9/10 the games I play are a 1 sided stomp with almost nothing I can do to win because my team or enemy team is just that unskilled.
There also doesn’t seem to be ANY consistency on the skill level of my teammates
Take 5 players from the que, average their MMR (so 1900+2300+2800+2100+2400 create an average of 2300 when divided by 5) and match them with another team made up of five other players whose average is around 2300. Their highest player could be close to 3k while the lowest is way under 2k.
Block mirrors if possible, force tanks on both sides if one side has a tank, force a healer if both teams have a healer. Force support if possible (but rule can be broken if que is extended, like mirror rule)
Those are the only rules I know that exist and are documented on HotS wiki through dev comments and patch notes.
Quick Match has at some point apparently also tried to match people playing a higher level hero against people on higher level heroes, or a low level hero with low level ones etc., but I don’t know if that rule exists with the current population.
QM also has MMR so it’s not a consequence-free mode like AI or ARAM, you will get worse games if you lose a lot.
Players should be matched with and against players of slightly lower skill when playing a Hero for the first time, and slightly higher skill when playing Heroes with which they have extensive experience. This results in a higher percentage of evenly-matched games, in which each team has a 45% - 55% probability of victory, and should encourage more players to try out a greater variety of Heroes in Quick Match games.
(I don’t know if this last rule exists anymore. It was first documented many, many years ago when this game had much more players in April 2016)
The matchmaking system will now account for each player’s Hero level in order to better define match quality and calculate win probabilities more accurately when attempting to create Quick Match games.
5 Likes
Meanwhile now matchmaker pair two solo que sup heroes together againts Aba
And bruiser vs non-bruiser teams are more common now when both team has a tank.
And teams with multiple high mmr players can force low mmr players into enemy team if they 5 stack.
Just look at this from my last game.
Enemy team:
Valeera: 2942
Falstad: 2735
Naz: 3130
Murky: 2505
Lunara: 3002
My team:
Kerrigen: 2769
Thrall: 2115
Greymane: 1833
Hanzo: 1664
Maiev: 2681
So I get 3 low mmr players on my team and only me and Maiev are high mmr againts a team with multiple master mmr players.
Averange mmr of both teams 2841 againts 2212 so a staggering 600 mmr diffrence on my team againts the 5 stack.
Just proves there are something really wrong in this game. And the game was found after it went extended que for 1 sec.
More hopeless matchmaking I have never seen and its not the first time I experience this.
4 Likes
That looks stupidly unfair, double checked the math and there really is a ~600 MMR average difference. One can only hope this is because some of those players don’t have enough games uploaded.
Though if the very low rated players (1,6k and 1,8k) kept feeding or doing stupid stuff repeatedly, it is likely accurate. 
I would rather wait in que 1 hour and not get a game at all, than be thrown into that match.
Did not check the rank of my own team but Naz was GM#36. Valeera was Diamond 2 and Murky was plat 2
All of them had winrate around 56-77% on the hero they played with Valeera as the highest.
So yea this is just one those games I just wanted to quit.
These rare cases happen, not related to your waiting time, but theirs.
You somehow managed to go to lvl 19, which shouldn’t even happen.
Even with the obvious loss, it only took away -0.41 MMR on your side. The other team even loss MMR despite their win.
A worst case was this one:
5 stacks vs a single 2 stacks, similar MMR, but you lost 9 times more MMR. There’s like no compensation whatsoever for 5v2.
Yeah that massive mmr difference is very very common. Obviously theres nothing the significantly lower mmr team can do.
MMR has always been a lie. They organize games by player/hero win rate and party sizes. Win too much? Its time to lose to a 5 man. Perfectly balanced and fair.
1 Like
I meet the same people playing 4 or 5 man a lot and if that only happened because of winstreaks it would mean i actually win a fair bit!
I certainly do not. Im average at best.
Not related, just small playerpool and unbalanced matchmaker that accepts a MASSIVE difference in mmr.
Since you can only look at team composition after the match there are clues to almost always knowing when its a 4 or a 5man.
-Abathur, no counter character in your team.
-Mediv but no abathur in your team
-Zarya, once again no counter character in your team.
-Double front line, melee and ranged assassins instead of a random and always double range assasin.
Double bruiser, only one bruiser in your team: The game only checks against having one bruiser so queueing as double almost always fights against a single bruiser team.
Then ofcourse theres just remembering the same names you meet over and over and over in higher QM mmr. (that means gold to high master QM mmr since it can range that much)
The truth is, stomps happen even in pro games and not because one team is better than the other. In a series where both teams play against each other multiple times, both teams managed to stomp each other alternately (because different map/heroes/strategy).
The snowball effect, a small lead makes it easier to increase the lead. Stomping is pretty natural in a game like HOTS and is not necessarily a sign of bad matchmaking.
On top of that, stomping isn’t really defined. When does a decisive win not count as a stomp? ^^
1 Like
Unless u improve ur fundamentals of the game u will remain a ~50% winrate player because when there is no strategy involved it’s a bullfight where the stronger bull(comp) will almost always win which is indeed a 50/50 chance.
It is. When a match gets so uneven that one team are forced to defend whole game and cant win anything then you call it a stomp. Winning againts a team that dominated you whole game but you were lucky to catch them offguard and win the match is not a stomp.
1 Like
Well i would consider this a good try ^^
So a 30 minute game can be considered a stomp, if one team is always ahead? To me, this definition is still too vague. I had always thought that games over in 12 minutes or less were a stomp, regardless of levels, kills, or the like.
But given the number of threads that seem to be nothing but whinging about the MM, I guess most people consider any sort of a loss a stomp.
3 Likes
I agree with this sentiment for the most part. The only exceptions are when a team (usually a high MMR 5 stack) is deliberately drawing out a game to run longer. I had one game on TOD where the the 5 stack wouldn’t cap any shrines to win or push all 3 of our forts to win. They kept 2 forts down and simply farmed anyone who tried to cap a shrine.
I know this is an extreme case, but here and on Reddit people have related the same experience. I’ve been in less extreme games where a skilled 5 stack was at our core and could have easily finished, but farmed my team and just waited for our re-spawn to do it again and again. In doing so they made the match go from what should have been 12 minutes max, to 20.
1 Like
Nice to know I am not the only one who has expericed that kind of crap.
I had a 5 stack with 3 stealhers and Butcher/Reghar comp againt my team that only had mages and low hp heroes like Hanzo. We got farmed out of the game but it did not end there. They would afk in the middle of the map and wait for us to come out while transporter on Hanamura was active and then have Butcher charge the first one he saw.
the 3 stealhers would then late game walk up to our core and dance infront of us where core cant hit them while Butcher/Reghar would afk in mid again. Waiting for us to come out.
In the end minions came so far into our core that minions alone was killing our core while I was sitting in spawn with 3 bots which left the game after 20 min of getting farmed.
But the abuse did not stop there. One of the stealhers then spammed me with laugh emotes to me on /w.
So 40 min game where we got farmed like it was 1 hour AI game. They properly also got much of exp from that match I bet.
But I have only tried it once in 14000 games so Its nice it was only this one time it happend. You need to more evil then BW to do such crap to others.
1 Like
No, no, the difference is Brightwing doesn’t know she’s evil, they do! Yes, it’s been completely rare for it to happen to me too, but it does happen. The second example I gave where a dominating team deliberately draws out the game to extend it by a few minutes happens more frequently, but it’s still not often.
I am surprised to hear this stuff happens on the EU server and I saw your post about the 600 MMR gap you experienced. I thought that would only happen on the US server or smaller ones like SEA.
1 Like
*Force Supports on both side if one team has one. Checked only if there are no Healers.
Same with Bruisers and Tanks. But Malthael, Ragnaros, Thrall and Xul can be matched against Melee Assassins as well.
1 Stealth / team unless parties.
Cho’Gall, Murky and Aba cannot be matched in the same team unless parties.
Try to match parties against parties.
If parties are matched against solos the team of solos have a higher mmr average.
3 Likes
Yeah that was a typo, meant forcing on both sides if 1 has one. Surely it does not force supports into each match 
Though it seems less strict than tank/heal rule and can be broken like mirror rule, if the extended que runs very high.
But I can’t remember if I have had support vs no support games lately in QM. If so, seems kind of rare.
2 Likes