Smurfs: The Good, the Bad & the Ugly - Q&A

General Discussion
84

That’s more or less what I thought. To take an example, it won’t make much of a difference to a black belt karate-ka if he faces a yellow belt or a green one. And a gold player will feel quite strained against opponents from just one league above.

However, I think that a master player would know better than the lowleagues about the progression in the leagues close to masters. But it have to focus on those to notice : and while it would indeed be easier for a master to do so upon casting a replay or a stream ; when playing himself, in the fog of war, it might be way harder to differenciate a gold 3 from a silver 2 since both won’t have enough micro/reactivity, and so will lose in the first minutes anyway.

So far it seems that those who opposed this indeed were smurfs themselves, regularly or not. :mag:

I don’t see how having a matchmaking doing a better job at match-making could be seen negatively from regular users. The only possible hinder of having a freelose-detection system would be to Bronze-to-Masters tutorial makers. But considering that a new account requires only 10 days to get (more vespene gas or additional pylons put aside ^^), and that those account could be led to bronze MMR in rather few games, they could get around this without too much trouble. :thinking:

Getting a second account is extremely easy, you just have to play on another server. As for having it one league or two below yours, as long as it’s not deliberate (for example people may really be 2 leagues under theirs in offrace situations), that’s not a issue. The issues come with the freeloses, that’s the heart of the problem.

1 Like
85

Watching the replay post game yes its easier to tell. But in the actual game itself it is near impossible for me to make an accurate determination of if this is a silver/gold, or a gold/plat player. Nothing these players can do will even remotely tax my abilities so they are all going to feel on average the same to play against.

A good player will naturally make someone in plat look like a gold leaguer. There is just nothing the plat player can do to me when i am significantly better at every aspect of the game. So they will feel like a gold player. But if the plat plays vs a gold those small differences in skill will come through allowing the plat to win the majority of his games while smurfing. Someone of higher league in the low metal leagues just does not have the same perspective that an actual player of that skill has.

Yeah thats what I ment. Like as long as you aren’t deliberately keeping it there I don’t see a problem with it. Like just for example. For me to keep an account below diamond 2 (hell even diamond 1) I would have to actively try. No matter what I do I can usually crush anyone below diamond 2, even if I am doing just ridiculous builds. So nobody should have an account more than 2 leagues below their real account in my opinion.

I could for sure keep an account in Diamond 1 legitimately, if I am doing builds out side of my norm however. Diamond 2 I feel like I would have to be doing some very weird stuff. But I’ll grant people the wiggle room and say 2 leagues below is fine even though I feel like 1 league below is probably the more realistic and fair compromise.

1 Like
86

So if you’re master 3, and that you feel having a diamond 1 or 2 account for occasional builds experiments ; you’re not meaning two leagues, but two tiers below your actual one. The precision is important !

Because, if on your main race, two leagues below (for example gold 1 if you’re diam 1) would seem unlikely if not deliberate.

For my part, in my current 1.1 procedure for my smurfs reports, I’ve considered that two leagues (4-6 tiers) or more below your past level, with your main race, can’t be anything but deliberate. For example if I spot a Terran in silver 2, while he has been several times platinum in the past, I do straight categorize him as a smurf. And I think current criteria are good, as it often happens that when someone is a smurf, he often meets several of those criteria simultaneously. :mag:

This isn’t true for offraces though. I’ve now screened a lot of profiles, and it seems that for Protoss or Zerg mains, Terran is a bit more awkward to handle. And so it’s not infrequent that Protoss or Zerg players have a terran offrace two leagues below their main. So as long as its offrace, I’ve not set any league floor limit. :slightly_smiling_face:

87

Yes sorry, my fault. I meant 2 tiers. 2 full leagues is absurd. XD I would never lose a game to a plat 3 player. I could lose to a D1 though, D2 maybe.

Yeah agreed. What I mention mostly goes for the players main race. Offracing is a bit different. The races are quite different from each other. So being a league or 2 below your main race isn’t to weird. Though I feel like you will eventually get within at least 2 or 3 tiers of your main race if you played more than 100 games.

I for example am M3 toss, D2 Zerg, D3 Terran. I’ve got close to 8k games toss, 400ish games zerg, and like 200-300ish games terran. I recently switched to playing all the races more often. My terran lags a bit but I mostly chalk it up to the lack of experience with them. I have no doubt with more games Ill be equal skill with zerg and terran. I doubt I’ll ever catch up to toss though unless I just stop playing toss.

As for this I feel like it would depend on the time gap between when they were plat to their now silver level. I could see someone taking a year break and coming back much much worse than they were. But they will likely regain the skill relatively quickly. When I was D3 I took about a 5 month break. When I came back I was low plat level for sure. But after like 2 weeks I was back to D3.

1 Like
88

At any rate though, keep up the work bro. Keep voicing the concerns about this. Blizz needs to do something about it, and screw everyone who thinks this isn’t an issue, they are likely part of the problem to begin with.

1 Like
89

I think blizzard should solve the problem by changing your MMR proportionately to the length of the game. That way Freelosing wouldn’t affect your MMR at all, or at the least would be much more time consuming and difficult to do.

The problem with this is that people would just instantly surrender vs players or match-ups they don’t feel favored in. Also, it opens the opportunity to purposefully draw out games you’ve already won to farm a greater amount of MMR. Most solutions offered to ‘solve’ Smurfing just create bigger issues. The only viable solution is to just keep playing and never assume/care that your opponent might be a smurf. Eventually you’re going to hit an opponent that is better than you.

90

There is an easy way to fix the problem. Increase the que after a game that lasts less than say 3 mins. Then keep increasing the que time the more games consecutive games after that, that last less than 3 mins. Example: after 1- 1 min que time, 2- 2 min que time, 3- 3 min que time, ect ect ect. You can apply any value to it you want, just an example.

This would not impact cheese games much because most cheeses arent over in less than 3 mins. For the ones that would be, it is a shame but I doubt they will fight it more than once in a row so its a necessary casualty.

It will force the smurfers to be more committed to the smurf. Either they gotta wait out the 3 mins every game they want to auto lose or wait out the ever increasing que times. Yes obviously this is not going to stop everyone. There will be those that suffer through this. But it will deter many of them I am sure.

Its either this or blizz needs to take reports more seriously. Actually investigate reported players and if they see a correlation give the accounts temp bans. But this will never happen so option 1 is the only way.

Blizz even has a similar system in place already that they could easily apply to que times. Currently if your match lasts less than 3 mins you get no xp from that match. So they already have a system in place that has the same basic principle. All they need to do is apply it to que times.

91

Cheese games can be decided in under 3 minutes. Losing is punishment enough for people who didn’t scout and didn’t notice a cannon rush for instance, I don’t think they deserve a longer queue for it. You will never convince me that there exists such a thing as a ‘necessary casualty’ relating to this conversation.

You guys are suggesting developer work that will both not solve the problem but also create additional problems.

1 Like
92

Even cannon rushes hardly ever are decided in 3 mins unless the person getting rushed just doesn’t want to deal with it at all and leaves the second they see a forge.

The only cheese I can think of that can actually end the game this quick is 12 pool in PvZ if the toss doesn’t wall. But even then if the toss try to hold it at all it will likely last longer than 3 mins. But even if you arent for the 3 min then make it 2 or 1. Something, literally anything would be better than nothing.

93

3 minutes isn’t a good cutoff for a freelose. I’ve been screening for those since quite some time now, and though when I can I just look at what happened build and army-wise, I’ve never seen a regular game where the opponent had realized he had a reason to surrender under a minute. Each time the game is lost with a duration of 0:xx, it’s a freelose. Earliest motivated quit from raging players were upon scouting, which can happen around 1:30 (which is my cutoff for the API searches).

I’d set a freelose definition for a game lost < 1:00. A guy who leaves in 10s doesn’t do it because he’s being cheesed. :cowboy_hat_face:

As for adding 5 minutes queue time for each consecutive freelose, script wise that would be quite simple. It’d be a bit longer to set in place a system to detect games lost < 1’, but I don’t see it taking that long for a Skyrim modder to do, nor having a noticeable performance impact. Just make the script run once (for the player) at the end of each match and we’re good to go (plus as Nordy said, there’s already a similar script for XP). For SC2 devs, that would probably be child’s play.

There are people who leave upon scouting a counter build, so the leave timing of such games is their scouting one… or rather the moment where the opponents first hits. Can’t happen under 1’30 from what I’ve seen. 3 minutes would be the timing for a player who actually tries to defend. :mag:

Thanks Nordy. I’m glad the topic receives some attention as there’s work behind it, but it’s even better to feel supported. :wink:

94

I think your confusing the statement “its much much easier to tell someone’s actual skill level the higher you are” with the statement “you will be able to accurately tell someone’s skill level if you are a master’s player”

95

To sum up :
— You’re saying that the higher level you are, the better you are at knowing what skills belong to what league (but not necessarily at noticing when being matched against).
— Nordy and I do state that the lower level you are, the more keenly you will feel if you’re playing an user above your level (but not necessarily at knowing from which league they are from).

It seems to me that those propositions aren’t mutually exclusive. :thinking:

96

Alright, back to the main course :

SMURFS REPORT W33-35

SILVER LEAGUE

Using a randomized procedure (similar than for the gold league) :

How to read : cf. first post (part II)
Method used : cf. dedicated post.

Quickie :
Over 105 users at silver MMR :

  • 20,95% of confirmed smurfs
    (13.33% of freelosers + 7.62% of non freelosing). :cowboy_hat_face:
  • 79.05% to 56,19% of regular players. :slight_smile:

Some thoughts about this :

  • Contrary to what I thought, this data suggest there isn’t more smurfs in silver than in the other metal leagues. Random variation due to the minute statistical power put aside, for the moment evidence suggests that the gold players would be the one suffering the most of smurfing issues. Aside the currently modest sample size, two other factors should be taken in account though
    ◦ The fact that I did not set a minimal career games to reach silver league (as I believe one could perfectly be placed in silver with 50% winrate right away). So that makes one less tool to spot smurfs, and could translate into less non freelosing smurfs detected than there really are.
    ◦ The fact that for the moment, there seems to be more freelosers.
  • This put aside, the results are still in the same range than what was gathered in gold, and in my observational platinum report. This suggests our estimates aren’t that inaccurate after all.
  • About the method used : finally, I didn’t reduce the amount of MMR randomizations compared to v1.1, as I feared it would decrease the screening power of the survey. I did very minor changes regarding discarded players, as I do not discard users anymore if their profile already gives me enough info to conclude ; and do also classify users as « undefined » when the data is to scarce for me to reasonably conclude. More info about current 1.1b procedure here.
  • A lot of players who met one of the smurfing criterias in fact met several criterias simultaneously. For example, one that would have been gold with too few games career would also have a suspiciously high winrate, or on the contrary have ≥ 3 freeloses. This makes me think we’re in a good place regarding those criterias, and I feel satisfied with current version.

FUN FACTS

  • If an user reaches a league with too few games in his career, that can allow to confirm him as a smurf (for example platinum MMR with 10 games in his whole career)… but it seems the reverse is also a hint. For example I’ve confirmed a freelosing smurf who had 12 000 career games while still being around silver3-bronze 1. Even extremely non gifted users would’ve gotten at least gold by that time. I’ve seen regular players staying in silver for more than three years though, so this should be only considered as a hint.
  • It also seems that an irregular progression (bronze → gold → bronze → gold) is a good clue.

Now for the funny part :

Once upon a time, was a lad known as « Ok ».

Ok was a high league player, but stagnated. So, one extremely gloomy day, he decided to stop playing in diamond 1, and to freelose 1000 MMR+ in order to feel like the man again. And it worked wonders, he finally could get the 20-30 games winstreaks he had always dreamed of. By continuously slaying unsuspicious lower leagues, he felt he had regained his virility. He felt like a Flash among the men.

But good things aren’t meant to last. One Saturday in the evening, he was matched against another lower league player, that we will call Pony. Ok didn’t noticed, but the pony seemed shaken by his manly apparel, as if he had seen a ghost.

Full of confidence, Ok deployed a splendid exemplary of skill : with his 235 APM, he hit every macro timing magnificently. He even killed the opponent’s scout who was trying to harass his CC building SCV. Such a foolish attempt !

But upon preparing to lay waste on the pony’s mineral line, Ok noticed the wall was made with three supplies. Suddenly, he didn’t feel ok anymore. An alert immediately sounded : his base was being assaulted by a small squad of marines, and they weren’t friendly. Listening only to his 200+ APM fingers, he got back, and tried to temporize an SCV pull. But while doing so, the second Pony’s SCV had (barely) had the time to land a bunker. He had just lost the game to a vulgar pony. And Ok wasn’t ok with that at all : he did ragequit, and by that had his first real lose in a while.

What was the difference between the pony and the 20 previous players ? Skill ? Nah, otherwise the pony wouldn’t belong to an intermediate league. It was information, as contrary to the others, the pony had learned to recognize smurfs right away. :cowboy_hat_face:

Moral of the story : would the devs let the ladder be ruined by smurfs, we regular players would still have a chance to deal with them by ourselves. Know your enemy, goes the saying. Even more so in a strategy game.

PS : first post updated. See you in about three weeks for Bronze league ! :smiley:

97

And here we go for Bronze :

SMURFS REPORT W36-38

BRONZE LEAGUE

Still with the randomized procedure :

How to read : cf. first post (part II)
Method used : cf. dedicated post.

Quickie :
Among 105 users at bronze MMR :

  • 28,57% of confirmed smurfs
    (22.86% of freelosers + 5.71% of non freelosing). :cowboy_hat_face:
  • 71.43% to 41,91% of regular players. :slight_smile:

Some (important) thoughts about this :

  • I found 28.57% of smurfs over the 105 users I randomized, with 35 for each league tier. Would those league tiers be around the same size, one could extrapolate that proportion to the whole bronze league, however that is not the case : there were way less users in bronze 3 (500 players, 49% smurfs) than in bronze 2 or 1 (1600 and 5700 players respectively, 17 to 20% smurfs). So for my stratified sampling to be correctly interpreted, I must weight the tiers differently according to the proportion of players. So here are the corrected estimates :


Corrected by league’s population, there would be :

  • 19.74% of smurfs.
  • and 80,26% of regular players in the whole bronze league.

This is the estimate I will retain from this report. :pen:

  • Though of course imperfect, I still feel my criterias are a good compromise between specificity and accuracy, and so am satisfied with them. However, as I was closing-in to bronze 3, I was confronted to situations where there were less and less players by MMR point, and so I had to define rules for the randomization to apply efficiently. Those changes are minor, but for those interested, can be found here as version 1.1c.

FUN-FACTS

  • Surnames pseudos/battletag seem to be common among smurfs, and that’s a specificity as the main benefit of pseudonyms for regular players would precisely be to conceal their usual name. So you’re matched with a player named John, Walter or Maria, that should rise an eyebrow.
  • As I was surveying bronze 3 this week, I stumbled among a rather typical profile at first : he had a name that sounded like a silly pun, and with 2,400 games in his career, he felt too experienced for a bronze leaguer. Interestingly enough, his clan name also felt similarly silly. So I started to dig into his profile, and wasn’t disappointed to find an impressively high amount of freeloses (more than 30-40 I think). But what caught my eye, was that he had freelost to a guy of the same team… and when he won, that was the very same dude providing him a freewin. So I checked the clan, only to find that the third member also was a smurf : we had a silver-gold and a diamond who smurfed in bronze, and a master who smurfed in gold. I had found a smurfs clan. :roll_eyes:

FINAL THOUGHTS

That’s it folks, I said I wanted to survey the 4 metal leagues, and after many weeks, the objective is complete. I must say that task required for me to get knowledge and tools I didn’t have a first, and that coupled to the hints I had to define, and the hidden nature of my targets, I sometimes felt like some kind of secret service analyst (lol). Anyway, it was an intellectually stimulating challenge.

  • Overall, with the exception of Bronze 3, there aren’t more smurfs in Bronze than in the other leagues. However, Gold put aside, we’re around 20% among the metal leagues, and this is a lot. Someone said « This isn’t League of Legends », but with one player over 5 smurfing in those leagues after only 2 years of F2P, who could say if one day SC2 won’t find itself in that very same spot ?
  • There are about 50% smurfs in Bronze 3, and what happens there confirms what I thought about circumstances where the smurfs ratio becomes high : they end up playing each other. The smurfs team is an illustration of that, but the feedback one smurf had the honesty to provide on this thread a few months ago (where he said he did play other smurfs, and freelost in that case) also is.
  • I have found the smurfs of some members of this very forum. I won’t give names of course, as I don’t want to have anyone individually targeted, but what struck me is that those were often among the most virulent individuals towards the balance, or daedgamers (users claiming since several years that SC2 is a dead game), or trolls. When I saw that, I suddenly understood that those people claiming the game was destroyed were in fact first hand witnesses, as they were among the very players who were trying to ruin it. :bomb:

Lastly, the previous report didn’t raise any reaction. However, with it the thread did attract lots of views (more than 320), and that even while being in the abyss of a three weeks inactivity. That’s remarkable, and I can only hope some of the eyes apparently so interested were those of the devs. If you guys ever read this : don’t assume F2P won’t bring to SC2 what it did to LoL. Don’t let smurfs ruin the game from beginners to intermediate levels, or you will lose the former.

Now, of course it could be interesting to survey the diamond and masters leagues, possibly to redo platinum. I could also do a best-of about the hints you could get to spot a smurf on the loading screen. And I’ve got something about the different motivations behind smurfing that might interest the devs.
But some views on a thread, without comments, without likes, and with the blue ones remaining ever silent on the matter, aren’t enough to renew my motivation. So unless there’s some change, I can let my work as is. I will still wander around to exchange as any other community member though. And, since I’m now trust level 3, will use the opportunity to transform the first post into something really neat. :star2:

See you around, folks !
:smiley:

1 Like
98

Concerning Leaver botter on EU where a smurfer ruins my RT games by leaver/afk botting

  1. FIX YOUR IN-GAME REPORT SYSTEM. LOOK AT IT
  2. Punish mass leavers/AFK. Look at the ZERO APM. Holy @#$% give me one reason why there is no penalty for this
1 Like
99

They did delete and lock all the posts in that thread except the first two, so either because of the replays or something else happened after I went to bed

I didn’t get any suspension, but it shames me I can’t even discuss leaver botting. The account name of the person was removed from the thread. Massively grieving other players should be punished, not ignored and censored when it’s reported. You disappoint me, Blizzard

100

Told you there should be nothing linking to the individual himself ; otherwise he could be targeted by the most angry readers. I might add that there are a lot of users around here who don’t see with a keen eye when their doings are denounced… and so if the smallest rule is broken in the process, they will report it so that the author gets silenced.

Now, you can’t denounce that individual ; but we can still discuss the case without breaking any rules. And that by debating over the phenomenon in general ; and anonymizing any source material.

I took a look at it, and I must say this user’s procedure was quite unusual. However, after careful inspection, I’m not sure there is botting involved, and I’ll explain you why below :

First, the profile :
https://i.postimg.cc/bYQR8bYq/2020-10-04-AFK-smurf-Profile.jpg

  • The key point here is that he has zero APM, but not 50 resources. Meaning that he keeps mining, but does nothing of it. That’s one way of ensuring defeat, though in teamgames there is no guarantee of it considering the others players might still outdo the opponents (as seen in the upper right example). Anyhow, the point is that’s an attempt at freelosing, but without insta-leaving.
  • The resources left are about the same from one game to another. This might be due to the mining being a bit random rather than him not leaving at the same time, and it’s at ±5 minerals each time which is extremely close.
  • The guy is master, has been twice GM. Yet he has a bronze team league. This suggest the user has the skill to go way beyond his current ranking, and loses deliberately in order to get matched with lesser opponents.
  • His teammates aren’t the same everytime. So its random team 4v4, and he is doing his freeloses alone.
  • What is not shown here is that the games he’s involved in can last 20-23’ minutes, but he still queues them about every 10’. So he not only AFK but also leaves them before the end.
  • Also, he already has all the teamgame achievements, so this is not farming.

Ingame :
https://i.postimg.cc/Bnhgk0SD/2020-10-04-AFK-smurf-ingame.jpg

  • From what I’ve seen, he slightly moves the camera/screen downwards at the start of each replay. On the second or third replay you initially provided, there was others truly AFK users, including one who apparently was on the phone because he started playing a few minutes later. And for this one, the camera didn’t move in the least at the start of the game. However, the camera move is nearly the same each time.
  • On the illustrated example, he missclicks the camera which as you can see above ends up well North from his base. That means he’s there at some moments for sure.
  • Side note, there was another AFK user in this one. Which suggest this issue isn’t infrequent… :roll_eyes:

From the whole analysis, we can deduce that :

  • He’s imputing manual commands at the start of some the games.
  • He does not only AFK, but also leaves them early, at about the same moment each time.

So my conclusion is that he’s there at the beginning of some games for sure, but not all, while the games’ ends look somewhat automated. So I can’t rule out an external program, particularly for the end.

As for the why, well it’s made to lower his MMR so he can play lesser users (and as you could see decreased in leagues progressively, becoming gold, and then bronze later, which suggest he left league and then got placed again).

And the reason why he does AFK 5’ instead of leaving right away, IMO could be that this way the other users could not take control over his units, and that making sure that his resources wouldn’t be shared with them. So some sort of method to ensure the game is a defeat, and that his MMR indeed lowers.

And I who thought I had seen it all regarding smurfing… :sweat_smile:

101

So after your, I must say thorough investigation concluded there is no botting involved? Just some smurf leaving games after 5 minutes? Well damn. The guy must never sleep then

I mean, I went to bed and he still at it for at least 21 hours, as far as the match history is able to go back

102

I think he’s there at some moments at least. I am not sure he hasn’t a sort of timer to close the game (which doesn’t necessarily requires a bot, but could be eased by an external program).

As for not sleeping, he could just alt-tab from SC2 to windows at at the start of each game, as suggest the slight camera moves. So he could be watching Twitch, reading at his emails, literally anything and then just closing SC2 process around 5 minutes. :mag:

103

Hmm. That’s disturbing. Looking back at what I saw, he left most games at nearly the same time as well, at about 5 minerals each time. This is smurfing, and he sometimes does some inputs by himself… but that could be him just watching at how things are going from time to time, and doesn’t rules out some parts being automated. :thinking:

To make sure, I took a look at the post-game debriefing you first linked (the one of 11:50 Old Estate), which is a different one from my Fortitude one, and the amount of resources mined was 1747… for 1747 in Fortitude, and 1742 in my 26:52 Old estate.

So this is too close to be done manually for 21H long. There might be automated parts to this 4v4 smurfing after all… :neutral_face:

1 Like