I see very cool
Well i cant say i disagree with your post, i just dont understand why i should be forgetting stuff you wrote.
We were talking about “carrying” in one elo below yours or even at yours. I say its absolutly possible. I am proof of that statement so are my friends who i occasionaly see playing and even carry some games in their elo.
Again it really depends on definition of “carrying”.
welp, I agree on that.
I will give you two examples of me carrying. Playing with widow and having a lot kills, enemy team called me out as smurf, i was 300 sr below my usual dps elo.
Second game as roadhog, totaly destroying enemy team for two rounds, they even called me “cheater”, it was like 200 sr below my usual elo.
1 Like
Rank is no longer an indicator of how well you play on average. Actually the system can be compared very well with manipulated casino gambling. There’s only one little difference.
Your bet are the mysteriously calculated MMR values.
You don’t know how much MMR you can win or lose but you know that you win about 50% of all matches over the period of 1000 games.
(This is one of the main tasks of the Matchmaker. If everyone wins 50%, all the games are balanced to the maximum, aren’t they?)
The “calculated” MMR values pointing in one direction and your following matches will be constructed accordingly.
It is calculated in advance whether you have a theoretical chance to carry a game, based on your previous performances.
From time to time you’ll run into a “balanced” match. MMR data is used for this. These matches are more important than others.
These matches are used to determine if you still follow the rules or if you are allowed to take the express route to a higher or lower rank.
This is how bronze to GM streams work:
The streamer plays clearly and persistently above the stats of all other players on this rating for a number of matches. Variance of MMR values is very small.
If this player can now dominate in a balanced match, the matchmaker considers this a mismatch of the MMR value of the player. (Major adjustment takes place)
On the whole way from bronze to GM, there are probably 5-8 games that lead to the approximate SR value. The rest is pretty much random noise.
There is at least one problem here
A player who plays better than his teammates on average but does not dominate clearly gets less balanced matches, because the matchmaker still considers such matches to be “okay”. In this case the random noise is much bigger.
2 Likes
Sorry but no. Matchmaker doesnt work like this.
There is no such thing as matchmaker creating a games for you to consider if you are supposed to win or not.
Matchmaker is trying the best to create a games where both teams have ±50% chance to win with -+same MMR on both sides. He doesnt care if you are top500 smurf or bronze player who is stuck there 17 seasons. He just see actual mmr and will balance both sides to have same.
As i said , he is dping his best. He will not most likely have enough people to create a game and find a match for top500 in bronze to other side. He cant even now person is top500.
In my experience matchmaker cant tell difference between plat in bronze vs top500 in bronze. He just know you have higher stats snd mmr, thats all.
2 Likes
You’re missing the whole point of “if you’re good enough, you can carry yourself out of a rank” Of course 6 gold players will win against 5 gold and 1 silver player more often than not. The point about carrying your teammates is not about your teammates, but against the enemy. If you think you’re a gold player, but are in silver, proove you’re a gold player by beating silver players, not by taking their hands. I, as a Diamond player, could of course also not carry Silver players against Diamonds, but I could snack on Silvers.
In general, I always assume that everything that is possible and strategically clever is being used.
You need to consider which features are important to the Matchmaker from Blizzard’s point of view.
- An extremely accurate rating and complex search algorithms and as many hard limits as possible that would prevent the creation of an unfair match.
Does it look like that’s the way it’s working right now?
This would lead to long waiting times. And a lot of players would stop playing. In addition, complex calculations cost a lot of money.
And all this just to keep the rating as accurate as possible, which nobody knows how it is calculated anyway. It’s hard to determine if a player has just had bad luck with his matches. However, one thing is certainly not desired, and that is that a player loses 10 games in a row, because these players stop playing.
It would be much better if you had something like a simulation of the following games and “corrected” all errors. You have to beat the matchmaker to get to a different rank reliably.
Whether this is manipulation or not is a matter of definition.
1 Like
Wait the minute. So you are actualy saying they have system in place which will not let you lose certain number or games even when you dont belong into elo you are in? You do realize they would be affecting all 11 other people to do that right? Or maybe i did misundersttood you?
A streamer, I don’t want to say the name, demonstrated this very impressively for a long time.
Outside the stream a Grandmaster player played with her account, during the stream she played by herself. At best, she was a gold player playing on masters rating. And, of course, she didn’t lose every game.
Why she should lose every game? Just like i lost games in low ranks on alts someone can win games higher. I would say its even hard to drop from higher ranks if you are boosted because they is not PBSR punishing you seriously for under performing. This streamer really doesnt prove anything because as you said, we didnt see what was jappening outside of stream.
Are you talking about sniperwolf?
I don’t want to give names, because that could be unfair and the person does not play that game anymore.
If you’re looking for it, you can find the whole story for sure.
It would not affect any player, as all matches except for balanced matches are built in this way.
Balanced matches are super rare, as already stated.
Well, i dont agree 
i guess we can end it here.
1 Like
For reasons I posted above, matchmaker will generally have at least 1-2 crap players per team. Sometimes it’s me (Had a Lucio game where I was bronze heals behind our Hog and Ana. But we still won), and moreso if I dps, as I’m far too high on the role right now.
Matchmaker is forced for sure. Played with a GM smurf in gold and we either faced enemy smurf or got the most rancid thrower team mates to balance it out. My GM friend said that GM was easier to play in.
But he would get there eventually on that account.
He’d smurf as Pharah in gold and get 90+ elims per game. He’d lose some of those games, but with that kill power he’d inevitably climb.
Someone mentioned Mercy being a bad carry. Certainly harder than others, but certainly doable.
The term carrying on this forum refers to deathmatch play, and people fixate on that and don’t understand that there are other ways to carry.
The centralized problem is that to the majority:
Tanks create space -> ask anyone what this means here and they’ll most likely reply, “A shield that people can shoot from behind”
DPS -> “Supposed to kill things”, true but I don’t think that statement is fair.
Support -> “Heal”. There’s a reason why raw healing output is less favored in higher tiers.
A mindset of I’ve done my job (held up a shield until it breaks but doesn’t truly create any space at all, healed and damage boosted a DPS that can’t hit anything, cleaned up 3 kills and then died after my tanks and supports have died why are we not taking the point), is a dangerous thought process to hold as it only promotes not helping your team win unless your team vastly outskills the opponents team (someone able to solo all 6 or you feel as though you’re getting carried).
This method of carrying is only valid when you’re placed in a tier so much lower that you could probably win one handed. If you aren’t doing so then you need to switch up your mindset faster or you’re going to lose because the kill all carry mentality is not going to work in this match.
Next identify what the problem is and solve that problem first. Not everyone is going to help you in solving this but if you do so, you’ll make a new friend and at the very least have a 2v1 partner for that game. If Rein charges in a lot and you’re on DPS, target confused targets or look at where the Rein is starting his charge and most likely is going to end his charge and focus that area. As Zarya I think that’s pretty easy to bubble him, D.Va can follow with Matrix, Orisa/Hog can pull targets that have turned around etc… Use everything that is going on to your advantage, as dumb as that sounds if Rein moving in with a shield to only have it break immediately isn’t creating space then him charging in and having enemies scatter and confused is his way of creating space. As soon as you have chaos the less likely the supports will be able to focus on their healing prioritization and thus your team should have an easier time with getting kills.
Truthfully I feel that no matter what role you play, the focus should always be how can I make the other roles feel better about playing the roles that they are. When a teammates is in danger, are you helping bail them out? Are you in a good position to get a synchronized attack in?
The whole thread topic of I put in 36% effort in so I should win only applies when the game has the solo potential, Quake/CS/UT, games which give you all the tools to do everything. This game, the carry potential is no longer in the sense of DM based games, it is solely around how carry is defined in Moba games. That means, you identify someone and everyone helps them to succeed. Zarya can be a carry, Rein can be a carry, Lucio can be a carry, Genji can be a carry. The issue currently is that out of those listed people only see Genji as the carry because he’s in the damage category and is “supposed” to get kills.
1 Like
Well, the only thing matters is win percentage. And amount of matches played.
90% matches won with 5 matches played mean nothing.
10% matches won with 5 matches played mean nothing, who knows, may be it was his neighbor playing.
You win 50% -> you stuck and need to add effort.
You win 60% -> you will rank up until you reach point where your win percentile reaches 50%
To carry -> effectively means to increase team’s winning chances.
Was this battle won by one single player? No.
Did he increase chances to win? Definitely.
GM vs gold player -> GM has astonishingly higher chances to win. Gold team vs gold team with a GM smurf have terrible chances.
Gold vs silver -> gold has a bit biggest chances to win. silver team with gold player vs pure silver player has a huge advantage.
If the system was flawed as some say, GM played would not easily reach GM whenever they want. Well even diamond player will wreck all the golds -> which leads you back to single statement, to rank you have to carry.
The term “carry” is short-hand that isn’t strictly correct. Players can only hard carry if they are playing ~500 SR below their ability, but players can rank up at any SR below their true ability.
However, you have a number of misconceptions in your original post.
This doesn’t happen. The matchmaker tries to make each match fair. If one team has a silver and 5 golds, the other team will have something similar to balance the match.
The system doesn’t match by rank, it matches by MMR, which is a floating point number. The matchmaker runs the MMRs of all players in a prospective match through some mathematical modelling machine, and won’t make the match unless the predicted win percentage is between 40% and 60%. Tier icon doesn’t matter.
The key to ranking up is for that machine to be wrong. That is, your true ability must be higher than Blizzard thinks it is. You will then win more than 50% of your games. Of course, as you win more, Blizzard will update your MMR and SR higher to do a better job of predicting your performance in future matches.
I go into this in more generality in How Competitive Matchmaking and Rating Works (Season 18) → Summary → Summarize matchmaking, ranking, and progression for me.
Yes, but a silver-skilled person can easily hop onto a gold account. Like I said, that data is real. It really happened.
To this, I disagree - You’re personal stats (healing/dmg/elim/obj time etc) can steady increase but you’re win percentage can stay at 50%. This I contribute to the matchmaker finding the average MMR/SR for your team - but the min and max values, where the average is calculated, are too far apart (in efforts to decrease que time)
EXAMPLE: say (out of a score of 100) team one has players that have separate scores of 10, 15, 15, 15, 15, and 30 - where as the other team has scores of 6 people at ~16.5 - the more well balanced team is destined to win because they’re all just contributing their average. Where as the other team, the 30-scored player shouldn’t have been teamed with the 10-scored player and due to the others not being able to carry more than their average - the team loses.
Just so you know, I enjoy and thank you for your well thought-out contribution to this subject.
More often than not, but if each are used to contributing extra (which is how they supposedly ranked up) then if each contributed extra during these matches, at least one extra win would happen. Especially if they’ve been in gold for a while, they should be able to carry (or contribute extra) an even higher percentage than they did when they first ranked up. Regardless, my point of mentioning all that is that that theory is flawed. You’re rebuttal suggests that the study was flawed from the start, whereas my study suggests that it should be good enough to show at least 1 extra win (as in the 9th win - one outside of the 1/3rd guaranteed group) in order to prove that there’s some truth to the 1/3rd theory.
For the record, I’d just like to clarify for readers, that my definition of carry that I refer to though out this discussion is
Carry: To do more than the expected amount of effort. For example, in the video that BestPapa provides, I would consider this carrying - not because soldier got substantial kills -because as damage he is expected to get substantial kills - but if these are more kills than he usually gets, and his co-dps teammate isn’t get many kills, mixed with the fact that he puts down his health pack for him and others to heal themselves - yes in my eyes this is carrying.
If a Baps was, saving lives with his immortality field, healing others not in the field, while cleaning up low health enemies in between (all while his co-support teammate is still not providing gold-medal healing)- yes this is carrying in my eyes.
Carrying: Not only doing the best in your role - but providing extra help in other roles.
You did not differentiate between ability and rank in your original post. Anyways, if a silver skilled person takes over a gold account, he will lose more than he wins until that account falls down to silver. The other players in that match may get an undeserved loss (or win), but that is just one game. Over many games (about 150) all these random happenings average out, and you get a decent measure of the gold players’ skill.
Yes I agree that the study would hold stronger if there was more data.