Plz explain the 50% win rate to me

There were actually a couple thread with responses from Blizzard similar to this one. We used to link them in every forced 50% thread, but people still didn’t believe it.

6 Likes

Yup. It’s questionable how you can definitively come to a conclusion that you personally performed better though.

2 Likes

As someone who has climbed out of gold as a support player and also enjoys Malf, sorry, no sympathy.

It is possible to win on an individual level, and still lose the team game. In another thread, someone had used the example of Dan Marino-- a football player with amazing stats that is regarded as being one of the best of all time, but despite this, he never won a Super Bowl-- Depending on the season, he could be considered to have low mmr under HotS’ system.

2 Likes

That’s something a community manager can do to give more weight. It was largely ignored because Blizzard also ignored it.

Because I survived my 1v3s!

It’s possible yes, but over a large number of games, it will be dwarfed in comparison to your contributions to you winning the game. If you are consistently performing at a level above your peers, you will, over time, and not through every game, able to be the person that pulls your team to victories more than someone than someone who is worse than you. It’s really as simple as that.

1 Like

I’m just playing with random stats here, but I’m pretty sure someone with the most hero damage, most kills, most take downs, and the least amount of deaths is better than someone with the opposite stats. There isn’t really a way to tell but you can kinda see it throughout the game or through replays.

I just always thought that it was determined by your personal performance with a win or loss just adjusting how much it changed, even though others have said otherwise on the forums. However, I wasn’t here on the forums long enough to see a Dev post on this topic. I just didn’t understand why my MMR should go down because someone else on my team didn’t perform well. There are too many variables to look at to see who is the cause of the loss or win. I was just asking to clarify it for myself because I really wasn’t sure how it worked.

1 Like

I think those responses also got lost since we moved to a new forum :confused:

“Possible” is an understatement. The odds are that if you can preform consistently, you will be recognized, even in games you lose-- by all but the matchmaker itself ofc, and that’s one of the issues AZ mentions in his own post.

The game does not actually rate you on individual performance; it rates you simply on whether your team won or lost. Nothing more.

2 Likes

numerical stats generally aren’t much an indicator of the context and consequence of those stats. One person could spend all game laning against a rexxar, murky or d.va and rack of lots of damage and ‘take downs’, but ultimately contribute nothing that team can use to win.

Stat bloat can come from the opportunity to rack up higher stats than the other side; a hero with lots of poke (or easy escapes) can generate a lot of damage dealt, and have their other team heal that damage up, only for it to be dealt again.

On the other hand, some heroes that have wombo/burst or team focus can only deal so much damage because once a target it dead, they can’t take any more damage.

Similarly, heroes like D.va, murky, diablo, leoric, etc can get value out of dying, so if they’re the ones generating take-downs for the other team, they may be giving up too much space to another team that then continues to win objectives, seize camps, take forts, etc etc

Yes, to some degree, stats can indicate more effort/success, but they’re certainly not a sole indicator, esp in the difference between roles, and momentum on a given side.

2 Likes

Yeah, I was one of those people on the old forums chiming in on these topics. This was a problem on the old forums too. Blizzard, could have and should have done more to dispel these myths. But I’m very appreciative of posts like these, but they should have been repeatedly shot down by Blizzard themselves, not us.

1 Like

How about grouped-MMR? Potatoes tag along with players that are otherwise equivalent skill all the time.

I feel that when I win more I am paired with worse teammates (not as you mentioned I am the worst person in my team). The match making system is solving: argmin(inter-team MMR difference + intra-team MMR difference). Maybe with the dropping number of players in this game, minimizing the intra-team MMR difference is becoming harder?

1 Like

To be precise, the famed Performance Based MMR had to be disabled after a few months because it turned out stat padders who didn’t have a real contribution were preferred by the system even though this is what the devs wanted to avoid. They want to fine tune this system and see if it works but most likely we won’t see this aspect coming up for quite long.

Performance based MMR if it would work would solve a lot of issues of the classic MMR. The problem is that defining what’s a good play is hard. From what I vaguely remember Hots had a neural network analysing performance and trying to break this down. It might still be active and would make sense it to be if the devs want performance based MMR to eventually return.

2 Likes

Your wins are what you get “recognized” with. Trying to distill other information beyond that is fluff and becomes a meta game of trying to game what you think you should do to that is perceived as beneficial for winning rather than actually winning. You’re too worried about the system misplacing you unfairly when EVERYONE has to go through the same system. You perform better, you will on average win more, even in spite of your team.

2 Likes

The other players can upvote you at the end of the game, right? The game itself can award you mvp, despite losing, right? Other players can tell you they thought you played well, despite everything, right? All of those are forms of recognition.

This is incorrect for the simple reason that a premade is statistically favored to win over solo-que.

1 Like

Why have all that convoluted and subjective stat analysis, when you can just go by games won?

That has been proven not true actually lol. Blizzard has said that the win rate difference was negligible. This is probably even more so the case especially after they gave groups an MMR boost. This kind of shows you actually don’t know what you’re talking about.

5 Likes

Just give us visible MMR already or at least visible PERSONAL MMR. Why do we have to beg for it for 5-8 years like Starcraft did when you guys KNOW it’s better for the game?!

If you just gave us visible MMR we could see EXACTLY what the matchmaker is doing.

This is why the forum needs more definitive blue posts on these issues. Too much work copying answers over and over which have already been answered or put on the HotS wiki. Premades will get matched with higher MMR solos.

1 Like