Anti-Boost and Riot's Invisible Hand: What 296,416 Accounts Do Not Tell You

A November night in Seoul, I replayed an old VOD of a Korean streamer. He had just lost a VALORANT ranked match and was digging through his opponent's match history. In the previous game, the supposed "high-skill" player had landed exactly one hit on an enemy across twelve rounds. But in the match against him, that player racked up twenty-three kills. The streamer paused for a few seconds, then said: "This is a boosted account." I remember the chill that ran down my spine, because I had seen a version of this truth elsewhere — on the ranked ladders I had been tracking for seven years as an esports writer. Galio once cried on an OGN night, and today I understand why games have souls. When an account is driven by someone else, what gets stolen isn't the points. What gets stolen is the belief that every rank on the ladder corresponds to a real person.

Riot Games has just released a figure that made the whole industry stop. As of late last year, their Anti-Boost system had processed 296,416 accounts exhibiting rank manipulation behavior across both VALORANT and League of Legends. The number appeared in an official publisher announcement, with no regional breakdown and no split by title. It is a single block of data, pooling a tactical shooter and a MOBA into one concept called "rank manipulation."

I understand why they did it. From the governance layer, pooling makes the message stronger, tighter, easier to nod along to. But on closer inspection, that pooling obscures two entirely different economies. Rank-inflation pressure in VALORANT stems from a strict individual ranking mechanism where every promotion hurts. In League of Legends, boosting incentives are tied to the season cycle, to Diamond–Platinum tiers where players are willing to pay not to fall. Two illnesses, one prescription. That is the first blind spot in Riot's picture.

But wait — let's talk about what's right first, because Anti-Boost really is an impressive machine.

Riot doesn't ban alt accounts. Riot bans the intent to manipulate rank. This is the subtlest distinction in the whole system, and what sets it apart from how many other publishers operate. A player creating an alt account and climbing with their own hands — that is normal activity. Anti-Boost targets only behaviors that leave traces of intent: a stronger player sitting on someone else's account to farm points, a weaker player intentionally deranking to sell the account, or account trades on the black market.

The penalty ladder has four tiers. Tier one: upon detection, ranked points and rewards earned by cheating are cancelled, the account is reset to its original rank, with a temporary suspension. Tier two: repeat offenses draw longer bans under an escalating mechanism. Tier three: account buying/selling or intentional deranking can result in a permanent ban. And tier four — the most contentious — associated parties may also be actioned, including the booster's main account and teammates who frequently queue with them.

I spent many nights rereading disciplinary notices from the era when I sat in the OGN newsroom in Seoul. What caught my eye wasn't the number, but the philosophy behind it. Riot operates Anti-Boost as an automatic judge: detecting, adjudicating, and enforcing all at once. There is no independent appeals body behind it. No third tribunal for the convicted to plead to. Governance authority rests entirely with the publisher, from defining the violation to pronouncing the sentence.

This is effective. But effective does not mean fair.

Son's backdoor didn't change history; it showed that history is written by audacity. And in the Anti-Boost story, the audacity lies in Riot's willingness to bet on an intent-based standard. The problem with an intent-based standard is that it is hard to apply consistently. A bright line — say, an absolute ban on all alt accounts — would be transparent, verifiable, defensible. But a standard that requires reading a player's intent through behavioral signals and telemetry always leaves a gray zone. That gray zone is where innocent people can get swept up.

The joint-liability clause is the sharpest thorn. When Riot says that teammates who frequently queue with a booster may also be actioned, they are extending the net beyond the directly manipulated account. In theory, this hits organized boosting rings where multiple accounts coordinate to farm points for each other. In practice, it can also catch friends who happen to queue with a booster without knowing it. A pair of close friends climbing together all season — if one secretly takes payment to boost, does the other get penalized? Riot offers no tolerance threshold. Riot also publishes no appeals mechanism. That gap is not a small detail. It is the entire question of the system's legitimacy.

There is another angle few notice. When Riot publishes the 296,416 figure, they are sending a double message. To players, it is reassurance: we are working. To investors and partners, it is a competitive signal: we govern our ecosystem more seriously than other titles do. But both messages rest on self-reported data, unaudited by any independent body. The number is a claim, not third-party-verified evidence.

Worse, that number has no baseline. No prior-period data. No quarter-by-quarter comparison. When an article writes that "Riot is tightening the screws," that is the author's inference, not a conclusion from data. Data gives us a total, not a trend. To know the trend, we must wait for the next disclosure, set it beside this one, and only then dare to speak. The difference between "total" and "trend" is not academic. It is a matter of trust.

Esports taught me that emotions have cooldowns too, but longing does not. And what I remember most from my years tracking ladders is not the beautiful wins, but the silent losses. Real players, day after day, climbing one rung at a time, then suddenly finding their ladders flooded with names that belong to no one. Anti-Boost exists to give the ladder back to them. That is why I still stand with this system, even as I see its cracks.

But there is a question Riot has not answered. If an automated system both detects and punishes, with no independent appeals mechanism, what protects the innocent when the algorithm is wrong? Riot is scaling Anti-Boost — adding match-level detection of boosting signs, continuing to cancel fraudulent results. That is the right direction. But greater scale also means more automated decisions, and more automated decisions mean more potential errors without a pressure-release valve.

I once sat in Kazan Arena and watched South Korea beat reigning champion Germany 2–0 on a night no one expected. I wrote that history is written by audacity. But history is also written by moments no one sees: a referee blowing the whistle at the right time, an editor checking sources before publication, a publisher daring to disclose its own flaws. Anti-Boost, at its best, is one such moment. At its worst, it is a machine without brakes.

Those who write about transfers don't sell players; they sell unfinished stories. Those who write about governance are the same. We don't sell punishments; we sell questions about the line between protection and control. 296,416 accounts is a respectable number. But behind it lie 296,416 times Riot decided on its own who was right and who was wrong without anyone else's agreement. That is what I will keep tracking in the disclosures to come, because if one day a real player is wrongly punished with no road to appeal, that number will no longer be an achievement — it will be a crack in the trust that a billion-dollar industry cannot patch with a single update.