These are the type of updates we really need from the Dev team. They did not hold back answering the toughest questions the community asks (for the most part). I don’t agree with all of the responses but I do appreciate it. It does provide some clarity because I really did think player based SR was still a thing (PBSR). I also really wish we would see our rank go up or down after it moves instead of 5/15, but I guess we shall see how much this helps.
What did you learn from the post?
What would you change based on the info they provided?
What difficult question do you wish they would answer in part 3?
That every conspiracy theory spread on the forums is nonsense. And those players that took the time to tell those people they were talking out of their back ends were 100% spot on all along.
Only thing I really want to see right now is data around map pools and if they actually have any impact on the game (play rates, and so on). I personally know that I play less because I get bored of getting control and push a lot.
I learned that what I called was deficient in OW1 and OW2 (mmr role average) was spot on.
I am also learning that performance is not used in MMR? very odd. But perhaps I was spot on again in “uncertainty”. If your destined to lose, and you go big, perhaps the uncertainty gets reset.
Oh and predication in lower elos is lower than higher elos… Elo hell anyone?
“For example, we have data showing that matches become less predictable the lower the average rating of a match.”
You know what HELL that could bring if someone was favored to win but has a new player, ends up losing? It artificially puts people lower than say if they were not favored to win but still end up losing anyways. Combine that with players who play lot are mathematically magnets for new players…
Ya, that’s a new change and I’m not convinced it was well advertised. Performance-based SR was a thing in OW1 for plat and below, apparently no longer and honestly, I agree with their logic. They recognized that players would get awarded for farming stats rather making legitimate contributions to the win condition.
Ummm… I think that statement is more about how inconsistent lower-tier (bronze/silver) games can be. Consistent performance will lead to consistent winning and gaining rank.
They also talk about striving to match MMR more closely, with less impetus on queue times in an attempt to reduce the number of steam-roll games. They also mentioned MMR will be matched across each team, per role ie a plat tank should be matched to a similar MMR plat tank. This will have the most impact on groups playing together since they found that people in groups can have vastly different MMRs and are trying to ensure each role, despite the group’s MMR range, is an even match.
No performance based mmr seems out of character, but I’m glad they did it overall. I don’t expect new heroes to be reliably balanced when they hit comp. Imagine if Sojourn players were getting that performance based boost ontop of the busted character boost.
Also nice to hear that they’ve realized, massive rank decay + mmr only matchmaking + visible rank only group limits = horribly composed matches. I’m sure only the smartest of people could have predicted that lol
I learned that to increase SR, you don’t have to win, you just have to get good stats.
So SR doesn’t show how good you are at OW ; it shows you how good you are with the heroes you play.
That might lead players to learn one or 2 heroes and never adapt to a given situation, it might lead players focusing their stats and not winning games.
Its a choice they’ve done but it seems kinda bad to me as OW is a coop game, I feel like jack of all trades, adaptative players should be rewarded.
But its a though choice as people want to be rewarded when they have good stats, even if they’re losing…
What would I change is to show more values like : your Career MMR number shown by a rank. And for each hero, a hero-restricted rank to show how good you’re doing with every hero. With that, I think that some players would be ok to adapt by switching heroes if they want to grow their overall rank without tainting their “main hero” rank.
Lately I felt it that way : when I play Moira, my SR increases even if my winrate with her is lower than Brig or Ana. But when I play Ana, I feel like way behind all other Anas I meet so I don’t feel good cause yeah I’ll win, but the game gonna show my Rank decreasing…
I’m not sure there’ll be a part 3 but basically, they didn’t say anything about Skill Rating. So, now we know about MMR, so I wonder how SR is calculated, especially since how can the game give you ONE Skill Rating if you play several heroes equally ?
Also, I’d ask if the firemeter is taken in consideration in any way ? It might be a good way to spot the “best” players in a game.
Tbh, I really don’t like that disconnection between MMR and SR. But I would admit that I think I see what they’re trying and I admit that its just difficult to find a system that’d make everyone happy.
Q: Does my competitive rank affect matchmaking?
A: We only create matches based on players’ MMR, not the visible competitive rank. A player’s visible rank will move towards their rating over time as they continue to play during a season. When we decay ranks at the beginning of a season, this has no effect on players’ underlying ratings.
Q: So you don’t take the number of eliminations, damage dealt, healing provided, or any other scoreboard stats to adjust my MMR after each match?
A: In Overwatch 2, your MMR adjustment after every match is not impacted by your performance in each match (regardless of your skill tier).
With both of those quotes, I think that SR is based on your stats but not on your winrate. That explains why some people stay at the same rank even with a 60% winrate or why the rank changes even after a 50% winrate.
What’s very important with those 2 quotes is that MMR and SR are 2 different things. MMR is hidden and depends on winrate only. SR is shown to the players but has no direct impact on MMR.
I guess the devs built it that way cause if you have high stats, that means you’re more likely to carry your team and win.
So basically you don’t think they are telling you the truth?
Easy way to know. 5 players create fresh accounts. They play together through qp and rank. If their rank is always the same, then they are telling the truth. We also need 5 players who do not play qp together but do play ranked always together. If their sr remains the same, they are telling the truth.
They specifically mentioned “scoreboard stats” not being used. So I am not sure if that implies there may be other stats being used to measure performance.
The biggest conspiracy theory is that ranked matchmaker limits climbing by “mirroring” players against someone of the same mmr. This actually matches what is now described in the dev posts.
Since OW2 started with the new rank system, the completely missmatched end of season rewards (plus more) provided clear evidence of mmr consistently not matching SR, to at least a full tier.
The result is a gold rank player with plat mmr has harder games than a plat ranked player with gold mmr.
This means the matchmaker is failing in its premise for competitve ranking.
I learned that since mmr doesn’t use hidden performance data (suddenly, if you believe that?) then we can drop it altogether and go with SR system.
Also I had no idea they would be providing a way to opt-out of data-collection since it’s irelevant to the MM/ranking.
I suppose we’ll get better games in S3, or at least have no excuses. Or it’s one big lolplacebo to get people to sweat more. Fake is fake and this company isn’t exactly in the business of honest.
wait how? The latest blog says mmr doesn’t use data. and it says random players.
how is perf data still a thing? i thot it was only win. they random u into lobby via mmr and u win/lose and that’s it. no history, no data-collect. your rank is your skill, and it’s transparent for you to see (or others, if you open profile).
or are we still in fake world ft. aaron and the layoffs ???
how can they confuse tf out of smart ppl?
just post the code pls omg~
