To prefice this, I am going completely based off of past experiences in regards to my thoughts. All of the following could be wrong. If you have additional/better ideas or understandings please comment!
(No TL:DR! Prepare for a novel!)
Now the matchmaker seems to be getting worse lately. (Without knowing participants 5v5 ranks we really cannot know for sure if its mismatches or boosted players) I have noticed, especially in the last few days, a large spike in players who probably shouldn’t be anywhere near my games. (POA Bronze currently)
Not going to lie, today tilted me hard. I got mad, stopped my stream early, and seethed about what has happened over the last few days. Once I cooled off I started thinking about it constructively.
Here are some possibilities I can think of that I believe could be causing all the havoc.
- Smurf guard.
This is a system in place that accelerates gains to be sure those who are high tier players cannot bully new/low tier players. The problem arises when a new player is accelerated due to substantial luck when it comes to the teammates they are getting, allowing them to win most/all of their early games skyrocketing them higher than they should be as the system thinks they are a better player.
Aside from removing said system, its a time game before luck runs out on those boosted players, eventually leading to them dropping out of the higher tiers and placed where they belong.
With how buggy the MM was on release, this is something I see as a very possible reason to the problems being seen by high tier players.
- Queue variance.
Since all groups but full parties are in the same pool, this adds a level of difficulty for the MM to get through.
For example lets say team one consists of 2 duos and a solo, while team two consists of 1 three man party and 2 solos.
The match, according to the program, could be perfectly matched based on all parametres accounting for the mix of solo and partied teammates.
The problem is how it is weighted. It is assumed that partying gives an advantage so to compensate the MM may be a little more lax on who will be put into the game.
This too me is possibly part of the problem, but not really a core issue to it. What I see happening is the matchmaker having troubles placing a group of three properly, so it adjusts the range assuming that the advantage will make up for the skill difference.
This to me is another likely culprit, increasing the problems from #1.
Since LPQ now bars players from ranked play, the pool of players in every level effectivly shrinks down. This leaves less suitable matches for the MM to find, causing a widening of parametres, and really really bad matches.
Even if the MM is running perfectly, a decreased pool of players will obviously make it harder to find optimal matches.
There isn’t much else I have to say on this one, other than this problem is causing problem #1 to amplify.
Now #1 is what I would like to touch on once more before concluding.
Since this system is new, and there is a lot of variance at play with 5v5 when it comes to winning and losing, IMO it is only a matter of time before we see it level out and start working as intended.
Barring a legitimate bug in the programming, this seems most logical to me.
These same problems occur after a hard reset, and why they are such a terrible idea, as some players end up boosted into tiers they dont belong in based purely on the luck factor with teammates.
Does this sound reasonable to anyone else? Any other ideas or insights? Let me know!
If this is in the wrong place I am sorry!