New matchmaking test this week
Behaviour just announced there is a matchmaking test this week where each day they’re trying different setting on the algorithm’s patience, so to speak, in trying to find a close match versus a quick match. And they’re asking for feedback on how the match “feels” each day.
Good for them looking for feedback, although I’m not sure how useful that feedback is when it’s impossible for us to know what our relative MMR is compared to our opponents. If I felt like my matches weren’t fun is it because I got stomped because I played badly, or was it actually a mismatch where I was against higher rated opponents? Or was someone playing really, really well but their MMR is much lower than it should be but I can’t see it?
Maybe they’ll get something useful out of the feedback, it can’t hurt. But honestly if they want us to be able to evaluate MMR from a personal perspective we kind of need an indicator of where our MMR stands in relation to our opponents’ in a given match. If not the actual number then at least a rough indicator of who’s in the higher or lower brackets than whom.
Comments
-
Though cool that they are trying new things, but is it really going to matter? If the current matchmaker works as it should but heavily skewed due to role saturation, how much is really going to change?
0 -
I believe until the kills/escapes based mmr is gone,nothing will change.
15 -
Yeah, I also think that this experiment is not altering the kills/escapes based MMR. It only changes how the system picks players into a lobby based on their kills/escapes based MMR.
9 -
Sweaty players should play with each other and good players should not team with baby players. Baby killers should not face with sweat lords. Hope they fix issues.
2 -
The system of kills and escapes as the on,y metric hasn’t changed therefore this is essentially meaningless
3 -
I understand what's being said on this thread...can I suggest trying the MMR tests and supplying your feedback based on those tests - without preconceived ideas about whether it's going to work or not?
I can assure you, as you can see by the fact we created a separate section for this alone, the feedback is something we're taking very seriously. And I will be monitoring it daily.
11 -
i thought they're just implementing backfill improvements because apparently it just finds people randomly for lobby dodging. the other aspect they said they're improving is disconnects & ping. if your able to play near 2 servers.
0 -
Just a quick suggestion, it doesn’t look like the matchmaking forum subfolder has any instructions or anything, just a folder for March 28. You might want to pin a short message saying what sort of format or feedback you’re looking for if you want anything specific.
1 -
I gotta say, something feels better. I can't quite put my finger on it, but matches just feel good and not sweaty as much as prior. Whatever they've done feels better.
0 -
That's encouraging to hear.
1 -
For once it's good. Hope this is just the start of more to come with more tests coming.
0 -
Exactly this.
0 -
Can we be given some kind of indicator of what our MMR is and what our opponents MMR is throughout the test?
0 -
An indicator of what MMR is going to be is not going to be part of the test no.
1 -
Hahahaha you do know that will never happen right?
They won't show what your MMR is because of elitism/toxicity, so little to no chance that it will be shown during a test.
If they do show it during the test, then i will eat my cap 🤣
0 -
The problem is "not influencing the test with preconceived ideas", but how to pass the feedback accurately to devs without knowing what has changed (and this is no double-blind clinical trial)
I'll explain better, even with normal matchmaking, it's not like it's random enough to have good matches in a row and the next day all bad.
1 -
On a tangent I never understood argument personally since they still show the Grades which leads to just as much elitism as the MMR, only of course it’s sillier in that case because Grades aren’t even part of the matchmaking at all but a lot of people still see them and assume if you’re higher/lower grade then it means whether or not something you say matters.
2 -
With how the current system works (Kills and escapes= Skill), this is not gonna change anything.
0 -
You can give your feedback based on things like "how long the wait time was - better or worse" "how did it feel the match went - did it feel that you were evenly matched with the other players etc".
2 -
Will each day have a different MMR calculation,because I may think that the first MMR way to calculate being a crap and the other one being awesome and say it doesnt work.
0 -
Yeah i am with you, i never understood it either.
Like there is nothing gonna change from when we had ranks.
Now they still think the grades is an indication and use that in an argument, or they say they are high MMR so they know what they are talking about.
But there is no indication of an mmr score, and even if there was it would still remain the same.
Whether or not they will show the score, people will use it anyways to strengthen their argument
2 -
If I had a guess, I would assume one of the days will include a test for survivor MMR having a team based component. Some days will probably just be a change to the matchmaking algorithm in how it uses the score. This is just a guess based on the last Q&A.
0 -
THIS.
Band-aids, or tweaks would never fix the MMR, it needs a total rework.
1 -
The real issue is survivor MMR isnt based on their skill it's based on escapes...
While the killer has more control over their MMR the survivors do not, if your team dont do anything you die, if a killer wants to hard tunnel and camp you, you die.
A single survivor has no control over their mmr, while playing good will increase the chances survivor MMR is still down to teammates and how the killer decides to play.
SBMM sucks. Waiting longer isnt gonna help it.
0 -
I worry the RNG aspect and other factors are impossible to see how it affects the test.
1 game I could play a killer with no addons and get stomped so it feels bad; the next bringing strong addons and stomping or survivors killing themselves etc.
Also surely factors like maps can affect both sides with some being easier to play effectively depending on killer chosen or survivor knowledge etc.
Can you say anymore about this or do you have to keep it all unknown in the interest of it being fair for everyone?
0 -
First game, as Nemesis, a killer I almost never use : awful game. After 3 gens I left the room and went for lunch.
In insight I didn't tunnel nor camp. A mistake I will not repeat.
0 -
The test is taking place in main game or PTB?
0 -
Second game, as Nemesis : F* this #########. I'll try with a killer I'm good with afterward. This current setting would make me leave the game for several months.
0 -
I don't know if this is still going on. But today was one of the worst days of playing survivor I have had in over a year. Super long match making times (fairly normal this time of night for US), and every single killer was sweating for a win.
Game 1: Fairly average, Nemesis with No-ed got 4 kills with that perk
Game 2: Wrath - Full slug build (knock out, third seal, silent bell, Able to see survs when uncloaking), After downing everyone he waited for them all to bleed out. I don't understand why there is nothing to stop people from bleeding out for 5 minutes.
(I even had unbreakable but the map gened with nothing on half the map. (I also ran windows of op, I thought I forgot to equip it because of how void it was))
Game 3: Nemesis (Scorge, pop, ruin, tinker)- slugged multiple times, all teammates urban immersion. Lost after 1 of the last 2 survivors got crows with 2 gens at 80+%. I stopped mid chase to get a glyph to show I was just playing for fun. I could have easily gotten to a window. He hit me as soon as he possibly could.
Game 4: Pinhead - Ultra rare addons (scorge, dead mans, pop) - 2 teammates camping chases with flashlights, never got a save. Lost when gens didn't get done.
Game 5: Wraith - (3 carry perks, forgot the 4th) Got tunneled out of the game before anyone else was hooked.
Game 6: Artist - (Scorge, dead man) Average, killer slugged for 4k. 2 of my teammates hid for hatch. All surv died.
Match making is awful. I get the worst teamates with the sweatiest killers. The only way to win in this game is to make a 4 man premade. Which isn't fun for killers.
0
