The fact that the matchmaking tries to make even matches is not a great conspiracy. It is it’s purpose. It’s the entire point of it.
You are not being held back by the matchmaker.
The matchmaker’s purpose is to find 11 other players with as close a match to you as possible, skill wise. If that is not possible it will try to find two teams with as close average skill as possible with a + / - 500 points range per player to do it.
That’s it. Again, there is no great conspiracy to make you buy new accounts or whatever. It’s purpose is to create as even matches as possible.
These are two different functions, and you are completely wrong about the first. Placing Overwatch players in matches with peers of roughly equal skill is what the Skill Rating system should be doing, not the Match Making Rating system. But at every tier of the ranking system there are players with widely different levels of skill, and that is because of Match Making Rating. My whole argument is that because of MMR, SR is not doing its job.
Btw, you should say “its” not “it’s” as a possessive determiner. Just remember, “it’s” stands for ‘it is.’
I don’t play Overwatch any more but when I did, I was not very casual.
I love all the people not playing anymore but spend all this time complaining…
As for the rest: That is what the Match Maker IS. It finds people of the same skill (SR and MMR), and tries to make an average match over two teams.
The Skill Rating is a number of VALUES. The Matchmaker USES those values.
People dont understand MMR at all. In a easy to explain way this is how MMR works.
Consider 2 equally skilled players using soldier 76.
1 of them plays solo.
1 of them plays as a duo with a Mercy who is always pocket damage boosting.
Although they are both similarly skilled, The damage boosted soldier will have a higher “damage per 10 minutes” therefore will have a higher MMR and will be considered a better player and will rank higher.
Thats basically how MMR works but taking a lot of other metrics into consideration.
*This is why MMR is flawed
*this is why having the same ladder for single vs queued up groups is flawed
*This is also why Blizzard removed the group indicators (so nobody complains they are getting put against groups and getting rolled)
Once you stop improving “your metrics”, your mmr stops moving and at that point your sr will hover around your MMR. (this has been confirmed multiple times by blizzard). This does not happen because you play bad or good or stop improving as a player in the game, it just means you have not improved your metrics.
Hidden? Duplicitous? Doesn’t literally everyone know about MMR? Hasn’t MMR been officially explained. What do you mean by hidden and duplicitous?
Handicapping? What do you mean by that. MMR doesn’t affect the way your character preforms in the game. Typically a handicap in video games refers to having some numeric advantage given, more lives, better damage, better item drop rates, whatever. I don’t think I’ve ever heard “Matched with other players of around the same skill level” referred to as a handicap.
I guess let me ask it this way, in what way would my gaming experience be better if there were no MMR. What would I see different that you think would be good? Sell me on it.
And you know all of this to be 100% for sure true? Can you cite a source on this? A lot of people just pull this stuff out of their fanny. Where did you get the information? I’d be interested to read it.
So lets say the system is solely SR based, where you are seeded at mid-pack SR (say 2250), are matched against similar SR players, and have your SR rise or drop depending on your wins/losses until you reach ~50%. Then you would have confidence that your displayed SR is accurate.
However this game uses hidden MMR to ‘balance’ matches within the SR bracket. In otherwords, they recognize that two players with the same SR are not equally skilled. For example, one might be a new seed on the way up posting metrics above average for the SR bracket, so he has an MMR that is higher than his SR. The matchmaker then looks for another similar ‘underranked’ player to put on the opposing team so that the game is still 50/50 - which is ‘handicapping.’ Based based on an even SR match without MMR balancing, the underranked player’s team should probably have a stomp so that he will move his way up the ladder until he plateaus off.
…How is that a bad thing? I am struggling to understand why that is bad.
If you are placed at 2500 SR, but you aren’t really that good, you just maybe got lucky in your placement matches, or it’s a new account or something, then you will lose more than you will win and you will go down. If you are better than that you will win more than you will lose and you will go up. If you are just about right then you will have a win/loss/draw of like 48%/48%/2%. That’s all working as intended right?
If they system sees you’re a 2500 SR DPS player but who doesn’t have very good damage stats, the assumption being you’re lucky or you’re being carried, but it pairs up another similar player on the other team so that the other 10 people in the game still have a nice game, why is that a problem, both of those player will still have their SR gradually decay over time until they are in the “right” spot, all the MMR is doing is making that transition less jarring for all of the other people in the game.
If you took away MMR then the first like 3 weeks of every season would be just an endless series of rolling teams one way or the other. In what conceivable way would that make the game better? How would my experience as an overwatch player being improved if MMR was removed? It’s not like MMR keeps a silver player in gold forever, they will still decay down into silver, it just takes a bit longer and is less jarring.
I, for one, and perfectly content to have my competitive experience stabilized even if it means a Silver skill player who somehow placed in in Gold gets to play and extra dozen games in gold as he gets slowly bumped down. Why is that a bad thing?
For my part, by the end of most seasons I end up at a play where my win/loss is about 50/50 and most of my games are at least somewhat close with only a few steam rolls here and there. So it seems like it’s working. What’s the problem?
I’m not going to say wether it’s good or bad. I’ve played enough that I’m confident my SR / MMR are in sync so it doesn’t impact me personally. Just what the handicapping claim means.
But the objective of playing competitive is to improve your skill - which mean’s its all about your ‘SR.’ The MMR handicapping ‘supposedly’ delays the progress of your SR matching up to your actual skill, defeating the purpose of playing the mode.
It also begs the question that if MMR is the better estimate of the skill, why not just drop SR and display your MMR instead? Preferably with details on how or why your MMR went up or down after each game.
I have never ever heard the term “handicap” ever used to describe a pick-up game where you balance the teams to be about equal. In what world is that a handicap? But whatever.
Now you ask “Why not drop SR and just have MMR” well I guess you could…but why not, and bear with my while I put forth this wild theory…why not have a system that takes into account BOTH your win/loss rate in games AND your metrics in your role and weights them TOGETHER while matchmaking…gasp.
I still struggle to see why anyone has a problem with that seems to obviously perfectly fine and appropriate.
Are people actually upset that MMR exists, or just that they can’t see it? But of course if you could see it you’d have people gaming the hell out of it in an abusive way, and we all know that’s true.
Well, the claim is the game isn’t supposed to be ‘balanced’ to be equal. If you are better than your SR then the game shouldn’t be trying to balance out your impact. If you are worse than your SR then it shouldn’t be trying to hold you up. The point of the match is to test whether or not the SR assigned to you is accurate not to try and make a fair game.
It makes more sense if you think about the claim that if you have a high MMR, the matchmaker may pair you with a low MMR if it can’t find a similar high MMR in your SR bracket for the other team - in other words it expects you to carry the potato to make the game even. Now, given that MMR isn’t shown there isn’t really anyway to show that this actually happens but it has been hypothesized.
I think people would complain about MMR less if they could see it, so if it goes up even if your sr isn’t going up quickly then you still get some feeling of progress.
The part about SR floating around MMR has been stated multiple times by blizzard themselves.
The other part is my own personal theory from playing every season mostly in GM or top 500 games and decided to test it.
One test I did myself was, first of all trying to determine which of the “per 10 minute” stats was easiest to manipulate. For me, the obvious one is the “objective time”, basically stand on objective as much as you can.
Next, determining which heroes typically hold the least objective time. Most tanks or healers and some dps like bastion probably have the most as they likely sit on payload a lot.
But there are heroes that rarely push payload, like tracer, sombra, phara, maybe widow etc. These are the heroes where it should be easier to have a higher “objective time per 10 min” compared to other players using those heroes since hardly nobody focuses on it.
So using Sombra I focused hard on increasing objective time on a low masters smurf account.
My results was, after a few hours of raising objective time, I went on a win streak where I was just winning 9 out of 10 games until I eventually hit top 500 and eventually #1 Sombra for a few weeks on overbuff before I just stopped.
This was a few seasons ago and I did it for maybe 3 seasons each time top 200, ive also posted this tip multiple times in the past after I got bored of exploiting it.
I mean, maybe I’m just a weirdo, but I’d rather have more even and balanced games even if it means someone coasts for an extra dozen or so games a bit higher than they really fit. I’d much rather that than having most games just be a steam roll one way or the other until maybe the last week of the season when things finally start to even out.
Maybe I’m the weird one here. But having a good game for the other 11 people is a bit higher of a priority to me than a draconian adherence to nobody ever being matched in games marginally higher than where they might belong. That extremely minor sin seems like a platry price to pay for a substantial benefit.
Again, I struggle to see why this is something to complain about?
Except that people on this forum will complain about literally anything.
That’s very interesting, but I find it highly unlikely your findings are applicable to the vast majority of players. What I mean is that
1- Your meteoric rise on your admitted smurf account doesn’t mean much, as you were smurfing, so what portion of your rampaging win streak was due to your MMR manipulation just you actually being more skilled is probably impossible to sus out.
and
2- The variability of these stats between different players at the levels you were playing at is probably VASTLY different than the variability between those stats for 90%+ of overwath players, so your effects might have been greatly amplified, or hell maybe even greatly muted, we don’t know, but assuredly highly skewed.
But still, a very interesting effort.
I think it best that MMR stays as opaque as possible.
maybe, or maybe not. All we know is MMR exists, it has a play in determining what the probability is that you win or lose, not taking into account wildcards like throwers or people disconnecting, and that SR floats around MMR. (this has been confirmed by blizzard)
SR will always go up or down what your current MMR is. That 100% means that if you do not increase your mmr, you will win a few games but then be brought back down below your SR to where “you belong”
He’s right tho. If you were actually a competitive player who cared about the game you would be trying to improve instead of trying to bend the matchmaker in your favor.
See, this is where you lose me. I think you’re pulling that out of your butt.
I propose that if we were able to create a hypothetical clean slate account with say a Solider 76 player (just cause he’s a simple example hero) and he came out okf placement at 2500. If we were able to carefully control his play such that his “metrics” for like time on payload and kills and damage and all that was almost exactly the average for a 2500 level player, BUT he consistently, week in week out, had say a 57% win rate due to things that don’t fit cleanly into a metric, gamesense, positioning, recognizing key picks and play making opportunities, that player’s SR would go up and keep steadily going up.
Now of course you would EVENTUALLY reach a point where 2500 SR average numbers just couldn’t cut it, and then they’d stall, but even if their metrics stayed almost exactly average to 2500 SR, they could get up to 3000 so long as they kept consistently winning at a greater than 52ish% rate. (52%ish cause not all wins and losses are equal in SR, and there are draws to consider)
But generally speaking, for the VAST majority of players, as your skill as a player increases, those metrics increase as well, so it’s not a problem, the two stay in pretty good sync. So I fail to see any problem. Like, I readily see the benefits, and have to really squint to see the downsides.
it has been shown time and time again that mmr is more likely to boost you up than keep you down. this combined with pbsr means its harder to stay in plat than it is to climb if you belong in a higher rating.
if the game thinks you belong at a lower rating then all it means is youll get the same treatment as anyone above diamond and rarely get above 24sr for a win. now the losses may seem unfair in comparison but if you truly deserve to climb you will.
Yes, you are right. But that would be 500 games to do so (52% winrate for adjusted SR gain/loss, so can be treated as 25SR movement on actual winrate, will go up by 100SR for every 100 games played as 4 more wins than loses, so 500 games to go up 500SR). And this is on a person that would obviously deserve to climb. Just checking, but do you see any problems with this?
Is there a better word that you would use to describe balancing out the teams to be about equal, when you have good and bad players who are not equal in skill?
Why create such a system at all for matchmaking? What you are describing is accomplished by SR adjustments after each match. Do not fall down the trap of thinking a matchmaking system is the same as a ranking system.