I'd Like Some Clarification Over a Hot Topic in the Community Right Now.
This is specifically requesting clarification from someone at BHVR.
Does MMR turn off at certain times of day?
Or is it that it expands its parameters (because of long queue times) so much that it takes whoever is available, regardless of their rating?
Comments
-
It expands. That is why there is a 1900 soft cap. If it sill can't find anyone at th soft cap then i don't doubt that it opens up even more.
3 -
I understand it expands, but my friends who have 4K hours shouldn't be getting sub-50 hour teammates while playing against extremely good Killers.
Likewise, my 4K hour friends shouldn't be getting literal 10 hour Killers.
6 -
A dev confirmed before here on the forums that the system prioritizes queue times over accurate matchmaking
4 -
The system apparently just ignores mmr if people quit out of a lobby and the lobby tries to find a replacement. At least that's the impression I got from reading the dev blog.
What is likely happening is your 4K hour friend is filling in a spot that someone dodged or someone is filling in a spot that someone else dodged.
1 -
I think it's important to remember that your side of the queue is only half the story. The newbie might have been sitting in queue for a while so the MMR system said heeeeey so you've been sitting here for a few minutes, go face this 4 man SWF. It'll be fine. Go play!
3 -
He was the first (other than the Killer) in the lobby. I could share some clips, but idk if it would be naming and shaming.
0 -
BHVR; Is this a feature or a bug?
Everyone; A disappointment, that's what it is.
10 -
I've joined into lobbies with 2 survivors in it and if no one joined within the next minute it wasn't uncommon to see the survivors leave (leaving only the killer remaining) and then shortly survivors start populating the lobby again.
It's still possible it was a lobby that the game was trying to fill due to people leaving.
The fact that there is a situation where the game ignores mmr just to fill lobbies ultimately makes mmr a useless tool.
1 -
I found myself from playing against some of the best survivors I have seen to playing some of the biggest babies I have ever seen as killer. As survivor for the the first time in a long time I have seen a survivor who was genuinely rank 20, was running around dropping pallets instantly despite the fact that the killer dropped chase with him a while ago.
So when I hear they say they have toned downed MMR I just personally feel like they have turned it off entirely, I will say that the game has been more enjoyable recently thanks to this.
0 -
I think it's to do with the fact that there aren't enough killers in the game right now to support all of the survivors queuing up for games at peak hours...
Meaning this throws off matchmaking for a lot of games at regular hours. Also means that it can only get worse at off peak hours...
MMR is really throwing a wrench into this game... but that goes to show the effect that survivors have right now. They really need to do something about SWFs if they want to keep MMR in the game.
1 -
Please don't make this an us vs them post.
2 -
I didnt think I did. I was explaining the problem with MMR, which correlates with SWFs and lower numbers of killer players. All a guess but I imagine it's not far from true
1 -
Its certainly an interesting topic. I've tested various times and regions and found no notable difference because each experience was completely random. I've had complete babies in the morning and complete gods late at night, with no noticeable pattern. I don't think its something you could ever prove one way or the other without thousands of games to mitigate the insane RNG that goes into the data. I've heard claims of literally anything which to me, means there can't much truth to any of them. I've heard only good survivors play in the morning in EU, but the good survivors are in the evening in NA.
If the claim is that in the morning there are people that don't have jobs / school that can afford to play DBD all day, that also implies SWF isn't as big of a deal as people claim it is as there would obviously be more SWF whenever people get home from work / school. I've also found the weekend has no real impact which also negates this theory.
tl;dr none of it makes sense because the system doesn't work to begin with. All discussions on the subject are pointless because nothing can ever be proven.
3 -
I have also noticed that the only thing consistent in my matches was how inconsistent they were, especially as Killer. However, it's definitely been getting worse recently, at least for me. My SWF usually plays around 7:00 PM EST - 2:00 AM EST. I have noticed that we get consistently new players. If those players are on our platform, we check their hours and we've only had one teammate that has had more than 200 hours in the last three days. Strangely, the Killer's hours are still inconsistent. We've seen as high as 8K hours and as low as 15.
The one time we played earlier in the day, at around 2:00 PM EST, we got two teammates with 6K hours and a Killer with 200 hours.
However, this means basically nothing. I understand that there's really no way to prove anything without gathering a massive amount of data.
That's why I've tried to appeal directly to BHVR as they are the only one's really capable of gathering that amount of data.
I also understand that the odds of them saying anything are very slim, but it's pretty much the only avenue available to me, so I figured I'd make the attempt.
0 -
You have to keep in mind that the system itself is designed to force you into an artificial 50% win rate.
The easiest way to accomplish this is not to make balanced match-ups. Quite the contrary. It's by making a series of lopsided trials that average out to the desired outcome.
So, have you been escaping more than your 'fair' share? Congratulations! Now you have to boost someone else's mmr who's got a bad streak.
Are you on a downswing? Worry no more! We'll put you in a gg ez game.
It doesn't matter how many hours you have. It only matters if you're winning as much (and only as much) you're expected to.
4 -
That's actually surprisingly consistent with the stats I've been tracking. Games that were stomps almost guarantee an easy 4E next game.
Consider this a sneak peak at the next Totem Experiment.
Post edited by Pulsar on0 -
Sorry, i couldn´t read your answer from the other thread because... well...
Could you repeat it please, so we can continue here?
0 -
Oh I really hope that's not what the system is doing. I'd prefer to know this is a combination of bad RNG and parameter-less matchmaking over something that manipulative.
1 -
This seems painfully accurate.
0 -
At this point, I sometimes think that it turns off completely. This was a match which happened some days ago:
This was before Grade Reset and you can see that the other two Survivors were Ash 4, meaning they did not play much during that month. But also, look at the Perks. I dont know who was the Survivor at the bottom anymore, but the one above was a David.
And both of them were really crouching around A LOT. The Killer on the other hand was decent. The match overall was a 4K, because it was basically a 2v1 between my SWF-Mate, myself and the Killer.
The thing is, I dont know if our MMR is high or not. Because if you die often, your MMR becomes lower, regardless of what you did before. And when you get more of those teammates, simply because they are actually at your MMR, you might drop even more.
But MMR or not, it should NEVER be possible that I (almost 7000 hours) face or play with players who have way below 100 hours. This just does not make sense.
I dont have screenshots, but I also have the same experience as Killer sometimes - good team, followed by 4 new players. Or, when playing Survivor going against Killers, suddenly there is a Killer player with 40 hours.
3 -
why do people keep saying this? this 50% winrate thing comes out of nowhere. if you lose 10 games in a row you'd lose like 200 MMR. a 1700 can match with a 1900, so even if you were 1900 and lost 10 games in a row you'd still be going against literally the exact same people. there's no "50% winrate switch"
0 -
There's no hard evidence for that, you are correct.
It does tend to feel that way that way though. If we got destroyed and the Killer got 9+ hooks and 3+ Kills, the next match was almost always a 3+ escape. That could just be the inconsistent nature of the system (it likely is) or it could be something else.
All we really have is speculation since BHVR doesn't want to talk about their system at all.
0 -
Yeah, I said you were more optimistic than me and that I wish you were right about the system.
0 -
The moment i lose faith, is the moment i quit playing for at least another 3 months
0 -
I'm already "retired".
It's a joke in my friend group that when I'm retired from a game it means I won't play unless someone else wants me to.
I never did think it would happen with DBD.
0 -
Idk,
For me, when I play on PC, my account has about 6 hours on it, and I am terrible with mouse and keyboard.
I get teammates and killers with thousands of hours. And I play like a player with 6 hours on mouse and keyboard.
It's why I am sticking to console, at least there I'm competitive.
0 -
It doesn't switch off at all at any point.
What is probably happening is that there's either an extreme on one side ie either too many players or too little and matchmaking expands to fill them....or there's "backfill" happening, where lobbies get dodged etc and them someone is put into those lobbies to fill them which then throws the matchmaking out.
3 -
All my friends stopped playing DbD. So taking a break is kinda easy for me.
0 -
My last 2 games.
Game 1: Baby Legion. He had all Legion perks with NOED. 60 hours player.
Game 2: Blight with full meta. +3000 hours.
Can anyone explain how this is possible? Blight had high grade, so i am sure he is playing killer so much. And it was normal to face with him because i had good winrate today. But what about this poor Legion? He had just 2 hooks and one of them was because of NOED.
0 -
The system is just so screwed up.
0 -
It takes queue times into effect, no? I would think it gradually expands as someone waits? In the extreme case when only the best killer in the world and the worst survivors in the world were available, you’d want them to play eventually, right?
Due to the nature of matchmaking, where there’s always going to be a difference in MMR, it would be nice if the game at least told the killer what the expected number of kills was for the match. If they get 0K/4K, they can at least know whether it was a particularly bad/good match for them or just matchmaking trying to shorten their queue time.
0 -
Other games solve this by showing you how your rating changes. Big positive rating change = you were outmatched and won a hard match, big negative rating change = the enemy was outmatched and you lost an easy match, and so on.
Of course we are not allowed to see our rating, or how it changes, because of reasons.
0 -
Yes, and it works well for other games. I would hope that if the player base of Super Smash Brothers could handle knowing an indication of their ELO, so could DBD's. I was trying to think of ways to make MMR more transparent without explicitly telling the player their MMR.
1
