I see a lot of posts saying that players are constantly running into smurfs and it is ruining games. Sometimes people are claiming they see smurfing players every other game. So first, I want to point out that it's extremely unprobable you are seeing smurfs at anywhere close to this rate.

Let's do a estimate of how many smurfs you might be seeing, using player distributions and basic random sampling. Method criticism is welcome.

According to Riot's player distribution API numbers for March 2021, 11.8% of players are above gold (this is the most common reported range for "the smurfs in my game are from ____"). Lets be generous and say a third of all of those players are all playing on alt accounts at the same time. Let's also say that by some method none of these smurfs have the ability to rapidly climb out of their rank back into plat+ (either by constantly playing derank games or rapidly making new accounts). I'll ignore that this means a significant portion of our 'one third' will not be playing smurf level games because they are either playing unrated games to unlock ranked or playing bad games to maintain rank. Now lets say those smurfs enter the iron through gold player distribution, with each division getting its fair share of smurfs according to how much that division makes up of iron to gold.

Since you are not a smurf, each game has nine other players that could potentially be a smurf. Let's take a random sample of nine players in each division's sample - any game where one of those nine samples is a "smurf" sample, the game will be flagged a smurf game. I want to add that this is pretty unlikely since smurfs more often have higher MMR and have less than a random chance to be selected for match made games at that rank, even if they are at a "smurf" rank (but hey let's just give benefit of the doubt). Over 1 million sampled games, we get:

Probability of smurf games in iron distribution: 5.8214% (about 1/17 games)

Probability of smurf games in bronze distribution: 9.6085% (about 1/10 games)

Probability of smurf games in silver distribution: 9.6085% (about 1/10 games)

Probability of smurf games in gold distribution: 5.9822% (about 1/17 games)

But you don't have to actually take a probability sample to understand that players who get 30+ kills are very rarely smurfs. Iron to gold players are usually that rank because they are not consistent. A player top fragging one game might be bottom frag the next and barely be able to maintain a positive win rate.

Not only that, but all non smurf players have the same chances of having smurfs in their games as other players of their rank. That means while these games feel bad no matter what, they are weighing on players the same. They are not "preventing you from climbing" over anyone else because everyone else is also dealing with them. Actually, statistically slightly less than half of those smurf games are being handed to you when they are on your team.

But I didn't even have to make this post, because people with the real information have already been looking into it for a while:

“Most smurfs will struggle to stay at low ranks. It's very hard to have an account maintain a rank, while also smurfing on low ranking players. It's kind of a complete counter to how ranked works, because if you are truly smurfing you would increase your MMR and no longer be smurfing. We can detect this very quickly, for example TenZ getting radiant MMR in less then 20 matches. It would happen for anyone below radiant in even shorter time (the extreme high ranks, like radiant, take the longest to shift MMR into). This is even harder in our system because we don't put barriers on ranked/MMR - and match you based on your MMR. On top of us having a performance MMR which can tell if you are performing outside of your current matches.

So then if players are going to be pushed to a higher MMR, due to smurfing and having "insane" games against low ranking players, they would have to make an another account. Which again people will do, but becomes less frequent because at some point making multiple smurfs takes a lot of time not playing the game (or money to buy a botted account that will just get banned eventually). It's just another barrier that reduces how many people smurf.

About 4% of all ranked players are above plat, about 7% of players are plat.

I've seen people say there are a "ton" of smurfs, when we investigate we see that players just have a good game. I've seen players with consistent 10 kill games, randomly, have a 30+ kill game for reference - these are players with less then 40% winrate as well.

Even if every single player above plat smurfed, they wouldn't even be able to fill half of all games that occur in plat - let alone all the ranks because they are only 4% of all players. Yes smurfing occurs, I'm not denying it, and we also have systems in place to try and reduce their frequency and how long they can sit in low ranks. I think it's just overblown, because in tac shooters players can have insanely good games. I am about Plat/Diamond, and play against immortal/radiant players every day in playtest - even I have games where I'm 20-30+kills in those games(against players way above my skill level).

That being said, the perception is the problem so we still do investigate smurfing concerns in case there are issues. We won't ever stop, and hopefully we can help players feel like smurfs aren't as common as they think!

https://www.reddit.com/r/VALORANT/comments/m3lbsp/i_feel_like_actually_bad_at_the_game_now/gqqamj9/?context=3

https://www.reddit.com/r/VALORANT/comments/m3lbsp/i_feel_like_actually_bad_at_the_game_now/gqpszdq/?context=3

Criticism Edits:

• If players have multiple smurf accounts, the number of players in low ELO is artificially slightly higher than the true non-smurf count, even though there are not more actively playing smurfs since you can only play on one account at a time. This would mean there are an additional X amount of false players from anyone who's bothered to make more than 2 total accounts.