Stop calling it forced 50%

I use forced 50% for boost my ranked game, play QM keep losing 3-4 streak, then ranked and you will easier win

1 Like

I notice that too at one stage but the opposite of yours. I remember going on a mass winning spree in QM last year and almost got a page of green wins and because of that, I shot into platinum. Coincidence? I’m not sure, but it’s something I might go for again. why or how does QM connect with Ranked?

It doesn’t. Literally at all. MMR is separate.

I’m guessing your winstreak in both are because you are a good player

2 Likes

So it is fair to match a player with people who are much worse than you to force you to either carry them and thus furthering the boosting problem or having said player lose cause of people that are really bad making it seem like the good player isnt as good?

Yeah that is gonna bring a ton of new players to this game. I am sure people love that.

Blizzard answered this already. Stop your complaining.

1 Like

That’s me there.


Part of the issue of any mmr system is that it is conceptually flawed from the onset.

That’s part of why you’ll see similar topics, use of words, gameplay experience (etc etc) on other games that also have multiplayer team-based matching (such as lol or dota)

However, since many of those sites have abandoned their forums, some of the accessibility to the archive of those types of posts (and animated gifs) are lost to the ages :frowning:
While the blue post has some merit to it, it does echo some of the issue of mmr: the claim that it tracks “skill”

MMR does not track “skill”. Now there is a correlation between players with more skill having a higher mmr, but the qualities we associate with ‘skill’ are not apart of the metrics of how a system makes mmr, and then uses it to create its prediction of a ‘fair’ game.

Because we’re told to associate mmr with ‘skill’, when [we] see something different from what [we] expect (disparity in skill,) [we] assume that the matching has failed to make a ‘fair game’ and it is ‘forced’ (or compelled) to use weaker players instead of stronger ones.

As much as I am associated with being a stickler for definitions, computers (or ai) are even more so the case, and you can’t really argue that with them. Artificial Intelligence, Match Making algorithms and other computer-automated processes use a different rational than we do, what is ‘logical’ for us isn’t the same ‘logic’ that those systems use.

So part of the issue matchmaking systems is that objective evaluations of itself, and even outside comparisons, that look at the whole picture, can see the system as “working as intended” when using the metrics the system actually uses, rather than basing it on the ‘rational’ that we feel from the matching. (our empirical evidence if you will)

And that’s part of defect of the whole ELO/MMR/TRUSKILL systems: it has to make certain assumptions on a starting point, a progressing point, and then it has to compared those against other people already in the system.

So long as there isn’t an objective standard to expect/enforce for exact skills needed at specific MMR values (to reduce the assumptions the system has to make,) players are generally going to run into gaming systems that “compel” their “fair” matches with ‘weaker’ players than others expect because those players don’t get to see how that player got to be where they are in the ‘mmr’, they make a different assumption than the system makes, and uss a different term for the conclusion observed.

You also seem to be confused. The match making doesn’t happen by MMR alone. It happens by MMR + uncertainty + time constraints. This means that having potatoes to balance you out is not out of the question if there’s not enough players at your own MMR.

Whatever AZJackson said in the other thread is a lie, either be omission or by lack of knowledge.

1 Like

On one hand, I do notice this kind of behavior, forced or not. Sometimes it appears to be. The tough part is, that if you run the theory, actually putting together a team of 70% winrate at 2000 MMR vs a team of 30% winrate at 2000 MMR, the latter is likely hard favored: the first team is probably lucky, and not as good while the second team has seen bad luck.

What I think the issue is that it’s ultimately random, and there are many elements to it:

  • Say, about 50% of the playerbase is a hard brawler. Means that a good chunk of time you get the whole spectrum of roles, but quite often not. Heck, five man push might even work, but soaking makes it a four man push which doesn’t.
  • Fewer than 20% knows how to play tanks so it’s likely your tank is just doing a half job, trying.
  • There are various types of players, divers / committers and carebears.
  • On top of that you have the rest: mood.
  • On top of that some heroes are really underperforming. I’m recording heroes with 20-30% winrates.

At the beginning of last season I had great winrate with two heroes, then both heroes have been buffed. As a result, my winrate plummeted with both, sitting at 50% and 40% respectively. Totally logical. No, I didn’t climb from Bronze to Diamond to explain it with challenge.
In contrast to that, for a whole 5 months I couldn’t win a single tank match, surely I was tilted by the end. Now it’s my best performing role, with a 9:3 hero leading.

Doesn’t just happen to me. I’ve recorded 33:32 Li Li performance by other players. That’s essentially a 10:20 followed by a 20:10.

Fun fact. The other night I got in a foul mood about the game and had a drink, so decided to fool around and drop some rank. I could barely type or walk to the loo but scored 6:6 (was 6:5). Either all my gameplay is deep down below habitual, or you make little impact anyway.

So. While I do notice that sometimes the matchmaker creates matches where it’s rather clear that I’m supposed to carry, there is plenty of random alongside the tilt and one might see a pattern where there is only a line.

And call it by what it is referred to by the devs: Compelled Fairness . It’s when a player is overperforming, and instead of putting them with and against better players, they are placed with equally underperforming players so both players have an average of 50% winrate going into their next game. indent preformatted text by 4 spaces

This just isn’t true. The only time I see a low rank and high rank player together is when they are partied, and the rest of the team is filled with either:

  • Other people at the average rank of the party
  • Other parties with the same average rank

Blizzard isn’t matching potatoes and skilled players on the same team, parties are. I agree with your other criticisms that it takes WAY too long to converge to your true MMR, hence skill seems to rarely match rank (MMR) in this game (at least up to gold).

Just to add, rank is the same as MMR now. There’s been Blizzard posts about this. If you play SL you’ll notice the above is almost always the case (rank disparities almost always follow parties). Also, you can no longer see when the opposing team is partied.

1 Like

There’s a difference between “there’s not enough players so the MM finds whatever so you at least get a game” and “the MM puts potatoes on your team on purpose for you to lose and stay at 50% mmr”. The first one is what happens, the second one is a myth.

Nope, it is actually the truth.

No. It’s not. He said you are matched by MMR alone and only MMR. In truth your aren’t. Blizzard uses a system similar to trueskill to match players. That means MMR and uncertainty.

There is no technical difference between the system putting potatoes in your team cause there aren’t enough players at your MMR and intentionally putting potatoes to make you lose. The end result is the same: frustration

:rofl: :rofl: :rofl: :rofl: :rofl: :rofl: :rofl:

Funniest thing I’ve ever read in my life.

You just don’t understand, I’m hard stuck bronze 3 I’m suppose to be GM.

3 Likes

First one doesn’t happen.

It can happen when you queue on non peak hours or in smaller regions. In these 2 situations it is common to see rainbow matches.

But for NA and EU the queues are fine and the people you are matched with are close to your mmr. The game doesn’t know who is gonna feed next game, who is gonna throw next time, who is gonna afk next game. MM isn’t capable of that, it just matches people by their mmr.

1 Like

You just don’t understand: I’m hard stuck silver, but I’m supposed to be Diamond / Master. Ow wait, I beat GM on stream with highly unfavored team as silver. Ow wait, I picked Arthas only for 250 games and went to Plat using that (reached diamond next season), loosing all my actual skill in the process.

The problem is simply that bad teammates throw hard, and stupid people keep defending a bad system that works.

To be fair can’t tell if you are sarcastic or not.

But anyways I’ve climbed from silver to master in 6 months wih 70% win rate.

If anything, it is possible to climb, the question is, are you able to?

4 Likes

It is funny how people keep trying to put these 2 things together:

1 - bad teammates
2 - matchmaking

Like, bad teammates are just bad teammates. Most of the time people are just having a bad game, like everyone else. Most of the time people are just raging because they are frustrated with something else, like everyone did one day.

The system is not bad because someone played bad in your games. People will play badly in your team even in GM, that’s just how things go. But the enemy team will also have these kind of players, so in the end it evens out.

Best thing you can do is improve so at least YOU aren’t having a bad game.

“When you queue on non peak hours or in smaller regions, you can have rainbow matches”

Agreed.