http://dbd.game/killswitch
Edit: The 1,800+ MMR range is the top 10%
Are we allowed to know what percentage of players are in the 1,800+ MMR range?
We just got official BHVR stats that list "MMR 500 and above" and "MMR 1,800 and above", but we have absolutely no idea what percentage of players this is.
When BHVR says Blight's doing well at 1,800 MMR and above, is that the top 5%? Is it the top 1%? Are we allowed to know how rare it is for a Blight player to perform over the 60% kill rate?
For the "High MMR" survivor stats, are we allowed to know what percentage of players are "high MMR"? Is it still 5%? Is 1,800+ MMR the top 5%?
Comments
-
I would also like to know % of how many SWF there are….
5 -
It would be interesting to see how many people are in high MMR.
Even better would be to show EVERYONE their MMR, so we can see how many players claiming to be there actually are there.
7 -
MMR is a derivation of Elo. While its not always 1 for 1. The calculations, formulas and such are the same.
The only real difference between MMR and elo, is that there is a lot of secret sauce that goes into calculating what the k-value should be (for those who are familiar with elo and kvalue)
Based on that knowledge, we can look at the chess elo distribution on a site like lichess.
Bear in mind ignore the "Your rating" field there, i have only played 1 game of classical chess on lichess and lost it, and lost like 400 points since it was my first game.
Based on this you can see that 1800+ is around the top 30% of a players.
And now you see the problem with DBD in general. If DBDs distribution is even somewhat similar to this, this means that someone who is 1800 is part of the top bracket, the same bracket that current best in the world magnus carlsen is in chess.
2 -
The big problem is that with the official BHVR stats that were released today, the killer stats said "MMR 1,800 and above", but the survivor stats said "HIGH MMR". Is high MMR the same thing as 1,800+?
And if we're making the cutoff point the top 30%, then 30% of 4-SWFs are overperforming?
And why did the previous stats say TOP 5% MMR, as if that was the cutoff point? Is TOP 5% MMR the same thing as 1,800+ MMR?
1 -
I literally have to be at that point on killer. I went for Skull Merchant adept a few days ago and lost 14 games in a row with the best outcome being a draw on Autohaven. No weak links on any team, coordinated saves, perfect pallet stuns and hiding it was miserable.
1 -
It was confirmed a while ago by some people that matchmaking has basically 4 brackets. And the top bracket is at 1700 MMR.
While the game does try to match you close in MMR, if you are 1700 you are still technically in the same pool as otz who is probably like 2500+ for example.
This is why a lot of streamers have such wild swings in play often. Where 1 game they are playing against a SWF comp team and the next is a bunch of randos, because there are so few people at that high MMR that the matchmaking has to quickly expand its search bringing in people that are way worse than you.
For context, even just 100 points of difference leads to a big skill disparity. For example, if one player is say 1800, and another is 1900, the probability that the 1900 player wins is 52%, the probability of a draw is 24% and the probability of the 1800 winning is 25% (these are rounded btw so don't add up to 100).
Realistically you'd want to match people within 50 points at most of each other. But in a game like DBD that is extremely difficult when there are so few players at the highest levels, so you are seeing big differences in matchmaking quality.
The point is, that the game has to balance around waiting a long time and fair matches. There was a point a while ago where they had a matchmaking test for example, and you saw streamers waiting legit like 30+ minutes to find a match. Which really isn't acceptable.
1 -
Knowing how wide the top bracket is, doesn't matter much if we don't know how often matchmaking makes giant leaps. If the giant leaps are rare for the majority of players, then it shouldn't matter much.
Ideally, the only people that should be affected a lot by an MMR soft cap, are the people at extremely high MMR numbers, that have way less people around their MMR number.
0 -
It's the top 10% as per their latest reddit comment.
https://www.reddit.com/r/deadbydaylight/comments/1k58ik7/stats_january_march_2025/mogwazx/
2 -
Thank you! Found the comment!
4 -
We do know that though, they did an experiment a while ago where the made matchmaking more restrictive and force more fair matches. And the result was streamersl ike otz, and scott and true and so on took 30+ minutes to get into a match.
1 -
It really sucks because I had some great solo matches :(
2 -
There is some evidence that there is a hard-cap in place at 2100 MMR rating.
1800 seems to be the "soft-cap" where it takes a long time to drop down from, if you happen to find yourself there and going on a very long losing streak for instance.Looping back to Choy's video from about a year ago, when he talked about the MMR brackets, he mentioned that a 1400 rating being the "gateway to high MMR". It is possible that an additional bracket has been added at 1800 and over at this point, making it 5 brackets in total.
But from what we know, the 4 brackets go from 0-649 (lowest tier / protected bracket), 650-899 (novice-tier), 900-1399 (casual tier), and 1400-2100 (high-tier) - If there is a 5th bracket, the distribution from 1400 and up would look like this instead:
1400-1799 (high-tier) and 1800-2100 (top 10% tier).
New players start at 650 rating, and move up or down from there, based on their performance.
At least from my own experience, I have faced much less sweats in my games lately. I did run into a couple 10,000+ hour survivors a few days ago, but it looked like they were actually just goofing around.0 -
That implies that when the current MMR system places an 1,800 MMR player with an extremely high MMR player, it's an EXTREMELY RARE event.
So I really don't see the problem here. Ok, so you say there's a big MMR bracket. Why is that a problem if it's an extremely rare event for an 1,800 MMR player to really be affected by giant leaps? And for the players with extremely high MMR, there really isn't a better solution unless you want them to be stuck in extremely long queue times.
0 -
So… why not make the graphic read "Top 10%" instead of "1800+ MMR", especially when no one can actually tell their MMR in the first place???
0 -
That's the opposite though. If the point was that say otz has 2100, or something which is rumored to be the cap. And if they have the matchmaking super strict on making sure its only similar MMR players, and it takes 30 minutes for him to find a game. Then when they turn it off it takes 1-2 minutes. That means that its matching him with far lower players most of the time, for example.
0 -
It would be nice to have an update of this for sure. The last time we got one was in 2019 I believe.
1 -
If an incredibly small percent of players, spends most of their games with giant leaps of MMR matchmaking, it still only affects an incredibly small percent of public games.
1 -
most players (especially survivors) are bad and i feel like the dbd community would throw a fit if they found out how bad they really were
-2 -
I'm gonna guess the majority of players don't care. Or are even aware of things like SBMM or MMR. I think people who take this game seriously would be more shocked at how many play this game at the most casual level, and just don't really care very much at all. And how these people end up in all of our games.
I think that's an internal and fundamental disconnect much of the player base has.
People like we on this message board even discussing this stuff are deep in the minority, I think.
They're never going to make individual MMR scores public, and that's a good thing. People would immediately start using their scores and the scores of others to be divisive as hell, and it wouldn't actually change anything.
One could say to just show people their own scores privately, but that would turn into a "why won't you show me your score?" sort of elitist thing.
I mean it would be kinda gratifying for some cocky people to find out they aren't what they though they were, but it wouldn't be worth all the crap that would come with it.
And regardless, BHVR has reportedly gone to some lengths to prevent people from seeing their MMR scores (Choy made videos about this), so it seems unlikely in the extreme that they're just going to start handing them out.
6 -
They comment 10%, as shown in the thread, but we still don't know how big of a player pool they are counting. Is the player pool everyone who has ever launched the game, everyone who played at least once during the trial period, or some other number?
0 -
Goodness gracious so much sweaty interest in a number none of us can even see 😔
3 -
I would love to see mmr and love to know how they work it out. The only reason I think I might be in high MMR is due to the frequency of swf I get (sometimes the same team repeatedly) but also the amount of wins I get. I literally went on a 50+ match win streak couple months ago with every match getting at least 3k but a lot of 4k. If I'm not in high MMR then it begs the question why so many swf and what does it take to get to high MMR if long win streaks and very few losses don't do it?
-1 -
That doesn't seem that low to me…
I would guess like 6-7% on PC for 4SWF.
0 -
I'm not concerned about that question, because other games that have MMR aren't concerned with that question.
For example, above is the SF6 MMR distributions, and it doesn't bother to tell us more details, and people really don't complain that there aren't more details to this chart. It's normal for games to release stats like this.
Having said that, it would be super cool if DBD released a chart like this.
2 -
I think one of the reasons players would question MMR less in a game like SF (or fighting games in general) is because the criteria are more cut and dry. Skill is quantifiable (and actually matters), win conditions are clear, measurables are all very measurable, etc. One doesn't have a lot of reason to question it.
I think the general reason the DBD community is so curious about the details is because it's pretty unclear how anything is quantified, and questionable as to whether what is known is even applied to SBMM with consistency.
I mean they say people above the soft cap comprise 10% of the player base, but it seems evident that the matchmaking doesn't reflect that, as new or lower skill players leak into higher MMR games with regularity. Knowing how many active players comprise the other 90% might help explain why.
In theory it's possible that the 10% of those at or above the cap comprise a much bigger percentage of active players, depending on how/who they count.
It seems reasonable to assume that BHVR is stingy with details because it wouldn't reflect favorably at all.
3 -
It seems reasonable to assume that BHVR is stingy with details because it wouldn't reflect favorably at all
I think the general reason the DBD community is so curious about the details is because it's pretty unclear how anything is quantified, and questionable as to whether what is known is even applied to SBMM with consistency
About that. The general consensus about the rating is that only wins and losses count towards your MMR
For survivor, you gain points if you escape, or lose points if you are killed/sacrificed. You can have the best match ever, clearly outclassing your opponent, until you exhaust all resources and get caught. I managed to double-pip despite being hooked and sacrificed in the endgame, but I know all too well that I will still lose MMR because I did not escape in the end.
And for killer, it is all about getting 3 or 4 kills to count the game as a win. 2 kills is considered a draw, and 1 or 0 kills are considered a loss.
The curiosity stems from whether or not this is the truth behind MMR gains / losses.0 -
The curiosity stems from whether or not this is the truth behind MMR gains / losses.
Yeah, I'm aware of the consensus, but there have been lot of conflicting reports/speculation about that, and a lot of people wonder (justifiably) what is actually the case.
Honestly, I kind of hope there's something more to it than K/E because that's astoundingly lazy and inadequate, even if all evidence points to it.
0 -
No way, I'm going to go ahead and invalidate your entire experience and tell you that it's extremely rare to run into a 4 man SWF like you'll almost never play against them so it's something they don't even need to bother balancing for
0 -
Honestly, I kind of hope there's something more to it than K/E because that's astoundingly lazy and inadequate, even if all evidence points to it.
You and me both. I certainly hope there is more to MMR than just the simple kills/escapes.
0 -
In a normal PvP game, I would not be so concerned, but in this DbD game, the time and quality of matchmaking is so low that I am not sure how the game is structured. But in this DbD game, the time and quality of matchmaking is so low that you can't help but wonder how it works and where you stand in the whole game.
0
