Riot Games and the 296,416-Account Invoice: The Economics of Trust in Esports

In March 2026, in a crisis meeting in Seoul, I proposed an experiment that the CFO called "a sociology master's student's joke." He was right about my job title, wrong about the value of the idea.

When COVID-19 closed every K League stadium, ticket and advertising revenue collapsed. The club I worked for at the time faced an estimated operating loss of 8.2 billion KRW in a single quarter. The proposed solutions all revolved around cuts: cutting salaries, cutting the academy, cutting communications. I proposed another direction — selling digital advertising space inside a virtual stadium built on an esports platform, inviting even opposing fans to bid. No one believed it. We raised 410 million KRW for one televised derby.

The lesson I learned from that was not about technology. It was about something accounting cannot measure: when sporting value is under threat, the end buyer is not paying to watch a match — they are paying to watch a fair match.

Four years later, analyzing Riot Games' Anti-Boost system for VALORANT and League of Legends, I see the same financial equation being handled at industrial scale. Three hundred thousand accounts. Two titles. One global ladder. And the question behind every sanction: what is Riot buying back, and at what price?

The world looks at the stars, I look at the value sheet.

Context — When the ladder becomes an asset

To understand why Riot Games spends engineering, personnel, and legal budget on an anti-boosting system, you need to understand the economic role of the ranked ladder within the entire esports business model.

In any competitive title with a ranking system, from VALORANT to League of Legends, the ladder is not just a feature. It is infrastructure. Three layers of value depend on it.

Layer one is recruitment. Every professional talent-discovery pipeline in esports — from Gen.G and T1 scouts to Fnatic — starts from solo-queue ranking data. Without a clean ladder, there is no way to distinguish a genuinely skilled player from someone who bought their position. Over years of observing academy tryouts in Seoul, I noticed an unwritten rule: every scouting profile begins with a rank number and ends with a verification video of real skill. If the rank number is under suspicion, the whole profile collapses.

Layer two is direct revenue. Rank is tied to seasonal rewards, badges, in-game resources, and in some cases tryout opportunities at professional esports organizations. The higher the rank, the greater the account value — and this is where the black market begins.

Layer three is media and sponsorship. Brands like Red Bull, Mastercard, or Samsung do not sign contracts with a title unless they believe the competitive environment they associate with is legitimate. If the public believes the ladder is full of bought accounts, brand image is poisoned.

In other words: the ladder is an asset with commercial value, and boosting is the theft of that asset.

How does boosting work? A high-skill player — sometimes a professional or semi-professional — logs into someone else's account and plays ranked matches on the account owner's behalf. Result: the account owner pays to be placed at a level they never achieved through their own skill. This transaction is part of a broader gray economy that includes account buying and selling, intentional deranking, and using smurf accounts to climb.

Against this backdrop, Riot published a striking figure: 296,416 accounts actioned for ranked manipulation across both VALORANT and League of Legends. The number pools both titles, pools all regions, and — this is the key point — provides no period-over-period comparison.

This is the point mainstream media typically misses. A cumulative total is not a trend. It tells you the scale of the problem, not the direction of the fight. To judge whether the system is working, you need to know what the previous period's number was, what the rate of increase or decrease looks like, and how it breaks down by title and region. Without that data, any conclusion about system effectiveness is speculation.

Numbers do not lie; only the reader misreads them.

Core — The architecture of the Anti-Boost system

Riot's Anti-Boost system operates across a four-tier architecture, each tier addressing a different category of risk.

Tier one: detection and rollback

When the system identifies an account engaging in manipulation, the response follows a specific sequence. Ranked points and rewards earned from matches flagged as fraudulent are cancelled. The account is returned to its pre-manipulation rank. The violator receives a temporary account suspension.

In essence, this is a post-detection correction mechanism, not pre-emptive prevention. There is a lag between the violation and the remedy — that lag is the real engineering cost Riot accepts in exchange for greater detection precision. In financial analysis, we call this "late-detection cost" — a form of hidden cost that is systematically under-valued when budgeting compliance systems.

Tier two: escalation by recidivism

Repeat-offending accounts face escalating penalties. A first-time violator may only be suspended for a few days; a second offense lengthens the suspension; and so on up the scale.

This technical detail matters more than it appears. The very existence of an escalation rule implies a non-trivial recidivism rate. If most violators offended once and stopped, there would be no economic reason to design a complex escalation system — its operating cost exceeds that of a simple punitive regime. Riot accepting that cost means they have data showing players re-offend at statistically meaningful rates.

Tier three: heavy sanctions for commercial violations

Two behaviors receive the highest sanctions: buying, selling, or transferring accounts, and intentional deranking. Both can lead to permanent account bans.

Why is Riot harsh on these two violations but more lenient on others? Because these are the two behaviors most directly tied to the financial incentives of the boosting economy. An account that is bought and sold is a commercial transaction — not an accidental choice. A player who intentionally deranks to serve another person's boost is running a shadow business, not simply playing the game.

A permanent ban is not a preventive measure — it is a price signal. It is designed to change the cost-benefit calculation of participants in the black market.

Tier four: joint liability

This is the most controversial tier in the system's design. When Riot actions a boosting account, the punishment does not stop at the manipulated account. The booster's main account is actioned in parallel. And — the most controversial point — teammates who "frequently play with" the booster may also be pulled into the enforcement scope.

The design principle behind this is economically sound: boosting is not a solitary activity. Boost accounts typically operate in groups, with at least one high-skill player pairing with the boost to guarantee a certain win rate. If only the manipulated account is actioned, the actual operator of the service continues with another account.

But this principle also creates the widest gray zone in the system. When does a player "frequently" pair with a booster enough to be considered complicit? Riot does not publish a specific threshold. Innocent players — a duo who happened to queue with a booster for a few matches without knowing — can be swept into the enforcement net with no published appeal mechanism.

Safe harbor: intentional carve-out

Counterbalancing the tiers above is a clear scope rule. Riot explicitly distinguishes between self-created, self-operated alt accounts — normal activity — and intentional ranked manipulation. Having an alt account is not a violation. Using an alt account to manipulate the ladder is.

This is an intent-based standard, not a bright-line rule. Intent standards are more flexible in theory but harder to enforce transparently in practice.

Expanding to match-level detection

Riot states it will expand the system by adding the ability to detect signs of boosting at match level — not just from individual account behavior, but from recognizable signal patterns within the match itself.

Technically, this is a shift from identity-based detection to behavioral-signature detection. Instead of only asking "is this account behaving abnormally?" the system will ask "does this match show signs of a different-skill player manipulating it?"

This is an inevitable direction, but also where false-positive risk rises sharply, because behavioral signatures are harder to verify than account identity. A player whose skill spikes suddenly might be training hard, buying an account, or having their account used by someone else. All three possibilities can produce the same behavioral signature in the data.

The economic scale behind the numbers

When evaluating this system, one reality must be faced squarely: the 296,416-account figure is self-reported by Riot, with no independent third-party audit. This is not an accusation — most game publishers self-report their enforcement data, and no industry standard requires external audit. But the number must be placed in its correct epistemological context: it is a claim by one party to the transaction, not neutral data.

Moreover, the figure pools VALORANT and League of Legends — two titles with very different economic structures and boosting markets. A tactical FPS like VALORANT has a different rank-climbing cycle from a MOBA like League of Legends. Buyers of boosting services in VALORANT have different motives, different geographic markets, and different price points from buyers in League of Legends. Pooling the two titles into one number discards the most valuable analytical detail.

Based on my experience watching ranked matches in both titles at Seoul PC cafes and via streaming platforms, I can confirm a structural difference: boosting in VALORANT tends to show up as quick rank spirals observable over a few weeks, while boosting in League of Legends tends to stretch across a season and connect to larger account movements.

Contrarian — When defensive self-protection creates new risks

Three structural problems with the Anti-Boost system that mainstream media does not look at.

Problem one: the detection-adaptation gap

Boosting is an economic activity with profit incentives. Any economic activity with profit incentives will adapt when detected. When Riot publishes its detection methods, boosters adjust behavior. When Riot increases inspection frequency, boosters move to harder-to-detect channels — off-platform communication, no direct queue pairing, operations dispersed across multiple intermediary accounts.

This arms race has no end. That is the nature of rule enforcement in a market where buyers are willing to pay.

Problem two: the joint-liability paradox

Joint liability — punishing teammates who frequently queue with a booster — solves a real problem but creates another. It can unintentionally punish innocent players. And because no specific threshold is published, players have no way of knowing when they are in the danger zone.

The deeper paradox: once players know that randomly queueing with a booster could get them punished, their optimal behavior is to avoid social interaction, not to play by the rules. This runs counter to the core goal of a competitive game — encouraging healthy interaction.

Problem three: the transparency paradox

Riot publishes the 296,416-account figure as a signal of enforcement effort. But that signal has two sides. On one side, it reassures compliant players that the system works. On the other, it reveals that more than a quarter-million accounts engaged in ranked manipulation — a figure that exposes the scale of the problem the system is fighting.

Publishing enforcement data is a brand action, not a preventive action. It reminds the market that Riot is at war, not that it has won.

Takeaway

If I had to decide the budget for a system like Anti-Boost, I would look at three questions the current report cannot answer.

First, what percentage of those 296,416 accounts are repeat offenders? If the recidivism rate is under 10%, the system is working well. If it is above 30%, the system is treating symptoms, not causes.

Second, what is the impact on the black account market? There is no data on boosting-service prices before and after enforcement waves, no data on account transaction volume. Without those numbers, one cannot say whether the system is an effective economic deterrent or merely a technical ritual.

Third, who audits the system? A publisher that detects, adjudicates, and reports results is a concentrated structure with no counterweight. This can be effective for a time, but it creates brand risk if a high-profile false-positive case occurs.

When data speaks, the whole world suddenly listens. But data only speaks when someone asks the right question. And the right question here is not "how many accounts has Riot actioned?" but "what is Riot buying back, at what price, and is there a cheaper way to achieve the same result?"

On the global ladder of the world's two leading titles, 296,416 accounts is an invoice. The remaining question is whether this invoice will keep growing, stabilize, or shrink — and in all three scenarios, who actually pays.