296,416 Accounts, a Four-Tier Penalty Ladder and an Unaudited Grey Zone: How Riot Is Trying to Save the Ranked Ladder

296,416 accounts. Riot Games released that figure in a single line of a statement with no chart, no date range, no regional breakdown, and no split between VALORANT and League of Legends. To an analyst, that is an incomplete input. To someone who has sat in a club's budget meetings, it is a far more readable signal than a complete press release would be.

I read it first in Gangnam, in the closing stretch of the annual season — a stretch in which a single win carries the weight of three weeks of climbing, and in which the price of rank-boosting services peaks. The annual season teaches an analyst one thing: every esports crisis runs on a schedule. It does not erupt on opening day. It erupts in the final week, when seasonal rewards are already within sight.

Three weeks earlier, I was sitting in a PC bang outside Mapo district at one in the morning. The student beside me was not playing ranked. He was negotiating over KakaoTalk the rental price of an account for ten matches — just enough to push its name across the boundary between two tiers before the season closed. Old-fashioned cheating requires software, requires malware, requires an invincible player crushing opponents. None of that was present here. There was only a transaction between two people and a ranking system sold out from the inside.

The problem Riot is solving is therefore not a technical problem. It is an ownership problem.

What Is Being Sold, and Who Is Buying

To understand why a publisher had to build an entire enforcement system called Anti-Boost, you have to understand what the market behind it sells. What is sold is not an account. What is sold is rank — a social label convertible into money, into internship opportunities, into a position within amateur scouting communities, into personal credibility inside a culture that treats a climbed rank as proof of ability.

Boosting runs on a very clear commercial logic: the skilled sell their time, the wealthy buy rank. No one gets hurt. No one is robbed of property inside that match. But there is always a third party harmed without ever being consulted: the players in the same matchmaking queue. For every match with one player far above the correct skill bracket, the four others in that queue are playing a game whose predictability has been distorted.

Across Vietnam and Korea, I observe two distinctly different demand structures. In Vietnamese communities, demand leans toward seasonal rewards and eligibility for grassroots tournaments — cyclical value that spikes in the final week. In Korea, where rank is tightly bound to PC bang culture and amateur tryout circuits, demand is year-round and sold in longer-term packages. Same behaviour, two different demand curves. And both sit outside the reach of any club balance sheet.

The notable point: Riot does not sell climbing rights. Riot sells the authenticity of rank. When that authenticity depreciates, the first thing to lose value is not the player experience — it is their own talent-scouting pipeline.

Four Penalty Tiers: The Architecture of an Enforcement System

The penalty structure Riot published divides into four tiers, and the way it divides says more than the content of each tier.

Tier one: when rank manipulation is detected, rank points and rewards obtained through cheating are cancelled, the account is returned to its original rank, and a temporary suspension is applied. This is the correction tier — the goal is to restore the pre-violation state.

Tier two: repeat offences extend the suspension period progressively. This is the deterrence tier.

Tier three: buying, selling or transferring accounts, or intentional deranking, can lead to a permanent ban. This is the economic suppression tier.

Tier four: associated parties — the booster's main account and teammates who frequently queue with them — may also be actioned. This is the liability-expansion tier.

The first three are standard design in any anti-cheat system in the industry. The fourth is the one that made me stop.

The Safe Harbour and the Intent-Based Standard

Riot draws a clear boundary: alternate accounts created and operated by the player themselves are normal activity and are not actioned. Anti-Boost targets the intent to manipulate rank, not the existence of an alt account.

This is a narrow, deliberate standard. It protects a common and legitimate behaviour — players creating a second account to practise a new role, play with lower-skilled friends, or experiment without damaging their main rank. At the same time it creates a grey zone in enforcement.

An intent-based standard is far harder to apply transparently than a bright-line rule that simply bans or permits. When the verdict depends on inferring intent from behavioural signals, the consistency of the system becomes a variable rather than a constant. And in any system where the rule-maker is also the detector, the adjudicator and the publisher of results, players are left with a single shield: trust.

I have seen that trust break. In 2026, while working as a financial analysis assistant at a club, I watched an internal disciplinary decision rest on behavioural inference rather than direct evidence. Logically, the decision was correct. In public relations terms, it cost the club three months of explaining. A publisher with hundreds of millions of players does not have those three months for every case.

Joint Liability: The Largest Blind Spot

The clause covering teammates who frequently queue together is the point I assess as the highest governance risk in the entire policy.

The logic behind it is sound: without action against accompanying parties, a booster could neutralise penalties by moving to a new account and dragging accomplices along. But the measure also strikes a group that has violated nothing — players who happen to queue with an account being boosted, or duos who play together all season without knowing their partner is operating someone else's account.

What worries me is not the clause itself but the silence around it. No pairing threshold is published. No minimum number of matches to establish complicity. No independent appeals mechanism is described. A player swept into a case does not know what they need to prove, to whom, or within what timeframe.

When an enforcement system expands the scope of liability without publishing a threshold and without opening an appeals channel, the cost does not fall on the violator — it falls on the innocent.

In finance, we call this contagion risk. You do not need to default to be dragged down by a debt. You only need to stand close enough.

Detection Asymmetry and the Problem of a Self-Reported Figure

Riot states clearly that it will continue to expand Anti-Boost and is developing the ability to detect signs of boosting at match level. That admission is itself an important data point: it means current detection methods are not enough.

Based on my experience following ranked matches and player communities in both markets, I consider the asymmetry here structural. The detection system must operate across a vast and heterogeneous behavioural set. A booster needs only to change one small variable in their process to fall outside the detection pattern. The defending side carries a large fixed cost and must be right every time. The attacking side carries a marginal cost near zero and only needs to be right once.

And here is the problem with the 296,416 figure: it is a cumulative number published by Riot itself, not independently audited, with no prior-period data for comparison. A cumulative figure does not express a trend. It expresses scale over an undefined period.

The world looks at the star; I look at the valuation sheet. And on this valuation sheet, the prettiest number is the least informative one.

Pooling Two Titles, and the Real Price of the Grey Market

Reporting VALORANT and League of Legends together is a communications choice, not an analytical one. These two titles have fundamentally different ranked economies.

VALORANT is a tactical shooter. Rank here is tied to mechanical skill and reflexes, the climbing lifecycle is shorter, and boosting demand concentrates around short promotion milestones. League of Legends is a MOBA with a deeply layered rank system, where account value at high tiers is more stable and the account trading market is far more liquid. A high-tier account in the latter title carries greater asset value, and therefore a stronger incentive to be stolen or boosted.

Pooling the two into one figure flattens all of that difference. It is not technically wrong. It simply prevents an analyst from answering the most important question: where the grey money is flowing hardest.

On market effect, I assess Anti-Boost as operating like a price-raising mechanism. When detection risk rises, the expected cost for both buyer and seller rises with it. In a market with inelastic demand, that cost does not disappear — it is passed into price. Buyers pay more. And as price rises, a portion of marginal buyers leave the market, while sellers professionalise to offset the risk cost. The result is not a market that vanishes. It is a market that concentrates, becomes harder to detect, and carries a higher margin.

I found the diamond in the rough data right here: the harshest penalty is not aimed at the buyer. It is aimed at market operation — buying, selling and transferring accounts, and intentional deranking to optimise profit. Riot is striking the distribution layer, not the consumption layer.

Vietnam and Korea: Two Ways the State Intervenes

This is where I find the source coverage missing an important dimension.

In Korea, trading game accounts is not merely a terms-of-service violation. The Game Industry Promotion Act stipulates that buying, renting or brokering the trade of game accounts is prohibited by law, with sanctions of up to two years in prison or a substantial fine. That means in this market, Anti-Boost does not operate alone. It operates layered on top of a state sanction regime that already exists.

In Vietnam, the legal framework mainly governs the player-publisher relationship through terms of service, without a dedicated criminal sanction for account trading. That places the entire enforcement burden on the publisher and on intermediary platforms.

These two models produce two different behaviours. In Korea, account trading moves into closed channels, at high prices, accompanied by genuine legal risk. In Vietnam, trading happens relatively openly in community groups, at lower prices, with the primary risk being account lockout.

Numbers do not lie; only readers misread them. But a number produced under two different legal frameworks cannot be read with the same yardstick.

The Contrarian Angle

The popular interpretation of this statement is: Riot is tightening the screws. I do not think that reading is backed by the data.

A cumulative figure does not establish a trend. To claim enforcement intensity is rising requires at least two comparable reporting periods, with the same unit of measurement and the same time window. This statement provides one total. The rest is the author's inference, not the data's conclusion.

The second, more important contrarian angle: if Anti-Boost's real objective is protecting the authenticity of the ladder, then the measure of success is not the number of accounts actioned. The number of accounts actioned measures workload, not outcome. An effective enforcement system should drive violations down, meaning the published figure should shrink over time. A system that only publishes ever-larger numbers is proving detection capability, not prevention capability.

And the largest blind spot: this war cannot be won with penalties. It can only be won by making rank less convertible. When a tier no longer opens internship opportunities, no longer serves as an eligibility condition for grassroots tournaments, no longer stands as a social credential worth displaying, the boosting market shrinks on its own. No publisher wants to say that, because it runs counter to the very climbing incentive they built.

That is the paradox at the centre of every online ranking system: rank must be valuable enough for players to pursue, yet cheap enough that no one wants to buy it.

What to Watch Next

Three signals I will track in coming disclosures. One: whether Riot publishes period-based data rather than a cumulative figure, since only then can a trend be read. Two: whether a pairing threshold and an appeals channel are published for the joint-liability clause, because otherwise contagion risk remains a hidden cost borne by innocent players. Three: whether match-level detection is described concretely, since that is the only indicator that the system is moving from correction toward prevention.

When data speaks, the whole world suddenly listens. But a self-reported figure, with no prior period, no independent audit, and no split by title or region, is still speaking a language only the publisher fully understands. Players deserve to hear the same story in the language of verifiable numbers — otherwise what gets protected will not be the ranked ladder, but the reputation of the party that published it.