VALORANT Anti-Cheat: Cheater, Reported!
Hey it’s Paul “Arkem” Chamberlain, your VALORANT anti-cheat lead here again with a quick spotlight on the importance of player reports.
One of the most powerful tools in the fight against cheaters is player reports. Reporting gives players a way to directly tell us, your faithful anti-cheat team, what suspicious activity is going on in their games. Unlike Vanguard’s automated systems (Riot Vanguard is our anti-cheat system) reports bring human insight to our anti-cheat efforts at an otherwise unobtainable scale and bring with it a direct look at everyone’s experience in game. It’s really important that players use the in-game reporting tools everytime they see something suspicious, it’s the best way to help us keep the game fair!
I promise some spicy numbers at the end.
REPORTS LIFE CYCLE
Reports feed into many of our team’s processes both automated and manual. On the automated side, the Vanguard backend uses reports to decide whether or not a player should receive additional scrutiny, such as a higher intensity game integrity scan. Vanguard also uses the number of reports (more specifically the number of unique reporters and number of games in which the player was reported) as an indication of confidence in its findings—which helps us ban players faster and, in many cases, without manual review.
Our anti-cheat analysts also use reports to drive the manual review process. Each day the analysts (sometimes guest starring myself) review the most suspicious players, usually starting with the most reported players. These manual reviews help us find cheats that aren’t automatically detected that we then use to improve Vanguard so that automatic detection (and bans) will happen in the future. We also use these manual reviews as an opportunity to remove the most disruptive players ahead of schedule. We often use “ban-waves” to keep cheat developers guessing as to whether or not their cheat is currently detected but we’re happy to early-exit the most disruptive cheaters to better protect the game.
Finally, we use reports as a measure of the health of the game as well as the effectiveness of Vanguard. If the overall number of reports is going up then that’s a sign to us as developers that players are feeling bad about the integrity of the game (regardless of the accuracy of each individual report). Likewise one way we can assess the effectiveness of Vanguard is by comparing the percentage of players reported for cheating versus the percentage of players detected by Vanguard.
BY THE NUMBERS
I know everyone likes stats, so I asked our anti-cheat data science friends to find some juicy numbers for you all to see.
First of all, 97% of players have never even received a single report—no one has ever found their behavior suspicious enough to comment on. Of these 3% of players that have been reported for cheating, more than 80% of them have only ever been reported by a single player. 90% have been reported by fewer than 3 players.
To put this another way, only 0.6% of players have received more than 1 cheating report and only 0.3% have received 3 or more. However, reports and cheaters aren’t perfectly correlated, many reported players are innocent and not all cheaters get reported before they’re banned. Right now only 53% of banned cheaters were reported before their ban and only 60% of players with 20 reports get banned after review.
But even without a perfect correlation, reports are extremely helpful.