1- No, I see no problem with that. If your win/loss is abbout 48%/52% then you are almost exactly where you should be and your climbing should be extremely incremental. If you managed to climb up 500 only getting ahead one additional win per loss every like 50 games, then yeah, that would, and should, take a long time.
2- Ok so I reject your key premise “good and bad players who are not equal in skill?” If you have say a Gold level game with players ranging from say 2000 to 2150 or something like that, then everyone in that game has almost exactly the same skill, with the differences between them being extremely marginal, unless an account is like brand new and hasn’t had many games to get sorted yet. So, grouping together a bunch of players that are all of very similar skill, with maybe Charles over there being SLIGHTLY less good than Danny at hitting his free throws, but only barely, I call that a good game, I call that balanced. I don’t call that handicapped because Charles only sinks 80% of his free throws and Danny sinks 81.75%, gimme a break. A system without MMR would be MORE prone to this type of uneven matchmaking, especially early in the season.
3- Because it stabilizes the system in a million ways small and large, creating a comp environment where games tend to be more even. A system based purely on win/loss and literally nothing else, especially in a 12 person pick up game…cmon man, use your head. I’m putting this in the big bucket of “Things people THINK they want but would despise if they actually got it.”
1- You quoted my statement of 52%, are we talking 52% or 57%? Whatever, it doesn’t matter. If your win/loss rate is barely over 50%, then yes you’ll probably climb very slowly, the more over 50% it is, the faster you’ll climb. Makes sense to me, what’s the problem? For example this season I’ve gone up about 500 SR so far from where I came out of placement and I have a 62% win rate. I feel like I’ve been climbing at a perfectly acceptable pace. If I was winning only barely over 50, I’d probably be barely climbing, and I would expect that to be the case. What’s the problem here?
2- I did give you a word, “balanced”, and a phrase, “good game”. That is what I wuld call matching together a group of people with very similar skill levels to play against each other. Nobody calls that handicap. The point of my whole diatribe about exceedingly marginal difference was that you struck me as a pedant, so I was preempting your pedantic reply, but instead I just got a different one.
3- No, I didn’t, I’ve been talking about both, and how the interplay between them works, for this entire thread (well to the extent of my participation anyway). That’s what everyone has been taking about. You can’t have missed that.
I took that as meaning you adjusted the 57% to an effective rate of 52% as “not all wins and losses are equal in SR.” Which especially works if they are playing outside of metrics as their PBSR adjustments will mean they lose more SR on a loss than they make on a win. If your example was meant to be two separate examples then no point was actually being made. Winning more than you lose is how you climb, or did you think no one knows this and spent two paragraphs describing it (badly)?
Ok, great definition for something else. What do you describe balancing the teams if they are not of equal skill levels. You know, when they are handicapped to get that balanced game. All handicapped games are balanced, but not all balanced games are handicapped. To be able to use balanced as you describe then you believe all balanced games are handicapped? Last attempt at this, I don’t think you’ll be able to even understand the question at this point.
You want metrics to be valued in the matchmaker:
That is nothing on how they “interplay” and is conflating the two. Do you believe:
a) A match should be made by MMR, and then use the metrics (which includes your MMR vs the average of the match MMR etc) to adjust afterwards how much MMR you move by.
b) A matchmaker should know exactly your SR adjustment value will be before a game even begins? (Regardless of whether you AFK or not)
OK, I’ve been on the internet long enough to recognize an intractable argument with a pedantic tar baby when I see one.
Long story short: Having both MMR and SR affect how people are matched up, as long as the actual game talent of people in the matches are within a fairly narrow range for most games at most levels most of the time, is good and balanced and fair and I see no problem with it, in fact I see it as actively preferable to a system where teams were generated based solely on a win/loss based SR, which seems to be what a lot of people on here want, which I think is a stupid idea and they would hate it if they got it.
insult people when wrong rather than admitting it, obviously.
Ok, again you are showing a basic misunderstanding. SR is not used.
Ignoring the SR/MMR issue, then can you explain how that is different to Diamond+ where PBSR is not a thing? You seem to not even know of PBSR and are just flailing at this point.
But it is arbitrarily set to have as close to 50% chance to win a balanced match every game. So the system is balancing to you, and not allowing you to progress naturally.
That is the conundrum. IF they want it to be balanced this way, SR should not be determined by wins/losses, but by stats alone. In my opinion.
This is where you are not seeing that in that 2000 to 2150 game the MMR is 1700 to 2350 (Equivalent) and hidden, so that the 2000 SR player that is really 2350 is being set with the 1700 mmr player to make the other 2000 SR/MMr equivalent players an “Even Match”. Because the inference of a balanced game within a set bubble of SR differences. This can occur.
You cant compare a 2 stack with solo because you know, there is 2 equally skilled guys but other has a outside constant source. Like wut? How good is the mercy? Is the mercy gold where you start your journey but the soldiers are dia for example? Or is the mercy as good as the soldiers? Not taking these things is consideration tells me you dont know how the MMR works.
Its not this simple.
Ineed.
No.
Kinda yes, but no. For example 2 equal soldiers goes to play the ladder on new accounts. Other plays solo other ques up with 6. The one with solo will rise to hes own level meanwhile the one who stacks is limited or lifted by the stack. Lets say both soldiers are 4,2k soldier OTPs. They que and the solo will ofc reach 4.2k. But the one stacking cant reach 4,2k if hes stack is for example made of plats. The highest he could get is some where in diamond, because there is no way he could carry hes team so hard unless hes plat friends would improve enough for them to climb 4,2k meaning hes now stacking with a GM team, not a GM + 5 plats. And this is why its so stupid when you see people claiming its impossible to climb and its so easy to climb in a stack. No its not. Only reason why climbing in a stack is easier is when you have much higher ranked smurf in. Those plats would be elevated a bit up because of the GM carrying.
Nope. You can still climb. You dont have to improve any metrics. In fact if you manage to keep them the same (assuming those are good) you will keep climbing. The problem is that when a gold player climbs it becomes harder to keep up the stats. Its not because its a rigged system like everyone likes to think, but because the player is facing harder opponents. This is also what some people are hugely mistaken. They only look stats and forget everything else. Ive seen like silver guys claiming they should be masters because they pull same numbers. Meanwhile anyone with half a brain understands that a silver cannot pull the same numbers in masters and a master player would destroy those stats in silver.
Would love to see this. Sure I think they have, but Im also sure you completely misunderstood.
??? You win game you gain SR, you lose and you lose SR. Its that simple. BPSR is something that helps the unlucky players and the smurf problem, but it doesent actually have anything to do with your rank, its not keeping you down. Only thing it does it actually helps you climb if you deserve it, even faster than a pure win vs loss based system would.
I think once the next patch goes live for a couple months and people continue on in 2-2-2, we will get a better feel. I will say right now, Comp beats QP hands down because the MMR/SR spread is enforced in Comp, its clearly been broadened to absurd levels in QP. Literally seen games with Master against silvers… that should not be.
Well, but you dont know that. Maybe game doesnt work like that. You might see higher dmg in profile but your higher dmg is not result of your skill, its actualy result of mercy skill by dmg boosting you so she can be actualy rewarded with better mmr.
I am pretty sure that comp system is very complex, it would not be so simple as dmg, healings, elims. There could be another very differents stats we dont actualy see. Like for example how much did you feed, accuracy can be used, how much dmg you actualy stoped by using your abilities can be used and many others stats.
Lets say you play ana and you will 4x slept ulting roadhog. You could actualy stopped like idk 3000 dmg. You will not see that anywhere in stats but game probably knows it and can reward you with higher mmr.
This is the most pretentious thing I’ve heard today.
How much of a non casual gamer are you? Are you t500? Do you play professionally? Do you get paid to play, coach, analyze, or manage people playing the game?
If you said none to any of the above then you’re as casual gamer as any of us.
“But it is arbitrarily set to have as close to 50% chance to win a balanced match every game.”
What do you mean arbitrarily. Its not arbitrary at all. Its based on how I play… That’s like…the opposite of arbitrary. My winrate with some heroes that I’m good with is 62+%, I climb with them, with others its lower, I don’t. Clearly how good I am with the hero, how well I can play it, how much I can add value, affects my ability to climb. That’s not arbitrary.
“This is where you are not seeing that in that 2000 to 2150 game the MMR is 1700 to 2350 (Equivalent) and hidden,”
Citation needed. You are completely pulling that out of your butt and you know it. You, nor I, nor anyone else on this thread knows what the average variation of MMR in games is. Not to mention you don’t know the range, or intended volitility of MMR as compared to SR. Maybe a person with an SR of say 2000 can, based on having off days or good days or lucky or unlucky matches, might have a very broad MMR variability throughout the season, or a very narrow MMR variability. You have no idea. You’re just making stuff up. Unless you have access to a resource I’ve never seen.
I will tell you what I’ve told other people. I feel, for myself, that I climb or drop SR at a perfectly reasonable rate based on how well I play, and that most of my matches are pretty well balanced with other players of a very similar skill to my own.
How would taking away MMR improve my Overwatch experience. Sell me on it, go ahead, we take it away starting next season, what will I see different that I will be happy about?
You know arbitrary also means strict ruling right?
Its math, you and I and others in threads of “Loss streaks” know that SR can drop incredibly in one session. If MMR can move slower than your SR, it is not that much to consider a win streak that will place you +200 or so or a lost streak that will drag you - 200 or so. All we know is that the system will not put you into games that SR values are too far apart.
And of course we have no hard data, but that does not preclude us from deductive reasoning.
Taking away MMR will only improve Competative. If you are in games that are relative to your SR and you win, lose, draw, you know you are facing opponents in the same range. MMR makes each match as close to 50% even if you are not in the SR range you should be at. Which can make you slip further away. Up or down.
Yes to all of this! And the bottom line is your first sentence.
It is hard to say what you would be personally be happy about, because it depends on what kind of player you are. If you are a lone-wolf type, a casual player, or a new player then the removal of Match Making Rating (handicapping) would cause you to lose more matches. If you are a team-oriented, competitive, experienced player then the removal of MMR (handicapping) would cause you to win more matches.
This change would make Competitive Overwatch less ‘accessible’ (i.e., artificially easy) for new players and possibly less marketable. But it would mean fair and objective ranked competition, for a change.
I don’t see the point of mmr really. It makes it harder to go up as well as down. It doesn’t really help with smurfs either and the pbsr difference is not as high as it used to be. I think it’s just a waste of developers time when they could do much more important changes to the game. If you are gonna keep mmr then get rid of sr.
And no, I don’t think mmr is keeping me down or anything, I just don’t see the point.
Very relevant. Competitive needs bigger swings. The mmr algorithm needs a validity check. Place a diamond player in bronze and vice versa to test how accurate it is. Place a diamond mercy with fresh players and see how good that player really is.
They also still “balance” heroes around mmr which is ridiculous. If you end up getting better the system flags this up and eventually your character gets nerfed and you are back at square one.
They should multiply or divide points based on k/d ratio or something analogous and account for pool of available players per user.