So I played a few games with a proper “smurf” last night. He just joined our group and started playing, obviously on an alt account. It took us a couple of games to realise. He wasn’t one of these people who announce to everyone they’re a smurf when really they’re like low diamond and can’t function without a mercy pocket.
It was a low to mid gold game and this dude could 1v6 the enemy team as hog without breaking a sweat, and at one point used his alt to force them all back through the spawn door.
It’s kind of changed my perspective on the game a bit. When people think they should be in GM and they’re stuck in gold. There’s a bigger skill difference between the two than they probably think.
I still think the matchmaker is broken as the majority of my games are either roll loss or roll win which isn’t much fun at all. But having seen the skill of a high level player I don’t think I’m ever going to be at that level no matter how many hours I put into the game, and I suspect a lot of these “GM stuck in plat” players are in the same boat. I guess we can’t all be the best otherwise the best wouldn’t mean much at all?
How did you guys come to the realization that he was playing out of his rank? Did he eventually admit it? Did you ask him (after doing some crazy play)?
Just curious honestly.
Have you considered going back and re-watching his gameplay to try and grasp his decision making?
Dave, I’m glad you’ve had this experience.
The difference between Masters (3500) and GM (4000) is probably equivalent to the difference between 2000 and 3000 SR (I know it seems like an exaggeration but I assure you it is not).
And in GM, Every 150-200 SR is effectively a new rank if only because you’replaying against players at the very top of the ladder who are all exceptional at their roles/heroes.
I’m not interested in arguing with you on this topic because we’ve discussed it previously.
I do want to ask if I may…
As you mentioned, the player was a great Tank/Hog.
Would you say the games were unfair for your enemies?
Would you say that a game with a group of players equal in skill to him would be a balanced match?
The reason for the second question is, in my opinion, the higher you climb, the more balanced the matches feel and the more rewarding it is to win because everyone becomes more proficient at their roles and the probability of consistent performance becomes very apparent (as compared to casual performance).
And no… Stomps are not fun at all. Nobody enjoys them.
Actually I HYPER disagree with you (I could elaborate on this). Depending on what role you are, its always about knowing what to do in what situation. There is a great amount of skill required for Overwatch, but more than skill is the second to second decisions that you make, and this is something players like myself, Abdullah9000, Violence, and some of the other Comp Forum regulars are trying to preach.
But the beautiful part? ANYONE can do it!!
I can very much assure you that no GM player stays stuck in platinum. Yes… it may take them a little bit of time to climb out if they have some unlucky games, but overall, they will not remain in low elo because of the decisions they make having such drastic impact on the outcome of the game.
Correct… As more people climb, the average level of performance to maintain/increase in rank, increases as people (and the community) become better players.
Not everyone can be GM, Masters, Diamond… But… ANYONE CAN!!
Including you!!
POST-SCRIPT
Real DPS Smurfs don’t ask for Mercy pockets because we’re gonna get kills one way or another. But if you pocket us (against a pharah for example) you’re simply making everyone’s life easier…
How are you defining Handicapping here? Second, since you are now the one making the claim, I’ll ask you to provide accurate, fact-based evidence that there is zero “handicapping” in the matchmaker. This should be fascinating, since short of seeing the running code base itself and relaying/detailing the relevant portions here, there is no possible way for you to know/prove this.
How would you know unless you worked as hard and as long as a GM? Wouldn’t you actually have to put yourself on the regimen of the average GM, and do that week in week out for as long as it took the average GM to gain that level of skill (barring some obvious debilitating mental or physiological limitation)? You don’t think there were a considerable portion of GM’s who thought they couldn’t possibly be that good when they first started out?
It’s only rigged in that if your stats are lower than they should be for your rank and role, you get forced losses. In a team game what other mechanic is there to keep people in the rank they belong? In a solo environment, you would either win or lose, but there kind of needs to be this mechanic in a team game otherwise you could just get boosted really easily.
The flipside that nobody ever wants to acknowledge after a loss streak is that if your stats are better than they should be for your rank, you get FORCED WINS. The system goes both ways.
I have a soloQ only account where I only play when I’m playing well, aim is on point, don’t play on weekends or late at night, etc and I have hovered in a 200 SR range for years. Never below, never above. But on my other accounts where I group, play when my aim is off or I’m tired, etc I’ll go on much bigger swings. If your aim is off, SIGN OFF or play QP. The system can tell that you’re putting up bad stats and will force losses.
The only purpose of SR is to serve as a human-interpretable rank. It does nothing beyond that. Until SR is a direct metric of the player’s skill and performance, it is not redundant.
Is a 50-50 coin “rigged” or not?
It’s quite strange that “forcing” fairness means “rigged” under your mentality. You’re asking for a system that wouldn’t even check to see if the coin is “fair.” Every person on the ladder is playing by the same rules and high elo players do naturally progress.
It’s not “cherry-picking,” it’s a detailed prediction based on the player’s historical and current performance. The closest thing I’ve experience that parallels this is an intramural soccer league. Each team was allocated 2 players from the collegiate team for the sake of balance instead of stacking them all on a single team.
Random != balanced. If you don’t like balanced games at your skill level, feel free to jump in custom games with higher elo players and get stomped.
It’s nothing near anti-competitive. You compete against people at your skill level–that’s it.
Is a 50-50 coin “forced odds” or is it “fair odds?” We know people aren’t being “held back” by “rigged odds” because a higher elo player can easily climb on that account. “Rigging” is just a lame excuse for people who refuse to accept their skill level.
No, it doesn’t. The mean shifts and the ladder moves. This is why it’s harder to climb now than in previous seasons with more casual players. That is, as causal players left, the game’s playerbase shifted toward long-term veterans that are better players on average.
Adding order to the system decreases entropy. SR isn’t finite nor is it controlled like a currency. The distribution remains normal with 60 million or 10 million players. That’s how the ladder works. If you introduce 1000 alt accounts at a higher SR, all it does is displace lower SR further down the ladder. It’s percentile based, not a hard limited number of slots.
Technically accounts; I’m using the terms synonymously here. It doesn’t matter if there’s a 1-1 player:account ratio or 1-3 ratio.
It must. Different numbers of players utilize the system every day, week, and season etc. That’s a basic input that must be ordered by the system.
It doesn’t have to “try,” it is a natural product of ranking players. You’re either below average, average, or above average in comparison to your peers. The analogy for a class exam: MMR = points on a test; SR = peer percentile ranking based on the number of points you got.
You don’t need a full reset–it accomplishes nothing. Everyone on the ladder is taking the same tests all the time. You can either study and increase the number of points you’re getting, moving up in rank against your peers, or you can sit around and think you don’t need to study.
Right–so if the curve isn’t changing, how is the distribution increasing in entropy?
There are several reasons why alt accounts wouldn’t endanger the fidelity of the ladder:
The ladder is reset every season and the ranks are only distributed to accounts that place/play. Not every alt account is going to be played
Player performance, on average, remains stable–meaning people end up in the same place over and over in comparison to their peers (like OP, who has been ~1700 for 7 seasons)
The number of alt accounts are greatly outweighed by the number of mains. That means their effects on shifting the mean is negligible.
I think many players struggle to understand how massive the skill gap is between ranks. For example, a Top 500 Ball–named “Ball”–did a “no-shooting” challenge and literally farmed low Masters players outside their spawn:
Imagine what this player could be doing in Bronze/Silver/Gold etc.
Being placed at a disadvantage.
If “handicapping” exists, players–regardless of skill level–would get randomly hard stuck in elos they don’t belong in. That doesn’t happen. A GM player is never going to be hard stuck in Silver, right?
If it’s “essentially random,” the probability of any account obtaining a rank should correlate to the probability density of that rank. Since it’s a normalized distribution, we can actually test that hypothesis for being “random.”
Meaning, anyone with multiple accounts in GM+ (like streamers who do unranked to GM series), we know it’s extremely unlikely to be occurring by chance.
Using some old (but otherwise methodologically sound) data from Season 9/10:
-the average SR was 2266
-the standard deviation was 610 SR
If only 20 streamers (a very conservative estimate) made alt accounts and climbed back to GM, the statistical likelihood of that occurring by chance is < 0.00001 using a two-tailed z-test.
i.e.–they magically aren’t being “handicapped,” they aren’t getting hard stuck, and it’s not random.
Because platinum players don’t even play the same game as we do. Platinum players consistently make so many errors that a GM player on a fresh account is going to consistently exploit and their individual contribution will kick in the performance based SR so they physically cannot stay stuck in low ranks.
You misunderstand the question (maybe that’s because you’re answering on behalf of someone else, which is a little strange since you can’t know why the person made the statement they did). The answer you’ve given here is a conclusion based on your observations. It does not prove the statement made by the person to whom I was responding. Quite clearly, OP cannot know whether or not a GM has ever been stuck in Plat, and neither can you. Why? Because it’s not possible. You can guess, you can speculate, you can reason, you can’t know, and that’s what I asked.
Clearly you do not know the difference between a supposition and a fact.
I didn’t ask why you came to this conclusion. I asked how you could know that your statement was true. Do you understand the difference between believing that something is true and objectively knowing it’s true?
Do you need the syllogistic form, more statistical testing, a philosophical primer on “objectivity” and the axiomatization of logic, or Popperian sophisticated falsifiability?
So no player in Overwatch’s history has ever been placed at a disadvantage?
This is just bad logic here. It could simply be that whatever handicapping exists isn’t strong enough to keep players of a certain skill level from climbing out of some rank. Or handicapping is engineered for grind, so that climbing takes much longer than it otherwise would in a better system. And again, you make this claim that no player ever gets hard stuck in an elo in which they don’t belong. You have absolutely no proof to back that up, to say nothing of how often it does or doesn’t happen. You don’t get to make definitive statements “as fact” for which you have no proof.
All of this is irrelevant because it’s a just theory which cannot support the statement “I can very much assure you that no GM player stays stuck in platinum.” This statement is empirically impossible to validate, so there’s no equation or calculation which can confirm it. In order to know whether or not a GM has gotten stuck in plat, you need access to data you can’t possibly have.
The competitive match making system does not create a disadvantage for the player nor team. The player’s skill level is never relegated to increase another player’s likelihood of winning. This means the player is never limited in their hero pool, given less health, never has reduced healing/damage, never takes more damage, or moves slower, is never given less time to cap objectives, or forced to play outside of their skill bracket, or forced to lose SR for draws, nor are they forced to win or lose.
This is illogical. You are suggesting that a player with a certain skill level will be at their appropriate rank. If nothing stops a more skilled player from advancing, then there is not a handicap.
SBMM/EOMM systems–especially those with win streak/PBSR/MMR rewards–are objectively less grindy than the open elo system you’re proposing.
It’s been demonstrated over and over and over and over by numerous independent players on alts and has been video documented hundreds of times by GM+ players. Get real.
Can you demonstrate with video evidence of a GM player being hard stuck in Silver?
You are denying statistical evidence. You are a dishonest interlocutor. Engage with the demonstrable data or concede.
Great question!! Let’s take a walk back to your general high school science class
Deductive reasoning - starts out with a general statement, or hypothesis, and examines the possibilities to reach a specific, logical conclusion.
General Statement - no GM player stays stuck in platinum.
Possibilities -
GM player makes new account, loses a couple games and no longer receives fresh account SR variance for gains. GM player exceeds the statistical performance of his peers and is pushed out through PBSR system via greater SR gains and losses when winning/losing a game
GM player makes new account, wins all 5 placement games and is automatically placed in Diamond.
GM player starts on aged account and overtime the system reviews the statistical data provided by the player as compared to the players peers and determines that, said player, is performing consistently above the current ladder position and must be moved upward. Therefore PBSR takes effect and the player experiences greater SR gains and lower SR losses for wins/losses.
——
——
I could continue with these but there’s like dozens of unranked to GM videos on pretty much any/every hero in Overwatch so I’ll refer you to YouTube.
——
——
Inductive Reasoning - Inductive reasoning makes broad generalizations from specific observations.
Observation -
Player in competitive game is far exceeding individuals performance on (insert role here). Game is won/lost. Individual reviews player’s profile and sees the player is starting a new account. Individual notes players current SR.
sometime later
Individual reviews player’s account and observes the players new SR to be 100, 500, 1000 SR above the individuals. Individual concludes, a new account with a high performing player. This player created an alt accounts to re-climb the ladder.
Player in competitive game is exceeding individuals performance on (insert role here). Game is won/lost. Individual reviews player’s profile and sees the player is starting a new account. Individual notes players current SR.
sometime later
Individual reviews player’s account and observes the players new SR to be between 100 and 500 SR above the individuals. Individual concludes, a new account with an average performing player. This player created an alt account in attempts to gain a ladder boost but isn’t sufficiently out-performing their peers and has plateaued as their individual performance is not enough to overcome the consistently more challenging matches.
Player in competitive game is exceeding individuals performance on (insert role here). Game is won/lost. Individual reviews player’s profile and sees the player is starting a new account. Individual notes players current SR.
sometime later
Individual reviews player’s account and observes the players new SR to be between 100 and 500 SR below the individuals. Individual concludes, a new account with an average/below average performing player. This player created an alt account in attempts to gain a ladder boost but is insufficiently performing against their peers and has dropped in rank as their individual performance is not enough to the consistently more challenging matches.
——
——
I could continue with observations but it’s unnecessary. I covered a majority of cases.
Naturally there are outliers but outliers don’t define the rule and generally are corrected for over time.
Lol, I stopped reading here. Do you understand that I can tell the difference between a theory and a fact? Do you understand that I realize that you’re taking limited experience, narrow observation, incomplete/possibly outdated information and trying to weave together some narrative that fills in the gaps of what you can’t possibly know, and then trying to pawn it all off as objective truth?
Why would I for a second ignore the fact that you have no means to verify (not invent some post hoc theory but actually verify) anything you’re saying? You write walls of text, but none of that conceals or changes the fact that you’re just speculating. You’ve convinced yourself that the illusion of knowledge is knowledge. But I have no such delusions about what you’re saying. Probably best that we end it here. We fundamentally disagree, not on the substance of your argument, because we haven’t even gotten that far, but on the basic soundness of what you’re saying given the sheer level of ignorance you have about the runtime code for Overwatch.
I’m a scientist and I’m happy to talk about it. A scientific theory is a provisional body of tested hypotheses. A “fact” is an observation independent of inter-subject consensus.
You’re trying to instill doubt about the system by shrouding it in the philosophy of science. Which, it seems, your view is predicated on absolute certainty. Nothing is “absolutely certain” outside of tautologies and even those require parsimonious suppositional axioms. We work through falsifiability, not by exhausting the entire search space to find an answer.
I’m getting the impression that these subjects are not part of your educational nor extracurricular background.
Nothing about the z-test above is a post hoc rationalization of the a priori. A player’s rank is statistically not random. Deal with it.
I work with data, not anecdotal experiences. If you get some data, let me know.
On this very competitive subforum certain individuals who have identified themselves as boosters (ie. they literally do this stuff day in day out as a professional service) and have tested things agree that there is a level of handicapping at work.
H0: 20 GM players achieving GM again is random
H1: 20 GM players achieving GM again is not random
If the probability is greater than 0.05 (our alpha), then we cannot reject the null. Because the probability of the z-test above is less than 0.00001, we reject the null hypothesis.
If the player is “handicapped” or the games are otherwise “rigged” against the player, the likelihood of the GM player achieving GM again should be random. We can even use a Bonferroni adjusted alpha for multiplicative comparisons: 0.00250. It’s still significant and we can reject the null.
