unbelievable 
1 Like
I think you should stop trying to control the direction of topics and let them take their natural course.
You weren’t even part of the discussion/thread and you show up telling people what they can talk about. Unbelievable.
You’re last offtopic derail contributed to a whole thread being removed. Why would you do something like that?
1 Like
Could you clarify what you meant by “unbelievable”? Are you referring to the RT vs AT discussion or something else?
From my experience, when RT players repeatedly face the same AT teams over a longer period, some may become frustrated and leave games because the matches can feel very one-sided over time. So when I joined, others had already experienced this and they started to leave RT. Me also after a few trials, switched to something else. Then later, when I came back, it was still the same farming AT team.
Like Grubby and others also suggested , better separate AT and RT queues again.
Continuing the above idea, I am not sure if the last proposed “MMR scaling / adjustment” has been implemented yet ?
I am reffering to AT being matched against higher MMR RT teams. Does anyone know?
Which MMR is which ? Is the “team specific” = sum of total individual MMR? or else? Can you give an example?
You guy are missing the real issue here. The problem is not how many mmr AT teams have, but that there simply aren’t enough strong RT players left to consistently compete against ATs.
Take some of the ATs I’ve been facing recently for example: teunis, aherren08, and gggl — basically a premade around W3C 1700 level. Against that team, I’m almost guaranteed to win no matter what side they’re on. If they’re on my side, they’re much stronger than my RT teammates. If they’re on the enemy side, they’re still much weaker than the actually strong ATs. So either way, my win rate ends up very high.
(Of course, W3C rating and Battle.net rating can’t really be directly compared, because the playstyles and map pools are completely different. For example, Sifo is around 5k on Battle.net but only about 1700 on W3C, while BobbyOG is around 4900 on Battle.net and 2300 on W3C.)
Meanwhile, when I faced the AT from Invigator and Dise, I won 2 out of 4 games. Against Nawt’s AT, I won 1 out of 3. Those games were extremely difficult. Their level is farmore stronger than the Teunis group. (I use Teunis as an example because their team basically represents the gatekeepers of high-level AT play. Compared to them, some weaker ATs — like kambill’s or kain’s teams or some unknown player’s team— are usually very easy to beat.)
The real issue is that I simply cannot get matched with three competent RT player. Most players in the 4k–5k mmr play likebots. They have no idea how to win games. They run around like headless chickens with zero understanding of teamwork, map control, timing, or win conditions.
And the funny thing is, I already have to wait 40 minutes just to get teammates like that. So if the system were actually trying to give me three decent RT players, how long would I have to wait? Four hours?
2 Likes
Never mind, just got the ‘no win counted’ bug again
Too bad, was basically a 2v4 comeback game as well
Sigh. Fine.
The thing that makes Warcraft III’s Arranged Team system different from the systems in other games is each team of different people you create has its own unique MMR. it is NOT based on the MMRs of the participants. Instead, the team is treated as a single entity- like a brand new player (or one new to the specific mode you’re playing).
Every new Arranged Team simply starts at 3500 MMR and plays 5 placement matches. Just like any solo player. On the loading screen, when an arranged team is in the game, every player on it will show the same MMR.
Herein likes the problem. Whether it’s smurfing veterans or just casual friends, most people who form a group are not completely new to the game, yet the game treats them as if they are. This is why AT vs RT games can feel impossible- their true skill is not represented by the MMR because their solo MMR is not taken into account.
Let me put that another way- If 2 people with 5000 MMR form a team, they’re now going to get matched to ~3500 MMR players and the game will prefer not to match them with higher MMRs.
No matter who you are and no matter what your intentions are, the game essentially forces you to “noob stomp-” even if you had no intention of doing so. And on top of that, if one of your friends leaves and another one joins, you’ve just made a new team- and your MMR is back to 3500 again.
This is incredibly easy to fix by simply using the player’s individual MMRs as normal instead of inventing a special Team MMR. Alternatively, it could also be fixed by using the MMRs of each player in the team to inform the initial Team MMR (i.e. instead of setting it to 3500 in the example above, it would be set to 5000)
Fun Fact: Because of this, Around 3500 is where you’ll see the most AT players. As you drift further away from 3500 (whether up or down) you will see fewer ATs because ATs tend to be near 3500.
1 Like
No.
IF AT’s MMRs were accurately reflecting their skill level, they would generally be getting opponents that can match them. This has nothing to do with the amount of players of a given skill level- if there aren’t enough players, you will just wait longer- and you’ve proven this yourself with your complaints about wait times.
But because most ATs are just permanently near 3500 MMR regardless of their actual skill level, the result is a lot of high skill players mixed in to an MMR range that isn’t supposed to be high skill players.
This is also why I, with my laughably bad MMR, don’t see as many AT players as most- because the game is matching me with other people who also got crushed out of the 3500 range.
in 1v1 I’m actually just staying right near 3500. I’m very average.
But in 4v4? I’m hovering around 2800, primarily because of people who leave when they see ATs biasing my win rate downward.
Not really, Team invigator or team nawt shall have a average mmr 5k+, according to your statement, they should be matched against a 5500mmr RT(Because they are at, they shall match RT with a bit higher mmr).
But as I said, finding a 5500mmr RT player requires a 4-hour wait; otherwise, a 40-minute wait will give me two 4k, 1x 5k mmr resulting in an average score of less than 5000.
（5500 mmr is in the top 0.1% on Battle.net. There are probably around 20 players with 5500+, with Orcs. Other 3 race, I am the only 6k+ Hum/UD, and NE have 0 player 6k+, I think other 3 race have like 10player 5500+ together.
I did not followed so deeply. I thought is based on individual MMR’s, computing some average or whatever formula.
But what happens if a new group of friends joins today: Eve, Adam, Cain, and Abel? Let’s say their team MMR starts at 4000. They win many games during the day and finish the night at 6000.
a) If tomorrow they continue with the exact same setup, will they continue from the last earned 6000, or the system compute again the 4000? (Assume none of them played separately in the meantime, so their individual stats did not change.)
b) If the next day, they lost Abel but Azura joins the team instead, will the new group continue from earned 6000, or will the system compute a new starting MMR for this New team, for example 4500?
(Footnote Remark : I was not so bold to use real names like Lutz did above, so all my names are invented. Is it ok to use real battle-tag names?)
When you play with the exact same set of people, yes, they will keep the same MMR they had when they played last.
It’s just every unique set of players gets their own Team MMR. You can see all of these teams and MMRs in people’s player profiles.
in other words, if you always play with the same people, the system works as intended. But if you play with lots of different people, you’ll basically always be around 3500 MMR regardless of your actual skill level.
So if you play with Jack, Jill and Joe for a day and you get to 5000 MMR from 3500, but then the next day you play with Jack, Jill and Jasper, that is considered an entirely new team and starts at 3500 again. If you then play with Jack Jill and Joe again later, you will pick up where you left off again at 5000.
It is needlessly complicated and confusing, on top of the other issues it casues. It would be much simpler to just get rid of team-specific stats entirely and just use everyone’s individual MMRs, and it would substantially shrink the group play advantage.
2 Likes
Another win not counted.
This time against a group of 3 AT players and one rando. Was a hard fought game with surprisingly no leavers, but getting the “failed to retrieve data” error of doom almost always means it’s not counted.
From the scoreboard after that error I checked one of the AT players profiles we played against, the loss counted against them, but no win for myself.
Fix?
https://imgur.com/a/jNrh7wH
Haha, road to 1hour
Why are you queueing using the CN client? Aren’t all regions linked?
Losses can not get counted too. It’s not specifically wins, it’s games. And it mostly seems to happen on EU. But that’s not what this thread is about.
My advice is don’t stress over stats and just focus on the game. You’ll have more fun that way, even if it is busted.
Well then perhaps I am just terribly unlucky - Every single a time a game has not counted, it has been wins, and never losses. Like I said I viewed the profile of the opposing team where a game didnt count for me and it gave them a loss.
I don’t care about the win/loss record overall, moreso if I am working hard for 20+ minutes carrying two braindead allies to barely get through a game, it is rewarding to see a win for my efforts.
”But that’s not what this thread is about.” It is - I included it in my original post
Its just chinese version, I search in EU Region, at last, 70min pass i found a 180ping us seaver game 
I think this is just confirmation bias. You want wins, therefore you’re far more likely to notice when you don’t receive them.
When this issue occurs, it is because the server is backlogged and didn’t process the results of the game. Sometimes it does in fact show up on your record later (Though, not always…) It has been proven to happen far more around the start of the ladder resets when activity is the highest.
1 Like
Great idea and thanks for the hint! I just ignored him ‘forever’ now. This joker is here all day (6700 posts!! LMAO ) causing nonsensical controversy and discusses every legitimate topic to death although no one asked for his opinion. Not gonna read him ever again. Next is Leviathan and BrotherRoga. They are kinda of the same (or are they the same??!).
1 Like