Riot Games Sanctioned 296,416 Boosting Accounts: A Ranked Ladder Is Trust, and Trust Always Has a Price

Late December in Seoul, the snow fell hard enough that I had to walk nearly two kilometres from Hongdae Station back to my small flat in Mapo. My phone buzzed inside my coat pocket. An assistant coach I know on an LCK team sent a single line: "Riot just published the number of accounts punished for boosting. Read it." I opened the article in the elevator and stood still on the third floor for a long time.

296,416 accounts. Riot Games calls that the output of its Anti-Boost system, spanning both VALORANT and League of Legends. I have sat in front of screens watching thousands of professional matches, written about plays in the fortieth minute, cried among Korean crowds at Kazan Arena. And yet an administrative notice from a publisher made me think harder than a grand final did. Because behind it sits the thing I still call the soul of the game, and this time that soul is being weighed on a ladder of ranked points.

Which layer Anti-Boost actually occupies

To read this properly you have to place the notice exactly where it belongs. Anti-Boost is Riot Games' automated enforcement system, built to detect and punish rank manipulation in ranked play. It is not a balance tool, not a patch, not a round of champion adjustments. It operates on an entirely different layer: account behaviour.

That distinction matters more than it looks. When Riot ships a patch, the community argues about win rates, about a champion's laning strength, about which team adapts faster. When Riot runs Anti-Boost, the question is no longer who is stronger inside the match. The question is whether the person behind the keyboard is actually the owner of the account.

Over years in this trade I have learned something about the ladder that few say out loud. A high rank is not only a psychological reward. It is currency. A Diamond or Challenger account can be sold, rented, or used as a springboard into a tryout at a professional team's academy. It is a ticket into community tournaments, into scouting sessions, into conversations outsiders never hear.

Based on my experience covering matches in both Korea and Vietnam, I can say the ranked ladder is the railway system of an entire esports scene. Professional teams run on it, rookies grow on it, audiences trust it. When the rails bend, the trains still run, but nobody is quite sure where they will stop.

The four behaviours Riot targets

Riot divides its enforcement targets into four clear categories.

Boosting is a high-skill player logging into someone else's account and playing ranked matches on their behalf, climbing on the owner's behalf without the owner playing a single game. It is the most common violation and the most commercial one, because money usually sits behind it.

The second category is buying, selling and transferring accounts. Trading accounts as merchandise is flatly prohibited. Riot treats an account as a competitive identity, not an item on an exchange.

The third is intentional deranking. A player deliberately loses matches to drag their rank below their real level. The motives vary: to be matched against weaker opponents, to produce cheap accounts for sale, or to act as a springboard for a boosting service. Riot treats this most severely, because it does not just ruin one person's experience, it ruins an entire sequence of matches whose victims never knew they were drafted in.

The fourth is using another person's account to manipulate rank generally, money or no money. A group of friends climbing on each other's behalf sits inside this zone.

The notable part is that Riot does not ban alt accounts. It bans the intent to manipulate rank. A player who creates their own alt and plays it with their own skill is not actioned. That is a narrow and very deliberate line. Riot chose to strike at intent rather than at the existence of alts.

The four-tier penalty ladder

When Anti-Boost detects a violation, it does not apply a single penalty but runs a four-tier ladder.

Tier one, for a first offence: ranked points and rewards derived from the cheating are cancelled, the account is returned to its pre-manipulation rank, and a temporary suspension follows. In other words, the system does not merely punish, it rolls back.

Tier two, for repeat offenders: ban duration escalates. Someone already actioned who violates again faces a longer lockout next time.

Tier three covers the two heaviest behaviours, account trading and intentional deranking: a permanent ban is possible. This is the line Riot uses to separate players who erred from people doing business on top of error.

Tier four is the part that made me read the page twice: joint liability. The booster's main account may be actioned. And teammates who frequently queue with that person may also be actioned.

Joint liability: the severity and its trap

The joint-liability design shows Riot understands how boosting services operate. A lone booster can be spotted and removed. A group of a booster plus regular teammates is harder to identify, because the group's win rate looks like an ordinary well-synced team. Riot chose to make the whole group carry the risk.

But this is also the largest blind spot in the entire system. An ordinary player queues with a friend, unaware the friend is boosting for another client. That player follows the rules, pays for the game, behaves properly. Then one day access vanishes. In the documents Riot published, no pairing threshold is stated, no appeal mechanism is described, and the word "frequently" is never defined.

I have seen something similar in another setting. Back when I organised amateur tournaments in Vietnam, we once disqualified a team because one member played on an account not registered under his name. The whole team cried. The organisers were correct. The whole team still lost. None of them had known in advance.

The pain of a correct penalty that lands on the wrong person, in my experience, lasts longer than the pain of an incorrect one.

Why you cannot say Riot is cracking down harder

This is where I want to stop the longest.

Reading the notice, many will conclude Riot is getting tougher by the day. The phrasing in the document itself points that way. But look at the data coldly and the story changes.

296,416 is a cumulative total spanning two games of different genres: a tactical shooter and a multiplayer online battle arena. Riot does not split the figure by region, does not split it by title, does not publish exact start and end dates, and above all gives no prior-period figure for comparison.

A single number does not create a trend. To say enforcement intensity is rising you need at least two data points. To say it is faster or slower you need the ratio of actioned accounts to active accounts in each period. Without those, the comparison is a feeling.

There is another small detail worth noting. Riot itself concedes that match-level detection is still maturing, resting on behavioural signals. That means current methods are not perfect. If current methods are not perfect, then inferring whether the past was policed more or less strictly is speculation.

Some teams lose by playing the meta correctly and win by daring to break it. Enforcement is the same: a system can be right in principle and still wrong in signal.

What is actually being protected

Set the penalties aside and the most interesting part of Anti-Boost lies elsewhere. The system protects the intake of a scouting pipeline.

Think about how professional teams find people. Very few are discovered through an open tryout. Most are found on the ladder, where an unfamiliar account appears at a high rank and starts beating familiar names. Assistant coaches, data analysts, agents, all look at the same board.

When that board is diluted, the signal loses value. A player can rise by paying someone to climb for them, then get a tryout and expose their real ceiling. The team loses a slot. A more deserving person loses a chance. Worse, the team loses faith in the scouting system and narrows toward internal referral networks, where personal relationships outweigh data.

For amateur teams that reach a final on a lucky draw and one explosive match, I stay sceptical. One flash of brilliance proves nothing about the development system behind it. But a clean ladder proves one thing: the best players are where people can find them.

The price of a clean ladder

Point rollback is a design choice with its own meaning. The system does not just block a violator now, it erases the past. In exchange, there is a lag between when the behaviour happened and when it was actioned. Inside that lag, matches were played, players lost, moments were stolen.

I once rewatched a ranked match in which one player's numbers were so far beyond reason it felt absurd. Only later did I understand that was the kind of signal detection systems feed on. The problem is that at the moment the match was played, nobody in it knew who they were playing with.

Riot controls both detection and adjudication. No independent appeals body is described in the documents. This is a governance model concentrated entirely in the publisher, consistent with how Riot runs its ecosystem end to end. The upside is speed and consistency of standard. The downside is that when the system is wrong, no voice outside the system is strong enough to fix it.

Enforcement data is self-reported by Riot and not independently audited. Self-reporting does not mean dishonesty. But it does mean every conclusion about effectiveness stands on one leg, waiting for a second source.

The counter-reading of what everyone is saying

There is one reading of this notice that I consider wrong and yet very common: treating it as proof that boosting is now more severe than before. The reasoning runs roughly that if Riot had to action so many, the problem must be bigger.

But the number of actioned accounts depends on two variables, not one. The first is the real rate of violation. The second is detection capability. If Riot improves detection and therefore catches more while the number of violators stays flat, the figure still rises. Reading violator count off catch count is a basic inferential error that an entire esports media industry commits every time new figures land.

There is a deeper counter-reading too: joint liability may be quietly turning honest players into unpaid monitors. Knowing that playing with someone who gets actioned could get you actioned changes behaviour. Players begin avoiding random relationships in queue. A ladder where people are afraid to play with strangers is a lonelier ladder, and that loneliness pushes players toward closed organised groups, exactly where boosting services run smoothest.

Ironically, a hard-line measure may be nourishing the very thing it wants to eliminate.

The kind of referee nobody sees

We are used to the idea that a patch is an invisible referee with the power to decide championships. A small change to minion damage can shift an entire season, turning a team whose style fits into champion and another into eternal runner-up. Adaptability gets mistaken for strength.

Anti-Boost is another kind of invisible referee, but on a deeper layer, where there is no stand and no caster. It does not change who is stronger inside a match. It changes who is allowed to appear in that match. Over time, the second kind of power is the one that shapes this sport.

Every match is a ballad; the storyteller's only job is to listen to its echo. But for the ballad to mean anything, the singer has to be a real person.

What I carry away from reading it

Esports taught me that emotions have a cooldown, but longing does not. I still remember the feeling of sitting in Seoul and realising for the first time that a ladder is not a list of numbers but a promise between millions of strangers.

Riot actioned 296,416 accounts. That number may signal a serious clean-up, or it may signal a detection system that has merely learned to see more clearly. Both can be true at once. Our job as writers is not to compress them into a single convenient story.

What I want to watch in the coming months is not the next figure. It is whether Riot publishes a pairing threshold sufficient to exclude wrongful penalties, and whether a player swept into the net of joint liability has any route to speak up. If there comes a time when innocent players get their accounts back along with an explanation, then we will have earned the right to say this system has grown up.

For now I keep one question for myself, every time I open the ladder: if nobody can check the thing we believe in, what exactly are we believing?