But, the effects are minimal, wouldn’t result in the practical complaints about the matchmaker that we often see, and would be very difficult to implement unless you magically knew when someone was about to quit playing.
The fact of the matter is, it’s literally impossible to guarantee an overall win rate other than 50% (without bots!). I think this study shows more about how fickle players are than how effective it is to manufacture wins…because it’s not. Those wins come from someone’s loss and the study shows that even matches, ones that result in draws, have the lowest churn.
IOW, the highest engagement rates are where you have balanced matches because players prefer balanced matches, and until this conversation moves on past the basics to explain how on earth balanced matches (balanced via algorithmic handicapping…ooooohhhhhhhh) are theoretically bad, it’s gonna be REAL HARD to convince me that they have to do anything more than create a perfectly normal system where players of equal skill are competing against each other, in order to drive engagement.
A nice little note from the paper below...
It is also interesting to note that SkillMM does not consistently outperform RandomMM, which is aligned with our discussion in the theoretical findings in Section 4, that is, balanced matches are not always optimal for engagement.
Now, if someone was accusing them of adding NPCs to drive everyone’s WR to 75%…then we’d be on to something.
You’d be an idiot not to understand that there’s some level of interference.
I’d argue its extremely minimal though.
As much as I agree with what Taleswapper is explaining, there’s also the point that the game is based around rules and hero interactions.
It is a “Hero Shooter” after all
Heroes are stronger or weaker against other heroes and the player playing those heroes can be at a disgusting disadvantage when equally matched against someone who is playing their counter.
That’s LITERALLY how the game is designed.
So if you have a team of players who are all a decent match against another team but one team is running a more synergistic composition (that happen to be their comfort picks) vs. a team who is only running Hog/DVa, Symm/Junkrat, Mercy/Lucio because they wanna play who they wanna play, well…
Its not the match makers fault you and your teammates are unwilling to diversify your hero pool.
Precisely- and this is what Jeff was bringing up in the bit that Cuthbert quoted earlier. The matchmaker has no way of knowing what actual choices the players will make. The legitimate complaints that people make (too many matches are stomps, leavers and throwers make for bad matches, etc) have nothing to do with the matchmaker. They are not caused by hidden MMR. They are not caused by seeking to make 50/50 matches. They are largely caused by 3 factors:
Erratic or non-productive player behavior- why is my Hanzo afk in spawn typing in all chat? Why is my Rein sitting on top of the shuttle on Gibraltar point two while we are trying to defend Gibraltar point 3? Why is my Zen one-trick now playing Ana but only hard-scoping and trying to duel the enemy Widow? Why is my Hog dancing on point A long after the enemy has already capped point A? Why did my Cass uber-tilt after one lost team fight and proclaim that they were switching to Bastion for the rest of the match, because there was “no carrying this team.” etc.
Or the perennial, why is there a silver Mercy on my team when I play high plat- Oh wait, they are grouped with our Ashe and the matchmaker had to try to balance that out somehow and we ended up with this weird Frankenstein of a lobby.
Decreasing player base, making it more difficult to make decent matches.
The inherent complexity of the game, which has been a real struggle since day 1. I think people severely underestimate how difficult it is to balance a game like Overwatch and how difficult it is for the player base to properly navigate all of the complexities baked into the game at a fundamental level.
Points 2 and 3 are in the process of being addressed (to whatever degree it ends up happening) by OW2. That just leaves point 1, which will probably always be the biggest issue in the game. All focusing on the matchmaker does and all insisting that the game is rigged does is to ensure that people are looking at the wrong culprits and to ensure that more people are tilted and not taking the game seriously (which exacerbates point 1- the biggest driver of terrible matches in the game.)
I’m aware that someone who is a competent one-trick can accomplish great successes, but at some point, that person will plateau purely based on the rules of the game. That’s not to say, these plateau’s cannot be surpassed.
It simply means it’ll take a lot of focus and improvement. Being flexible and competent on at least 2 heroes, tends to allow people greater chance for success.
Of course there are “some” unusual exceptions. But outliers DO NOT define the rule.
As someone who has climbed multiple accounts Bronze/Silver to Platinum and Diamond at least a dozen times, I can definitely say that match making “can” get wonky when groups are taken into account.
Example: A GM friend of mine was playing on a Silver DPS account and grouped with a Platinum Mercy player and a tank player who had recently dropped below 3000 (but still held the Diamond emblem. His other tank and support were silver and the DPS was gold. The entire enemy team was low/mid platinum.
Why is that allowed? Well my friend is playing WELL below his elo while grouped with people of a significant SR bracket in comparison. Because of the account he was playing on being in Silver (19xx) the tank player being 29xx SR, plus the mercy being mid platinum, this is basically FORCING the match maker to create a balance at the ABSOLUTE rails of allowable match making!!
You’re LITERALLY asking/begging the software to use the entire SR gap of 19xx to 29xx SR to find a balanced match. So of course the MM is going to do its best to find a match that this team would not simply sleep through. And they didn’t. It was a rough game actually.
#DeadGame
Seriously though… Around the time I stopped competitively grinding Overwatch, the player base still existed but was definitely dropping in size. A few months later is when ALL of the posts on the forums started popping up about how terrible match quality was becoming. Its not too hard to figure out the common denominator. Many people have left the game for better options.
Or as
Has done, some of us have moved on in life.
There are fewer and fewer people, every week, for the match maker to make use of while attempting to balance the individual matches. As a result, queue times have increased and match quality has decreased.
There are many players, even a couple on the side of the Rigged MM argument, who do understand that the game is complex and not well taught in the “introductory” tutorial.
I do have to say, I agree with them. However…!! The tutorial was MORE than enough of an introduction when Overwatch launched in 2016. The game has since evolved in a direction that the devs did not expect or control!! The player base did. The devs attempted to make corrections based on what they thought would be the best outcome
The devs took minimal input from the general community and mostly took input from the GM players (as they should since the GMs are the only ones who will push the game to its limits).
For Blizzard to go back and redesign the intro tutorial around the current state of the game in 2017, 2018,2019, 2020, etc…
Why? Why would they waste their time and money doing this? The Overwatch community has created SIGNIFICANTLY greater tutorials than any developer ever could!!
The last time there was a Blizz Q&A on the subject, it was eluded to that a majority of the Dev staff were only Silver/Gold/Platinum players, with maybe 3 or 4 being Diamond and above (I recall the highest being a 4.3 player).
There’s no point for a bunch of developers who don’t understand the actual game sense to be the same people who write the tutorials. Its an unending and evolutionary task.
There’s no money in it for Blizz.
There IS money in it for the content creators!! (Personally I recommend YourOverwatch)
As much as “I” don’t agree with it, 5v5 will simplify the complexity of the game.
Less screen clutter from all the action, explosions, abilities, things flying
Less Implicit/Explicit Tank Synergy so games are less based on tank composition
No tank synergy means other roles (DPS ) are more impactful
Less cooldowns for players to keep track of.
Less players present for team fights
Solo-Carry Potential is significantly higher because Tank synergy no longer exists (this is the biggest impact and the primary reason as stated by developers for the transition to 5v5).
Papa Jefe himself stated that 6v6 Overwatch was the ideal number of teammates to FORCE team cohesion. Well it works too well. Especially when a majority of the player base prefer’s to solo-queue.
Blaming the Match Maker for community issues is just plain misguided and unintelligent. And I have to be honest, I have no idea how any rational human being comes to that conclusion.
Its not the MM’s fault your Hog player has a nasty attitude because he’s upset IRL.
Its not the MM’s fault your Ana has been playing for 5 hrs without a break and fatigue is starting to settle in so they missed their shots.
Its not the MM’s fault the 13 yro kid finished his homework and gets 1 hr to play Overwatch before bed so he wants to bust out his Genji play.
Its not the MM’s fault that you’re stressed about the pandemic while playing video games which is affecting your mental capacity to stay focused.
Which is why I’ve tried for as long as I could to prevent new players who come to the forums from falling into this trap! Misguiding the community is not how you help people become less toxic gamers.
At the end of the day, I just want people to play the game, have fun, look inward towards self-improvement and stop using the n/k/g words already guys… its 2022 For crtihs sake.
You have a tendency to reply to my posts. Don’t. Or do, if it’s beyond your control, but I won’t see it (you’re what the ignore feature is there for). For whatever reason, you have this bitter, childish need to align maliciousness with disagreement, and it’s boring and typical – you’re like the fly looking to land in someone’s soup. It’s asinine to attribute “better” and “worse” to the character of people having a clarifying discussion. It’s unfortunate that something this basic needs to be pointed out to you.
I decided to do something completely unique within the Overwatch Discord space which is to have a server that revolves around intelligent debate.
You can access Discord even from a browser without having to even install it. It will create a temporary account for you.
This is the link to the group:
The good thing about Discord is that we can even host live debates sometimes.
@ taleswapper I also invite you to join if you want to participate and represent your points of view.
I would love to see a live debate between you two!
I think text exchanges are good but having it in real time you can address each other’s points immediately.
Usually not much of interest. We Host pugs from time to time. But some mature and well-reasoned debates could really spice things up and substantially improve the quality of discussions.
Just putting it out there. Don’t feel obliged to join but only if you feel like it. .
I’ve looked around a bit. It seems that the “rigged” viewpoint has become dominant, at least here. 4 years ago, I was trying to combat it because it’s a self-fulfilling prophecy. The less consistent people are, the worse the matchmaker becomes. Thinking your games are out of your control makes you less consistent.
There was no better way to destroy a 12 player competitive game than making sure that a few people always believed that they were destined to lose. Wars have been lost with less.
I would make an even stronger statement. Thinking the matchmaker is rigged or that the games are out of your control makes you a worse player (not simply an inconsistent player).
And I say that because that is the major component of why I have to have these conversations with my clients/students. When they think that the GMAT algorithm is somehow putting a thumb on the scales, it prevents them from improving. They spend time and energy convincing themselves that their skill is not the chief component that determines their score.
And that means they do not improve because they think improving is fruitless. Their investment is in attempting to “prove” that the algorithm is rigged. Which means the single most valuable thing I can do for them is to convince them that it is not. Then, we can actually begin work on meaningfully improving.
You can accept that the game is rigged and that games are outside your control and still try to improve your gameplay. It might even save people their sanity because when they fail to climb despite having improved they won’t become so frustrated. A person who genuinely improves and still loses games can start to doubt that what they are doing is correct. So knowing the truth can set you free and actually help you improve because you won’t second guess yourself all the time. I like to work towards improvement because I think it can make me a better player in future custom game scrims regardless of how messed up the matchmaker is.
To the extent that it is something you worry about in game (the matchmaker is rigged), it actively hinders your ability to play to your potential. And we see this all the time. The moment someone experiences some degree of adversity, they start typing in chat or yelling on coms about “what is this matchmaking” or whatever.
It’s not the only thing that people preoccupy themselves with that hinders people in game, but it is a thing that hinders people in game.
Similarly, to the degree that your OW time is devoted to pondering or railing against the matchmaker, some amount of time and energy that you devote to OW is being wasted.
Having said that, if you put the bulk of your time into improving. And you don’t worry about it during matches. And the out of game time you devote to railing against the matchmaker is a sort of side hobby that does not take time and energy away from improving your OW skill set… if all of that is true… then it’s probably not a big detriment to think the matchmaker is rigged.
However, for most people worried about the matchmaker, all of that will not be true. They will fall into one of those pitfalls. So, to the extent that we are able to spread the idea that the matchmaker is rigged, we will limit the community’s ability to improve and we will increase the likelihood that players in our matches are tilted/throwing/uninvested in the game/tilting other people in the game.
Tale I’ve actually come around. I did a lot of research on GMAT teseting this morning, specifically to understand why you keep bringing up the GMAT, and honestly to fact check you. That led to finally doing about 2-3 hours of research on SBMM/MMR and ultimately changing my views completely. I wrote a post about it below. I do mention a lot of people who are not so good at explaining this stuff, and arguing for argument’s sake, but I don’t count you among them. Though I will say that everyone needs to study this stuff for themselves, and that these forums, for a lot of reasons are not the ideal learning environment, because of time constraints, the frustration factor, etc.
In general, this renews my faith that debate and argumentation is great, particularly (and really only) if you go into it willing to learn and change your views. It also underscores the real danger of cognitive bias, so this is an important moment of reflection for me.
Things got a little dicey there for a moment, but I’m glad they stayed civil for the most part.
Ha! It feels weird to be coming at this from the other side of things, but I’ve changed my views on this. They may seem intuitively correct, but I don’t see how it’s possible to maintain those beliefs having done adequate research. Which I did this morning. Spent 2-3 hours reading everything I could about trueskill (e-al), MMR, SBMM, etc. It’s completely transformed my thinking on this, and the whole “rigged” thing just doesn’t stand up to scrutiny. These views are really easy to hold, and even understandable until you understand how and why it works, and not only why it’s better, but actually necessary.
I’m eating some humble pie this morning for sure, but still, it feels better to be dealing in facts and reality, rather than in misconception (NOT delusion, as some here will derisively call it).
Because we have something of a control group in GMs, we know that they will never lose the same number of matches that we lose at our rank, and from that we know that winning and percentages are all almost entirely skill based, and so that 20% number doesn’t really apply, or certainly shouldn’t be considered the gospel in all situations. For players who aren’t skilled enough to “take over a match” then yes, you’re looking at some average number of unavoidable losses – for you – but not all players.
I agree with everything Tale said, but to clarify here: There’s a difference between going into game thinking it’s out of your control and looking back on a game and coming to a valid conclusion that your team was vastly outplayed.
I mean the first, but yeah, given that you’re 1/6 of your team, only 16.7% is what you can control. I always interpreted it that way, not that you have 100% control over 20% of your games, but that you have 20% (rounded up) control over 100% of your games.
I mean, the first interpretation is obviously incorrect, right? You’ll never have 100% control.
It’s true though, you don’t win them all. You could be playing the hottest game of your life in that 20% and then lose anyway because your dog pees on your router or lightning blows up a transformer. Chaos is a fact of life. The important part is to realize what you can control and work with it.
When people lose faith that they have something in their control, or that the control they exert is not enough, they can become less motivated. If someone thinks the game is a rigged loss from the start, do they care as much to try? I can tell you I’ve seen a lot of people who literally give up after the first minute on 2cp because they feel they don’t have enough control.
The usual excuse is “to not waste my time” - by wasting it.
Only here in the competitive forums. Unfortunately there are more monkeys in the circus than masters
The general forums aren’t as misguided and Overwatch University (Reddit) is a MUCH healthier player base.
There’s something about this forum that breeds that toxicity.
When I first started coming here, I was hard stuck platinum (looking for answers) I’d always go up 3 or 4 (sometimes 5 games) and always drop just as many.
Not that I would expect you to waste your time doing so, but some of my original posts/convos were pretty much like that.
I can’t beat the Smurfs and tanks don’t help.
Toxic supports
No matter how hard I play, I’m just stuck
Never turned to rigging for an answer though. But I think that’s because I had friends who were GM players and watching them play I knew there was a difference between them and me.
That’s why I’ve always tried to offer VOD reviews. To share what has been bestowed upon me.
Eventually I met “the right” Smurf who put a lot of time into 1 on 1 coaching me to understand how the game works (from multiple perspectives).
Thank lord for this.
My (insert whatever the trash player with a trash mentality wants to blame) sucks.
gg
Bro…. We’re 30 seconds into the game!! CALM DOWN!!
Titan…
Remember I told you I was starting Grad School this semester?
Haha… I got rolled by the GMAT when I was applying. Fortunately my score was good enough to get me into the school I wanted. But still. Hard test.
Ctrl+C/Ctrl+V of Kaawumba’s post, with the parts you quoted in bold:
What’s the deal with MMR
Scott Mercer Oct 19, 2016
Thanks for the well written post on your matchmaking experiences, I’ll try to explain some of what’s going on.
The system does in fact try to place equal sized groups on opposite teams whenever possible. Your report for game 2 where both 3 player groups were on the same team definitely seems like something that shouldn’t happen based on the rules we’ve setup, so I’ll look into it further.
We do need to do a better job of not placing players into “unwinnable” matches. When the matchmaker creates a match, it determines the % chance for each team to win based on the match it made. The VAST majority of matches are usually near to 50% (especially if you’re a player closer to median skill rating and you’re not in a group), but I’ve definitely seen logs of matches where that’s really not the case and my eyebrows raise.
The unfortunate truth is that there is not always a “perfect” match for you, especially at very high (and very low!) skill ratings where there’s fewer players of similar skill. Then you throw in the desire to match groups vs. groups, with everybody having low latency, and doing ALL of this as fast as possible even though it’s the 3AM offpeak… it can get tough. We’ve tried different tunings with regards to wait times, and the improvements were unfortunately modest as we increased the time to wait. Still, this is an area we’re always looking to improve and tune better.
Fortunately, when we do put you in a match that we know isn’t a 50/50, we adjust your SR gain or loss based on your calculated change of winning. So if you did get placed into a match with only a 20% chance to win and then you lose, you shouldn’t lose much SR.
For matchmaking groups, there’s actually two separate issues that we try to solve. The first issue is “How do we handle groups formed of players with different MMRs?”. With season 2 we prevented players of REALLY disparate Skill Rating from grouping, but there’s still some variance we need to handle. Over time we’ve tested different models to try and see what’s best and are now using what tested most accurately. (Hint: it’s not simply averaging the MMRs)
The other issue is how do we model the synergistic effects of players being together in a group. As you noted, they have access to voice chat. Now here’s where things get interesting. This “massive” advantage actually differs based upon the skill rating of the group members. Based upon the data we’ve seen groups of low to mid SR players don’t see that much improvement to their win %. Higher SR players do see more notable improvements, but it’s not as huge as you might think. Still, we do take this into account when we predict the win% for each team. Regardless of how the data looks, we do know there’s a perception of a large advantage for groups. That’s one of the reasons why we explicitly try to match similar sized groups together.
So then why do points for losses and wins seem so random? Well, the amount of MMR (and SR) you go up or down isn’t simply a matter of whether you won or lost, and what was your predicted chance of winning. There’s a couple of other things at work. One is the matchmaker’s confidence in what your MMR should be. Play a lot of games, it gets more certain. Don’t play Overwatch for a while, it gets less certain. You go on a large win or loss streak, it gets less certain. The more certain the matchmaker is about your MMR, the less your MMR will change in either direction based on a win or loss.
As a minor factor, we also do evaluate how well you played the heroes you used in a match. The comparisonis largely based on historical data of people playing a specific hero (not medals, not pure damage done), and we’ve done a lot of work to this system based on the community’s feedback. In fact, I’ve seen some people indicate that they don’t think we’re doing this anymore. We still are. While it’s a minor factor compared to wins/losses ( The best way to increase your SR is still to play together and win as a team! ), doing so does help us determine your skill more accurately and faster.
So take all that into account, the SR gain/loss after any single match can be a bit more “noisy” that it seems it should, but we’re asking it to look at a lot of different factors to do the best job it can creating fair matches for you.
As an aside to all of this…
“Fair” matches doesn’t always mean that every Ilios match goes 3-2 and 100-99 on the final point, or each team gets the payload to the end in overtime on Dorado, etc. Sometimes when two evenly matches teams play, the result can be one-sided. It just means that at that single moment in time the enemy team played better. It’s not always the matchmaker’s, your’s, or your team’s(!!!) fault that you got stomped.
) are more impactful
