416 is accurate, or how consistently the definition of manipulative behaviour is applied. I raise this as an epistemic limit, not an accusation. When there is only one source, you are reading a party's statement, not a cross-check.

One technical detail gets skimmed over: the system is reactive with rollback. Ranked points are cancelled after the behaviour is detected, not before it happens. That means there is always a lag between the manipulation and the remediation. During that lag, honest players have lost matches, lost points, and nothing is returned to them. The system rolls back the manipulated account, not the people who ended up on the wrong side of the match.

At the top of this structure, Riot is simultaneously the detector, the prosecutor and the judge. No independent appeals mechanism is described. I raise this as an observation about the structure of authority, not a moral criticism. A system with only one referee will always have a side that cannot appeal.

The summer of 2026 taught me one thing: a meta exists only to be broken. That holds at the tactical layer, and it holds at the behavioural layer too. Boosters do not live in the meta of the rules. They live in the meta of the loophole, and every time Riot plugs one, they go looking for the next. The asymmetry between detection and evasion is close to an unbreakable law. Riot must improve its systems in ways it can explain to the community, meaning slowly, with testing and documentation. A booster only needs to change one small behavioural detail to muddy the signal, and owes no explanation to anyone. The attacking side always has the speed advantage in a race like this.

Over the past three seasons I have watched hundreds of ranked matches across many tiers looking for boosting signals, and what caught my attention was not the number of clear cases. It was the ambiguous ones. A player performing abnormally well for three games and then returning to normal. An account climbing so steadily it looks like a diligent grinder. Those cases cannot be concluded, and precisely because they cannot be concluded, they never appear in any enforcement report.

Riot's recognition of self-operated alt accounts as normal carries an assumption about ownership. The account belongs to whoever created it, and the right to use multiple accounts is a legitimate one. That assumption differs from how many other publishers view the issue, where alt accounts are suspect by default. The difference is not purely technical. It is a governance philosophy, and it is why Anti-Boost's standard is narrower on paper yet harder to enforce in practice.

Tighter enforcement is an inference, not data. Riot published a cumulative figure of 296,416 accounts but did not publish the prior period's figure for comparison. A total is not a trend. To claim controls are tightening, you need at least two time points side by side and a stable definition. Neither is present here. Riot's stated expectation that these measures will make the environment fairer is a forward-looking statement, and it is recorded as such: an expectation, not a measured outcome.

The next concern is the trust loop. Anti-Boost exists to protect honest players. Yet a system that can action the unaware, offers no independent appeals channel, and publishes no false-positive rate is asking those same honest players to trust it. That trust is not built by declarations. It is built by verifiable data and by publicly corrected mistakes.

Three things could change how this system is judged: a baseline figure for comparison, a statement on the false-positive rate, and an appeals channel with a defined processing window. All three are within Riot's reach. None appear in what has been disclosed. That gap is where future disputes will play out.

There is another direction few mention. A clean ladder is an input to the entire amateur talent-scouting system. Academies and teams still use high-rank solo play as a signal. If rank is inflated by money, that signal loses value. Anti-Boost was not designed for this purpose, but it incidentally protects a pipeline the esports world has never properly priced.

The topic returns in cycles. Each time Riot publishes enforcement data, the community flares up briefly and settles, because there is no specific scandal to sustain it. That is the signature of a medium-heat, medium-life subject. It will not disappear, but it will not escalate on its own.

From an industry perspective, a system like Anti-Boost is an investment in trust, and trust is the hardest asset to value on any esports platform. Players do not spend money to buy rank. They spend time to buy the feeling that rank reflects their ability. When that feeling disappears, the value of the entire ladder collapses with it, and so does the value of everything built on the ladder: scouting, amateur competition, and the community's seriousness.

A great rule-maker is not the one who draws the meta, but the one brave enough to erase it. For Riot, the issue is no longer whether it is tightening its grip. The issue is whether a system with no independent court, no baseline data and no public appeals mechanism can keep the trust of the very players it claims to protect. Fate never favours anyone. It only rewards those who know how to read the RNG, and in this game, the RNG is player trust.