One of the times when Blizzard was transparent was when they assured people that there was no truth (not even a kernel) to the idea that premades win more. The specifically said that the measures they did take were to attempt to resolve a perception issue, nothing more.
I’ll leave you to guess whether people actually believed them and whether revealing the information did any good or actually changed anything.
It’s because they were looking at the wrong data. What they should have been looking at was how often people in parties play together. They could hold MMR values for parties that play together often, it would only require like a few hundred bytes per player to store.
That way, people who play dozens of games together develop another MMR.
Even something as simple as a 50-75% reduction in MMR gain and loss while in a party would go a long way.
The only advantage groups have are coordinated communication, and it’s more likely a 5 man would be on comms.
None of that is being transparent. transparency would be releasing the API so that we the players can look at the data and come to conclusions. What Blizz did was TELL US something that they CLAIMED they measured and nothing more.
This is the equivalent of people accusing a police department of corruption, so the department investigates itself, comes to the conclusion it did nothing wrong and assures the public that their worry was misplaced. it is meaningless lip service that doesn’t show the public anything and doesn’t let the public come to any conclusions.
What good would that do? Then people would just claim Blizzard had set up a fake database for the API to access because the real database would show that Blizzard has been distorting matchmaking in order to corner the Canned Ravioli Market. Or something.
People determined to believe that someone is lying aren’t going to stop just because they get more data that shows they’re wrong.
Seriously, do you honestly think Blizzard doesn’t keep track of their own data? That they don’t have people looking at it? That they made sure to hire people who don’t know how to look at data?
It’s kind of funny to hear somebody advocating a system whereby people are artificially restrained from rising in rank. 
Here’s a crazy thought: what if they looked at the correct data, and it showed exactly what they said?
Again, there’s really no way to satisfy somebody who’s determined to believe that someone else is lying. Whatever evidence is produced will just be decried as another falsehood.
I’m talking about being constrained to being boosted by your peers and not tanking your MMR for playing with lesser skilled friends.
I say give us the data and let us decide for ourselves.
Not everyone that is suspicious are hardline conspiracy theorists…
You cant convince a flat earther the world is round, you still continue publishing the data to prevent the wide spread infection of their ignorance.
Of course they do, has anyone said they dont?
Was my “concerning alarm” for a bit when heard it from others but then watched the explanation in details where it was clearly stated that SOLO PLAYERS will be HIGHER MMR than the party MMR, meaning = if people team up they’ll get matched vs better players, and vice versa = when playing vs a party you’ll have better skilled allies, i.e.:
In order to ensure fair matches are created, one of the variables used to determine whether a match has an acceptable win probability is the party size. In the event the party sizes across both teams aren’t equal ([5]vs[1-1-1-1-1] for example), other variables will be weighted differently to make sure the match created is balanced. One example of this is that a team of [5] may have a lower MMR vs a team of [1-1-1-1-1] to help balance out the advantage.
It’s bizarre at first glance, but makes sense kind of on another
Don’t get me wrong, I’m Insert HUGE quotation marks here "glad " they’re “experimenting things with the Matchmaker”… Actually WRONG
Mainly because THESE kinds of “tweaks and experiments” should’ve been done FREAKIN’ LOOOONG TIME AGO, like, even before BETA if you ask me, but hey, better late then never KAPPA
SRS, it still pisses me off to high heavens they decided to listen/ask to this & that & everyone BUT their own fans, almost like they expected rage/negative-bias, well = maybe for one second they should’ve thought that THAT WAS THE TRUTH ffs, and usually that, THAT’s what happens when designers/creators/producers/you-name-it “split away” from their long-term consumers and turn onto “marketing experts” whose goals are to “chase new audience” instead of quality of service
Overall IDK, again, GLAD they experiment, but it may be (probably IS) tooo long tooo late now
Umm, all of that stuff they mentioned they’ve had implemented for years already.
They only mentioned it again to try and assuage fears about the system.
Can’t help but doubt it tbh
It’s been in place for a long time. They also raise or lower your MMR in QM based on your hero level.
Don’t think Hero level is the right measurement either tbh… It’s complicated but there are simply matchups that certain heroes can’t withold unless much superior
Yes I know/heard it depends on the hero but not on hero lvl specifically, if anything I think they once mentioned it depended on hero WL%, not sure
Either way, the bottomline stands = they should’ve experimented MORE with the matchmaker (and even ranked) during the preparement phases and instead of read reddit “positivity bias” should’ve looked for ACTUAL CRITICS even if it may have been a “needle in the haystag” but doubt it back at the time
There weren’t many players (10k Alpha then 30k Beta invited originally ?) and usually the feedback atmosphere was more sincere (kind of like reminescent to the situation now)
This is baseless conjecture and not worth taking seriously. also it would be pretty easy to tell if the API was giving us incorrect information. Unlike now where we have NOTHING but HOTSLOG.
Official data is much more reliable than developers who have shown to outright lie and refuse to be transparent. This is a pessimist argument of “NOTHING WILL BE GOOD ENOUGH!”. it holds no weight because it adds nothing to the conversation. “what if some people still don’t accept it!” isn’t a reason to deny us access to official data.
they obviously do keep track of their data since they’ve confirmed there is an API that lets them access it.
I’ve called out former lead hero designer Matt in a previous thread years ago for suggesting Murky was a “strong hero because of his winrate!” due to Matt purposefully ignoring his pickrate. He never responded to my comment nor posted the actual data either. With access to the API, I’d never need to ask a developer something like that when I could just look at the data myself. There is a lot of evidence of the developers not knowing how to read data as they have nerfed heroes declared “too strong” despite very low pickrates in skilled play that we could see on HOTSLOG.
kinda shoots your “it wont be good enough” when your resolution is claiming the numbers they gave are either misread, or they must be lying.
So did you think that all through before posting to notice the bit you’re trying to disagree with is inherently contradictes by how you act to this information, or was that supposed to be a compelling case for being more subtle about errant claim?
Even if this were true (I didn’t see it in the patch notes) it’s a massive red flag for someone to come out and say “I’m hiding something to change your perception”. You might as well say “I am untrustworthy”.
If you spend enough time lurking around you’ll find in plenty of communities there are people who will make a ridiculous claim. People will then call them on their BS only for the claimant to turn around and say “Well I had evidence, but since you won’t trust me I’m not gonna show you. Nyeh”. This reminds me of that, I kinda hope it turns out that way in an official capacity. It would be really funny to see a dev do that.
Would your base assumption be that communication doesn’t impact the result of a team based game?
Literally all they have to do is give us our actual MMR, and put it into the .stormreplay file and INSTANTLY HOTSLogs is now 100% accurate.
Exactly. I would probably trust an official API. What I don’t immediately trust is Blizzard’s INTERPRETATION of the data set. If I had the data set, I could draw my own conclusions.
For example, I could use statistical analysis tools to determine what the actual mean and standard deviations are, and then from there do some hypothesis testing to really determine whether or not the data is useful.
From what I understand from former employees, there aren’t very many people on the dev team that have an actual heavy understanding in mathematics.
Most game dev only requires basic trigonometry (Which is SUPER useful for game dev stuff like drawing, pathfinding, you name it) and I wouldn’t expect the team to know how to interpret data in an unbiased manner.
At least if it were a player doing the data analysis, that analysis could be peer-reviewed for correctness. We can’t double check Blizzard’s work, we have to take them at their word.
Every time I have ever made an analysis tool for this game, I have always released the source code hoping someone could check it for correctness. Give us access to the raw data through an API, and let us do our own analysis.
It’s good to know that you’re the sole arbiter of what is and isn’t ridiculous. I’m sure you gained such massive ability from completely reliable sources.
Woohoo! There’s reliable information!
Or something, anyway . . .
Would you have felt better if I said “Regardless of if you think this is ridiculous”. Because that wasn’t the important part of what I was saying. The important part was people refusing to give information that would help their credibility on the grounds that others don’t trust them.
I’m starting to understand the basic problem here. Let me take you through it step by step:
They aren’t doing a thing to try and change perception. They did that already when they tried to tell people that premades don’t actually win more often, and it didn’t work (see above posts on how Blizzard is lying and/or can’t understand their own data). All they’re trying to change is in-game behavior by removing a piece of information that accomplishes nothing other than causing players to tilt. The pre-game bias is “We’re going to lose because the other side has people in a group.” That mindset is unhealthy for the game. That’s the bias they’re trying to eliminate, because it directly impacts how people play the match.
The flaw in those points are:
If what you’re saying is accurate, is Blizzard in turn saying that the community is stupid and would misinterpret the data? If they say the data reflects the reality that premades don’t win more surely it would help them.
I like how you got offended by me saying “ridiculous” and you have repeatedly implied people who don’t trust the devs are crazies and conspiracy theorists. They pretty much said themselves they are not trustworthy by making this change especially when you add their stated reason.