Bad Matchmaking Thread

[Jekoz]

Match ID: 24152762

Infernus and Mirage basically didn't understand that's a Team game and you actually also ahve to fight and not only farm...
I checked the Valve Match Score from tracklock and they are basically way lower than the rest of the players and it showed in gameplay.

 

[j3sins]

this is ranked matchmaking

 

[will4zoo]

still getting games with players that are clearly brand new or extremely low skill - I understand this is early morning but ill happily wait longer so im actually playing on an even ground and not wasting 50 mins. This was a 3 stack that should not have been matched with us.

 

[gunslinger19]

Match 24178461

Just horrid. I had to take down 3 guardians and 1 walker myself. And those 3 guardians took a long time. They still had 4 guardians while we down to two walkers. Just bad players feeding again. The enemy team didn't even seem *that* good. It was just our team was so bad it was painfully unbalanced.

 

[maomao431787]

24181157

Feels super unbalanced

 

[i100n]

rating game 24187323

 

[Khanzy]

24193066
Just what the fuck. All five players were awful.

 

[mnkyjamjam]

24203594 most dumb game ever

 

[zkimi]

24204235
48 KILLS VS 8 KILLS

Very balanced game

 

[Cat-Shid-Hunter]

@Yoshi

MatchId: 24130808

I think I figured out something which I am almost sure you already know. The matchmaking algorithm is struggling with group/parties that have a wide variability in skill level. In Dota 2 they ultimately ended up not allowing matchmaking with parties over a certain disparity. I think ultimately as you can see they enemy team has 2-3 players in a significantly higher skill level compared to the average of our team. In theory the algorithm should match up the average of both teams to be roughly similar, however, I think past a certain skill level a few coordinated higher level players can snowball a game quite hard. I think it might be worth consideration to inflate the matchmaking number (for matchmaking purposes specifically in party games with high skill variability) for higher level players on an increasing basis for this reason. this multiplier may need some refinement and tweaking but ultimately should depend on the number of players that have a higher outlier skill level compared to the lobby average as a whole. I think this could provide more well balanced games.

I consistently experience this from the other side finding "stompy" games when playing with a friend who drags down the average lobby skill level.

Also our high level players significantly underperformed and their players overperformed, which may mean individual hero mmr may need some adjustments of course.

 

[finchshts]

I mean I'm not complaining, I can do this all day but since you asked... 24209539.
Personally I'm a fan of games where the non-ranked matchmaking isn't super intricate, just let good players be good players and sort of loosely match. Ranked is a different story, my first few placements were stomps like this but then it got much more even.

 

[dork1337]

I have 100 hours and just got matched up with a guy who didnt know what i meant by "go defend blue" he goes "whats that mean? blue lane?"

 

[SnappierSoap]

Only me and Beepbop above 20k



I mean i am pretty much guaranteed a loss at this point

 

[y3110w25]

Insane Stomp. My team didn't have any comprehension how to lane, farm, team fight or even communicate. When support dynamo is the only person remotely making something happen there is a problem.

The Wraith died several times with ult available, Yamato dove tier 2 tower at minute 5 3 times. The list goes on.

 

[Cmphp]

24224589



Every lane was one sided. 70k difference in 25 minutes.

Why are these players matched in the same game?

This cannot keep going on. How many more examples does Valve need to make this right?

 

[kululuslabo]

ID 24222339.
At least two players in my team fairly new, Abrams was on his alleged second game against people with probably 70+ matches under their belt. How is this normal?

 

[EmDash]

Match ID: 24227427
Reason: Seemed like the entire enemy team was very inexperienced. The enemy bebop kept ulting on the ground next to me (infernus) and I'd just leave fire on the ground and kill him instantly. This happened 4 times in a row. The enemy warden didn't know how to use his ult, and the enemy dynamo didn't seem to know where to be in fights. Just seemed very inexperienced. (They all went very negative k/d)

 

[SamaMac]

Match ID:24231183

got beat by 100k souls in 19 minutes.

 

[robsnake]

ID 23362793
24 minute game with a Seven who went 43-0 and was able to spawncamp our team inside the protected base area. Meanwhile the Infernus on my team had 5k total player damage by the end of the 24 minutes. I just watched the replay, and Seven spends the whole time running around jungling knowing exactly when/where creeps spawn, only coming out to get ganks on our team. Their Wraith abandoned early and he was still able to carry the whole match solo. Honestly there are times watching the replay where I wonder if he was cheating, but idk how to tell.

ID 24209289
Quick game where the enemy Warden stomped our whole team by himself. My team had four players who didn't deal any objective damage.

 

[elmeri.rusi]

Match 24230969

McGinnis and Shiv shouldn't have been in this match.