1.OP points out there is a flaw in the system due to a handicap.
2.Half the thread "IT DOESN’T EXIST TIN FOIL HEAD
3.Hold on folks here is an patent that shows they actually do have something like that also has a description of something OP was warning us of.
4.Oh Oh more bs conspiracies! More web fake news
Yep… These are the same people that will unfortunately vote one day.
Lol you know the NFL gives drafts pick based on performance right? The best team picks last, and vice versa, meaning the worst teams get to draft the best players.
The systems are so different it’s a stretch to even compare the, but the basic principle of “balance teams” is there in every successful sport or game. No one wants to watch the best players crush worse players every game.
Similarly, most players want a challenge, not to just crush players they are better than.
Of course, let’s address the underlying premise of your comment. Balancing teams in Overwatch is done by placing you with, and against, equally skilled players (At least, people the system THINKS are equally skilled, based on everyones MMRs). The fantasy that the game realizes your TOO GOOD! for your rank so it gives you worse teammates is totally false. Just delusions.
There’s no patent that has a description of what Cuthbert was warning us of. You’re making things up.
Every time I visit this forum, I brace myself to be gaslit (i.e., made to feel insane for expressing rational thought). Blizzard has told us point blank that Competitive Play is handicapped, only by using the word “balance” instead. It’s easy for them to throw feeble-minded people off their trail by mincing words.
We see this everywhere in corporate America. Corporations like Blizzard get away with unethical conduct because they can rely on the public not to read between the lines, to examine the meaning behind their glib words.
The people you’re talking to will never reverse their position because they are too deeply entrenched. But they are only a vocal minority. The polls in my threads show that the vast majority of players support the removal of Match Making Rating from competitive play. You are not alone FoxSC, you are loved <3
People want to play in fair matches, but only once they reach their correct rank, and only when all the players on both teams are of the same skill level.
That is not what is happening Overwatch.
Here’s what happens in Overwatch: fair teams are created by, at times, purposely combining good players with players who are not so good in order to offset their skill so that one team doesn’t have an advantage of the other. If you prefer I can swap the perspective, players who are weaker are teamed up with players who are stronger in order to create a balanced match so that both teams have a mix of weaker and stronger players.
This is TOXIC. It’s toxic because the stronger players are playing well and can clearly see that 1-2 of their teammates are not even close to pulling their weight. Good players can see that they are paired up with people who are making rookie mistakes consistently throughout the match.
I’ve lost matches where I felt like I was the strongest on the team, but also felt like my teammates did a pretty good job and the other team was just better. I don’t like losing, but these types of losses are tolerable. What’s not tolerable is losing because you have a couple of teammates who are clearly out of place in the match.
What’s causing this problem? The quote below from the dev team is a huge part of this issue:
The simple and primary goal of our matchmaker is creating fair matches. To do that, it evaluates potential matches by synthesizing an expected win %. The matchmaker is normally really good about creating matches with a win % that is close to 50%
There should never be any potential match win rate synthesizing. That code right there is exactly what creates ELO hell.
That’s a great summary, S23, thank you.
That’s exactly what happens in Overwatch 99.9% of the time. First and foremost, any MMR/Elo system tries to match people/teams with equal ranks. Overwatch has a massive playerbase, neither you nor Cuthbert nor anyone else has to deal with this issue in competitive.
I have actually had this happen to me in competitive CTF. A mode I really enjoyed, and got top 500 in, but was FAR less popular than normal competitive. In that mode, there’d often be one top 500 player on each team counter-balancing each other after a long queue time. It did get annoying as I had to carry to win, I agree with you there.
But what else could they do? You’re focusing on the extreme edge case where there simply aren’t 12 equally ranked players to put into a match. What should Blizzard do then? There are three options:
Is is not clear number 3 is the best option?
To summarize…
You’ve dodged this twice already, but I’m going to post it again anyways:
This is the topic where you got your Scott Mercer quote from. Where in that entire topic does it talk about “handicapping”? NOTHING in that entire topic talks about SYSTEMICALLY pairing “good” players and “bad” players together to make a balanced game.
Your only source of this phenomenon happening is Cuthbert, and as I pointed out above, he has no relevant source.
No one is actually saying that “good” and “bad” players can’t be mixed onto the same teams. What I’m actually trying to say is that there isn’t a system that INTENTIONALLY does this for balancing purposes. All the matchmaker does is find 12 people of close enough MMRs and RANDOMLY shuffles them onto two teams. It doesn’t give a damn how (in an SR2700-avg game) all the players got to that point. Why do you think there’s so many topics complaining about stomps in either direction? That’s contrary to Cuthbert’s statement of balancing where every single game has to be hotly contested.
It’s not super rare, it happens practically every match in every tier. When the matchmaker is putting a game together it looks at all the available players in the queue for that specific geographic location and for that MMR range, probably within 40MMR or so. Statistically speaking, 12 players that are within 40MMR of each other should be the same skill. I mean honestly I can’t tell a difference in matches that are 300MMR apart let alone 40MMR apart. The matchmaker should stop right there and do nothing else, but it doesn’t. Instead it synthesizes a win rate of the two teams. If the win rate for the two teams is for example 60%/40% (60% chance for one team to win and 40% for the other) then it will not launch the match and instead dips back into the huge pool of players waiting for a match and it will continue rearranging teams until the synthesized winrate is around 50% for each side.
It shouldn’t do this. It should have launched the first match of 12 players within around 40SR and done nothing else even if the winrate was 60%/40% because the team that has a 40% chance of winning has a couple of weak players on it that should lose the match easily and rank down.
Below is a direct quote from the dev team:
The simple and primary goal of our matchmaker is creating fair matches. To do that, it evaluates potential matches by synthesizing an expected win %. The matchmaker is normally really good about creating matches with a win % that is close to 50%
I agree that It should just grab 12 people of similar MMR and do nothing else, but it doesn’t. Instead it synthesizes a win rate of the two teams. If the win rate for the two teams is for example 60%/40% (60% chance for one team to win and 40% for the other) then it will not launch the match and instead dips back into the huge pool of players waiting for a match and it will continue rearranging teams until the synthesized winrate is around 50% for each side.
This example is incoherent. If 12 players within 40 MMR of each other are found, then they are effectively identical (as you state) and it literally does not matter how the teams are formed. The win % will be basically 50%.
The win-rate for a team being 60+% means that the players AREN’T all close in MMR. So again, your left with the options of…
How are you reasoning that the system finds nearly identical players, and then refuses to make a match with them? How are you reasoning that a game with nearly identical players is imbalanced?
Once again, the edge case you are concerned about is when there AREN’T 12 people close in MMR queueing. Period. There’s no matchmaker that can magically fix that and make new people queue.
Not really the main point, but something I’ve seen. You (and others) have seem to have an impression that imbalanced games help make ranks more accurate, but it’s entirely the opposite. Imbalanced games provide the least information to a matchmaker. The outcome of a balanced game provides the most information, from a mathematical perspective.
The intuition is pretty simple. If you put a good team against a bad team, you expect the good team to win. And they do, crushingly. What have you learned? Nothing. Your opinion about how good either team is should not change at all.
When you put two equal teams against each other, now you should suspect that the whoever wins is actually better. It doesn’t mean they are necessarily, if teams are close they would trade games, but it’s evidence to support the idea that the team that won was better.
And that’s just from the matchmaking perspective. As a player, let’s say you are underranked. Your MMR is too low for your skill. If you’re put into a perfectly balanced game every time, than the fact your actually better than your MMR means every game your team has an advantage.
If 50% of your games are one-sided stomped, than the difference between your MMR and your actual skill does not matter for those 50% of games. This is obvious, right? If a game is a stomp, you can’t make a difference either way.
So balanced games gives players the easiest and fastest means to climb to their real rank.
The point is that all 12 players in the game, assuming all solo-queue players, are all similar MMR, and thus, no matter how they are arranged onto two teams, it’s a 50-50 chance to win. Obviously, groups make funny things happen, but that’s not the main point of 50-50.
That being said, sometimes games end up not 50-50:
(Same topic that Cuthbert uses as “evidence” of “handicapping”)
The opening poster had an example of a game where he was on a team of all solo-queuers vs 2 three-stacks.
Or another example where you don’t see 50-50 is 2 six-stacks in a game, but one of the six-stacks is averaged at 2500 SR and the other six-stack is 3000 SR.
Those are the examples of non-50-50.
This is just my point and you have to pause and think about it to get it - There should be no reason for the matchmaker to synthesize a win rate with 12 people of similar/close MMR. But it does, as per this quote
The simple and primary goal of our matchmaker is creating fair matches. To do that, it evaluates potential matches by synthesizing an expected win %. The matchmaker is normally really good about creating matches with a win % that is close to 50%
The fact that is synthesizes a winrate reveals that it is not just matching people on MMR alone, it is calculating individual performances and taking that into account otherwise how else could it simulate a winrate?
At this point I honestly truly believe you can’t grasp the concepts that are in this thread. I think that you believe that we can’t understand you, that’s not the case. I understand the way you believe the matchmaker to work 100%, but you seem to consistently fail to understand the way we believe it work. It’s fine if you understand but don’t agree with it, but the fact that you actually are unable to comprehend what we are talking about speaks volumes.
I also believe firmly that you typically play in groups with people when you comp and that’s why your pov is skewed. There is no way you have made it to that high of a rank solo queuing, I wouldn’t believe it in a hundred years.
You seem to be working backwards here and thus, misunderstanding the 50-50 quote. What you seem to be saying is that, in the following order, the matchmaker finds 12 players of similar MMR, and then arranges them into two teams to make it a 50-50. I actually don’t agree with that. What I’m saying happens is that the matchmaker creates a 50-50 match BY finding 12 players of similar MMR. The difference here is that you’re saying that arrangement matters, while I’m saying that arrangement doesn’t
If it grabs 12 people that it can’t arrange teams to create a 50 / 50 match then it discards people and keeps looking.
If it only did that then why would there be a need to “synthesize a win rate”? There wouldn’t, they would just grab 12 players of similar MMR and no winrate would need to be synthesized, the match would begin.
This is a hilarious joke. You continue to push self-contradicting examples and ideas, and claim people pointing out the logical flaws is them not understanding it. Ridiculous. Let’s break down exactly what you are saying.
If Blizzard is already calculating win-rates, why would they also create some hand-coded hard-threshold on MMR to see if players are “close enough” to assume the game is balanced? That’s just redundant. They don’t need to define a threshold because if MMR is actually really close, the win % will be 50% anyway.
It’s not that they couldn’t do what you’re saying, it’s just a worse way to implement the same thing. There’s no need for redundant checks. Similar MMRs <=> 50% game.
Lol, you have literally no idea what MMR is, do you?
Estimating a win-rate is based entirely on the MMRs of the players. This is one of the MAIN purposes of any MMR system, to estimate the probability of who will win by comparing MMRs.
Overwatch’s MMR is basically Elo for groups with a few other minor tweaks. The fundamental concepts are the same. Seriously consider reading a little about Elo.
https://en.wikipedia.org/wiki/Elo_rating_system#Theory
Very ironic that in the same comment you said “I couldn’t grasp the concepts”, you revealed you have absolutely no understanding about what MMR even is.
Are you actually a super troll? How could you not believe someone gets to a moderately high rank solo?
99+% of my games have been solo, 100% for the last few seasons. Season 2 is the only season I actually grouped semi-regularly.
Plenty of players have STREAMED CLIMBING TO HIGHER RANKS THAN MINE SOLO. Anyone can go on youtube and watch people do it. How can you deny something you can literally go watch?
I’ll make you the same offer I made Cuthbert a long time ago: Add me on battlenet. We can join any type of game you want and I’ll explain how the system works to you. No personal info or anything needed, just chatting through Overwatch.
But I get tired of running circles on the forums. You say something objectively false, I point that out. You ignore it and move on with an example that is self-contradictory, I explain the flaw, you ignore it and move on to something else.
I promise I won’t be an a-hole, but with direct voice-to-voice I won’t let you get away with constantly moving on as your points are disproven. The discussion can progress 100x faster if we can immediately cutoff silly statements like “Blizzard estimating win-rates means they don’t just match on MMR” and address these misunderstandings.
Because groups still exist. Matchmaking would be simple as hell if Competitive was solo-queue only or clans/guilds only, but unfortunately, you can get anywhere from 2-5 stacks playing, especially when you consider the even more matchmaking-breaking 1000 SR difference between highest and lowest members in the group. Trying to make a game where you have a 5-stack that looks like (2750-2700-2600-1900-1800) is not a fun time.
But not a program assessing “criteria” to decide their mmr and then forcefully balancing based on that. So yes your example of one sport is bad.
Take OWL, balance teams using this automated system and see how competitive it is. Never mind, you’ll turn it into an argument about something other than how stupid that idea would be. If it isn’t stupid, then why don’t they do it? It is how comp works so either it’s stupid there or it is stupid for OWL. OWL players should all be close to the “same skill” right?
We proved this before. And this assumption you have is on how volatile mmr is. Basically factually wrong. Unless you actually know the code and are defending something we can’t possibly know.
Patently wrong there. Go read them. They are designed to both balance teams for extended gameplay, and to show skins to newer players as a way to encourage purchases. Both do not belong in a comp mode.
You’ve been given a quote (and patents, and a lead designer youtube clip describing this exact process) that this is what they are doing. Thus it is not redundant because this is something actively being done. For a reason even!
So, if it is not redundant then it is describing a handicapping system. Not the fringe upper or lower extreme games “how else do we balance it” like you believe.
Here’s a prime example of handicapping. The person in this video is by his own admission a Grandmaster smurfing in bronze. What happens? The match comes down to the wire with the other team almost winning despite the smurf obliterating the other team. This link should take you to the match at around 56:35 (Junkertown). The matchmaker has handicapped this player so that he easily has the worst players on his team. Watch at the beginning too, the smurfs team almost doesn’t make it past the first checkpoint despite him killing almost their entire team with widowmaker.
This is what makes the game rotten. This should have been a complete blowout with a grandmaster smurf. Now consider this, a person who should be around just 1000SR higher (say, a person who is in mid bronze but should be in mid silver) trying to carry this team so he can rank up properly, it would be impossible if the GM smurf barely could.
the fact that there is no single blue post debunking it after so many people saw it just proves this theory, besides english is not my first language but in Forums TOS there is this:
Inappropriate Language
This category includes language and/or links to websites containing such language or images which:
If a player is found to have participated in such actions, he/she will:
I think if this is not true and is truly misleading then this post should be banned for being “otherwise objectionable”, right? And it was created 6 months ago.
They don’t care blizzard would be don’t you guys have phones ???