Visit the Kill Switch Master List for more information on these and other current known issues: https://forums.bhvr.com/dead-by-daylight/kb/articles/299-kill-switch-master-list
We encourage you to be as honest as possible in letting us know how you feel about the game. The information and answers provided are anonymous, not shared with any third-party, and will not be used for purposes other than survey analysis.
Access the survey HERE!
Should MMR Factor in Devotion Level?
Since it roughly corresponds to play time, would it be beneficial to attempt to match players with similar devotion levels? Especially since it doesn't really do anything else right now.
Comments
-
Maybe hypothetically if the MMR is trying to assign an initial rating to you for a brand new killer, like say they want to guess how good you might be playing Blight when he goes live, then taking your playtime as killer or Devotion level (which is basically the same thing) could be a reasonable factor. But once you've already played a number of games it really doesn't need to see your play time at that point. Plus Devotion level isn't going to tell you too much for killers that have unique playstyles. Me being good with basic 115% melee chases doesn't help me very much when I play Nurse and tells nothing about how good my aim might be with Deathslinger or Huntress for instance.
2 -
I don't think playtime should be relevant to matchmaking unless you're an actual baby to the point where matchmaking hasn't had a chance to assess your skill. I agree it'd be cool to have devotions do something for you though... like maybe they could give you a big shards and/or BP drop every time you get to a new devotion.
1 -
This is the mistake they made when MMR first came out. They proclaimed that rank or playtime or whatever didn't matter when it obviously does in some form or fashion. Using something like devotion or rank as a baseline should have be a no-brainer.
0 -
The problem is high devotion players may suffer much longer queue times if it's made an important factor in MMR.
1 -
That's understandable. If it were used as a fallback for when the system had no data on a player's performance as a new killer, would that be agreeable? Of course each killer has unique and game-changing powers, but if you've been playing the game to Devotion 5000 you will know how to run chases and when to patrol gens better than a new player.
1 -
Not necessary. Although devotion level is already in the game, coding a new MMR system like CS:GO or Dota 2 is very easy to do unless DbD's code is a mess.
0 -
I don't think it should factor in. I have over 3,000 hours and i'm still a ######### killer. That's mostly down to me mainly playing survivor but still, playtime (which is exactly what Devotion is) doesn't mean too much.
1 -
Still saying that for a proper MMR system it would need a complete overhaul of several game mechanics to genuinely allow to measure things against each other. you can't really measure a players skill with e.g. a survivor against a killer as long as survivors can still switch in the lobby. neither can you measure a player's skill with a killer if they play each killer as camping and tunneling, never making use of the killer's abilities.
0 -
Devotion levels should only represent overall time in the game not determine skill
Now if they added a leveling system for survivors and killers they should repersent time on either side you play on would be cool
Like I spent more time playing killer then playing survivor and have reached level 62 devotion 2 (I think) also have about 700 hours play time
MMR/ rank should represent things that matter during a match:
For killers: Finding, downing, hooking and killing survivors as quickly and effectively as possible while keeping gen porgress to a minimum
For survivors: locating and repairing gen as quickly and effectively as possible while healing, saving, and keeping the killer from disrupting gen progress
Also survivors also (would) have bonuses for having long chases, time spent repairing gens and saving/healing teammates
Also Keys and Moris = gain less but lose more
Anyway that's my take on it...
0 -
If I only got matched with people who had a similart number of hours as me, I'd lose every game.
I wish my actual skill level increased with my devotion level.
0 -
I think probably not. I have a couple of friends who play this game who have higher devotion levels than me (I think one is 9 and one is like 12) and they are not good. One is a spine chill/urban evader who lasts about 10 seconds in a chase and the other usually gets grabbed out of a locker 20 seconds after the start of the match after trying a ridiculous head-on play. And as killer they're even worse.
They also aren't on the forums afaik but if they are, I'm sorry lads
0 -
Playtime is a poor indicator of how well someone plays. Different people have vastly different amounts of effort they're willing to put in to get better at a game. Once a person is past that point, their ability at the game increases very slowly with further playtime.
7 -
Couldn't disagree more.
This game's skill floor and ceiling are much more like an RPG or MOBA than any other kind of game. Information is king, and there's a lot to learn as a new player. Nothing is explained in-game and advanced tactics take a long time to learn and even longer to master. It's not an FPS where someone can jump from playing Counter-Strike to Overwatch and roughly keep that skill level. That's what placement matches are good for. A game like DBD is entirely unique, there's almost no crossover (apart from Deathslinger?).
A player who plays for <20 hours will NEVER be as good as someone who has 100+ hours. You can't say the same for FPS or Fighting games, which is why placements for rank work so well.
As people sometimes say "rank doesn't matter, it's not an indication of skill but rather time played". True, the ranking doesn't match up like other games, but I would say that time played is an indication of skill.
My idea:
Devotion should bracket players off into... brackets. New players should play with new players, and veterans should play with veterans- this was shouted from every rooftop during the MMR fiasco. Yes at a certain point the difference between a 1000 hour player and 1250 hour player isn't as large as 0 to 250, and we don't want people to have super long loading times.. so at a certain point everyone will be in the final bracket.
For example:
0-50 hours | 50-100 hours | 100-200 hours | 200+
Wherever that corresponds to devotion of course, not idle gametime played. If noobies are partied with someone above their bracket, they're going to get a killer that's on the highest bracket.
Only AFTER you get to the final bracket does your MMR kick in. This is when your individual skill will shine and put you with appropriate players. That is, of course, you even want an MMR in the first place.
6 -
Not to mention as well how fast some people learn different skill sets. Natural skill can't be shown in simple playtime.
1 -
I'll just give you one example. I came across a LoL player with 5000 games. He was Bronze 4, which at the time was only one rank above the very bottom. This guy has no business being put in matches any higher than Bronze 4 despite having multiple thousands of hours in the game.
Yes it's only one datapoint, but it illustrates the problem. Playtime cannot be used for matchmaking. Matchmaking is never going to be exact, but playtime just doesn't indicate anything useful about a player in that regard.
4 -
Probably was me :P
3 -
Yeah I'd agree that's one datapoint, and can be explained in a number of ways. My first guess is that they were smurfing to lower their rank. Another is that they had someone else play on or buy their account.
Let's say he's a genuine player- how can someone do something for 5000 hours and not improve at all? I find it backwards to design a system around those anomalies instead of the vast majority of players. Bad players should play against bad players of similar game-experience, the bracket system would ensure this as well as stop smurfing (to an extent). My whole point is that there needs to be a baseline level of knowledge before you can gauge their skill. Most popular games can ignore this as skills transfer over - so anyone who has played an FPS (which is almost everyone) has that baseline already.
What popular games are like Dead By Daylight? It's unique. Therefore every player needs to learn the baseline for the very first time, so it takes a long time before their skill can actually shine.
1 -
Twisted on that one.
I mean, when MMR was active, I went against players with under 1 hour of playtime (with Billy, who I played a lot). So even IF I would be really, really, really bad with that Killer, my 4700 hours would be more than enough to deal with players on their first game. Without any use of the ability, I can probably even catch them with Nurse without blinking.
But also, my high hours could hinder me. Lets say, I want to learn Nurse (I am terrible with her), but my high hours would pair me with stronger Survivors, I would have a harder time than I should have, if MMR is working perfectly. Because, most of my hours on Killer are on Pig and Billy, Nurse is completely different, I would not really get anything from my experience on other Killers, which would be influenced by my playtime.
Let alone that it is entirely possible that someone has a lot of hours as Survivor, but almost never plays Killer and would get matched with way stronger Survivors. And Devotion is not different if you are playing Survivor or Killer, so someone can be on Devotion 10 and only played one role, but would go against Players with similar devotion.
Just imagine someone who only played Survivor but tries goes against Survivors who also only played Survivor, all on Devotion 10 (or vice-versa). It would be a mess.
IF Devotion should be a part of MMR, it has to be very minor.
0 -
While that is a good idea it wont work like ranking its just a number to show how much you play. Baising it off that wont be good because those with a high devotion will never find games or allways go fight the top players and the only thing that keeps some players is running into the new players
0 -
Experience should be considered when rating a players skill level but only to some point. Experience alone doesn't make you a good player and there are many streamers that prove this but an average player (skillwise) with 1000 hours in most cases will be better than an average player with 100 hours.
It doesn't matter if you have 2000 hours or 4000 hours but it matters a lot if you have 20 hours or 500 hours.
2 -
On a different topic since you’re taking an interest here, any general word on the new MMR? Was it an issue with the initial seeding or how certain killers were rated or something more involved? I’m assuming it’s still being worked on but I’m just guessing on that and the possible reason for the problem obviously. “No comment” or “still working on it” is fine by me, I’m just curious, MMR systems are kind of fascinating to be honest. 🙂
1 -
They should ditch MMR, this game ain't built for it.
There's a huge difference between SWF and solo,(solo and SWF should have separate matchmaking anyway) there's a million and one bugs that still need fixing. This game its about as balanced as a McDonald's diet.
MMR will push SWF to the top and encourage toxic and scummy play from both sides.
0 -
The OP asked if playtime should be a "Factor" in the calculation.
Time = Experience regardless of whether the player is actively trying to get better at a game. A casual player with 500 hours is going to be more skilled than a new player with 5 hours in.
Those two players shouldn't be put into a match together for any reason which is why Time Played should be a FACTOR in the calculation.
Clearly, time played isn't the only thing that should be considered in an MMR calculation, but shouldn't it be part of it?
1 -
It's not a reliable indicator of performance, so it's not useful for MMR. MMR systems make mathematical predictions about the outcome of a match based on the incoming players' MMRs. Something as loosely tied to performance as play time doesn't help.
0 -
Don't you think that having rank or devotion even as a baseline to start off at is better than starting everyone off at zero? Because having 4000 hr killers slaughtering baby survivors was a good idea, right? Because time played somehow doesn't translate to skill in this game where the learning curve looks like 2020's unemployment graph?
0 -
MMR systems deal with new players in the system by allowing their rating to change swiftly during early games. So what you're talking about is already built into the system. I guess you could say that part of the system takes into account play time, but only relative to the number of games since MMR got turned on. It doesn't care about anything from before that time because the number of games it takes for MMR to settle is small.
It's a huge complicated subject. I recommend reading about these systems as they're actually really interesting. Like what happens to such a system when there aren't new players coming in. What if there are a ton of new players? Can someone abuse the system? In the case of that last one, the answer is YES. https://en.wikipedia.org/wiki/Cheating_in_chess#Rating_manipulation
0 -
So you really think it's OK that brand new players will be facing 4000 hour veterans? As a gauge of skill? I'm just baffled by the idea that you think someone can pick up DBD today and by tomorrow be as good as someone who played since January.
0 -
So what did your data say about the outcome of a match when "red rank" players (players with more time in the game) went up against "Rank 20" players (players with little to no time in the game)?
I ask, because when the MMR system went live many players reported going up against players with less time in the game than themselves and that period of time would be a good gauge of whether time should be a factor in the calculation or not.
0 -
It's not what they said at all. The fact of the matter is it's an unfortunate consequence of the system. However- it's designed to get you placed back into where you need to be quicker.
For example- I was a rank 20 killer after taking a 5 month break. I slaughtered like my first 10 games and was very quickly purple ranks. It's alot easier to pip and rank up between rank 20-10 then 10-1.
0 -
No, and here's why. I've been playing nothing but 2500+ hour survivors since last year, when I only had 100 hours. Because I'm insane and continued to play against those kind, I've reached a killer level on par with them.
If you were to put me against average survivors with my own devotion level, it would be a slaughterhouse. Wouldnt be fun for me or them.
0 -
But why does it need to be? Given your example, a bracket system would work great. You are a veteran player that knows the game, but given the time lost, you've dropped in your MMR. That's no problem, because you haven't forgotten everything. You're matched with the lowest ranked players at Devotion X and if you're a god player then your MMR will increase as it should. But at no point are you smashing brand new survivors who don't know what a pallet is.
In this new system, you're just where you're at now. Currently you were at rank 20 but you're not a rank 20 player. With the new system, you're at 0 MMR but you're not a 0 MMR player. You'll be matched with random MMR players to get a baseline of your skill. That might work out OK given that you know the game, but what about the brand new 0 MMR players?
They'll be completely lost, with the system throwing them into games with veterans who will stomp them. How many games will it take to "sort it out" and just get them to play other noobies? Who knows. How many games like that before they quit?
0 -
Yeah. At the very least, low playtime newbies should only be placed with other low playtime newbies, maybe raising the playtime max if the wait goes too long, but still keeping it to low playtime. Other than that, I can't think of a single good reason to do anything else. If their MMR skyrockets due to natural talent or being a player who was at for a while or something, then move them up into non-noob MMR.
The big issue with the first attempt at MMR, as I've understood it, was that not only was it failing to do what it was meant to, it was doing exactly the opposite, creating mismatches that displeased newbies and veterans alike.
Playtime, at least generally low playtime vs generally high, should be factored in to avoid this sort of thing as much as possible.
1
