SBMM and SWF.
I get it, you're showing the game to a friend and there is nothing wrong with that. But why do I have to deal with your 10 hour friend in my lobby when I'm in the thousands of hours?
More often than not, everyone is above 500 hours, even above 1000 (including the killer) and there's a survivor with 10-20 hours that someone decided to bring along, making it an obvious 3v1.
Why is SWF once against bypassing the ranking/SBMM system and what can be done about this?
Comments
-
What can be done? Well, nothing. Most games have issue how to handle this type of teams.
From what I know dbd just makes it average from SWF MMRs, which is probably best you can get.
You can't go highest they would get most likely destroyed and you can't get lowest, because killer would get destroyed.
0 -
So instead of the killer getting destroyed, I get destroyed because I'm in a 3v1. Fair, not.
0 -
To be fair, that could still happen even if they had hours. Team based games, man. Your only choice is to play killer and not deal with bad teammates.
10 -
In other words. New comer gets boosted, veteran gets boosted (since he is better looper then a killer in chases). Everybody is eventually high mmr.
0 -
You really want to tell me it would be better if matchmaking took lowest MMR from SWF?
I hope not...
Get 3 over 1k survivors as first game would have to feel nice for sure.
Average is just best middle ground. None of those options are really fair, but it's best we can get, unless you want to remove SWF option, which good luck with that.
1 -
yes it can still happen, but at least those people know how to hit skill checks while pressing m1.
2 -
How does veteran gets boosted by being a 3v1?
2 -
I never said it would be better. With the trash SBMM we have right now there is no "better" anything really.
I have no idea what you mean by the 3 over 1 sentence.
0 -
What you could do is take 3 veteran players and play with 1 new player. If matchmaking was by lowest MMR, you would put those 3 survivors against completely new killer, who would get super destroyed. They would either bully him, or just finish gens fast and game would be boring for both sides.
1 -
I never said in my post I want veteran survivors to go against new killers. I said I dont want survivors in my team with 10 hours when I have 5k hours and the killer is also above 1k hours.
0 -
I just told you three options how system can handle it. You didn't like how average option works, so I just gave up an example what would happen if they tried lowest MMR option.
Unless you want to remove SWF completely or limit who can join that group based on MMR / playtime. But this just can't be fixed by SBMM, or any matchmaking system.
1 -
Me having a 10 hour team mate, when I have 5k hours and the killer is also in the thousands of hours, isn't an average option.
Again, I never said I want to remove SWF.
0 -
You asked what can be done about it, so I am saying options. If you have better option, then feel free to share...
It's not about playtime, it's about MMR...
It's nice that you have 5k hours, but a friend of that noob could have higher MMR than you.
0 -
Playtime has a big thing about it tho, or are you telling me 10 hours is comparable with 5k in terms of skill?
0 -
it is regarding the SWFs and balance around mmr.
If you are in soloQ and have a chance to see his hours. You can dodge the lobby.
There is no proper solution then disabling SWFing with vast difference in MMR.
And yes. I dont underastand half of soloQ. While lockers and crouching is a main thing next to doing gens 10m away from you being on death hook. Beauty of dbd
0 -
SBMM doesn't give a ######### about your playtime. Only thing it cares about is MMR and MMR cares only about escapes.
If that 10 hours player played only with his better friends, then his MMR most likely boosted af. Then there is chance his friends just have higher MMR than you do, if they play SWF only. That's how you got there.
So you can suggest better MMR system how it calculates player's MMR, but SBMM for SWF should still be average.
And if I have learnt something during my time of gaming, then it is playtime does not mean skill. Your progress in skill is not that big after like hundred hours. It may be for games where you have to learn a lot of things, but I don't think that is dbd.
So if you have new player that got coached by his better friends and / or took his time to watch good streamers, then he could be quite decent at 10 hours easily. DBD is not that hard game imo.
1 -
Simple solution: remove SWF
On a more serious note, its better to drag the new player to the higher mmr than the other way around. Just imagine all the SWF squads that take a fresh player into their team to bully new killers.
6 -
With an ideal skill-based matchmaking system, numbers of hours wouldn’t matter. Someone that can press w and m1 at the right times, loop well, and has the “right” perks is going to hold their own regardless of time played. There is the problem that it takes a certain number of hours to get the “right” perks, but an ideal matchmaking system would evaluate a build and adjust accordingly. This kind of build adjustment would also be helpful for people that prestige.
The only solution I can think of when you have a large gap in MMR on the survivor side would be to take a weighted average that heavily weighs the low outlier.
0 -
If you factor in time played to matchmaking, then veterans get longer queue times. Idk what could be done. Plenty of other games have this issue as well.
0 -
Technically the best is probably something like using a neural net to look at games involving swfs and various sets of MMRs to calculate what net rating swfs have if they were two solo players with identical ratings. It might not be a simple linear relationship or average of the ratings but rather something for instance where an MMR gap of a certain size in a swf has the most significance but then beyond that even larger gaps don’t matter as much. Or that the difference means more or less depending on how low or high the lowest or highest MMR player in the swf is.
Averages are the simplest to program but probably not all that accurate an estimate.
0 -
I think this is always going to be a difficult issue....I do think the average MMR is probably the fairest of all the options available. Other survivors who are not in the group do still have the option to dodge the lobby if they don't wish to play that match.
Hours do correlate to skill to a degree...but in saying that, I don't think the 1000's of hours matter, as that's very person dependent. Some people with 3k hours are still not great at the game and they never will be - we all don't have to be great at everything. Some players with 50hrs are pretty damn good and can beat players with 1000's of hours, their understanding of the game might not be the same, but their actual skill can be comparable. (which is why hours are not a good indicator of skill - someone could have 1k hours in 2021 for example and someone else could have 1k hours since 2016, there's definitely an argument there that the player with the most recently accumulated hours will be more in-touch with what is currently happening in the game, hence we're looking at introducing the "decaying MMR").
9 -
Your version would work for normal 5v5 games, where you care about how big is difference, you don't want top 1% player end against two average players.
But in DBD you can compare it only with 1 MMR of a killer.
When you have 1 really high MMR and 1 really low MMR, there is not much to do here, probably best is to make an average and find two survivors and a killer around that value.
EDIT: You mean something different than I thought at first. You would actually want to track multiple games and results involving similar matchmaking and adjust optimal value for that game to have most balance results. That would be hard af to do.
0 -
Actually you can hypothetically look at all the games in the data, plug in all five player’s MMRs into a neural net learning algorithm along with a true/false input on each survivor being in a swf, and have it train itself to guess the outcome of the match in terms of escapes or kills. This should work since it’s reasonable to assume that very small differences in MMR probably correlate to very small differences in skill on average. And it would probably give better estimates of what the optimal rating for a killer should be against a given lobby of survivors since it’s unlikely that a simple mathematical average of the MMRs of the survivors is the actual best estimate.
0 -
Their result of SBMM is MMR based only on escape / kill. I think you want too much from them :D
0 -
I think the best way to go about this is to make it different for solo and SWF players to boost and drop their MMR, giving them different requirements.
People will escape more if they're in a SWF, doesn't mean they're better at the game, it just means they are at an advantage compared to solo players.
Saying someone will never be great at the game even if they play a lot, I dont think is very accurate. I know BHVR has that mentality when it comes to it, since that's why they nerf a lot of perks, because they are too hard for new players (a good example of this being the ruin nerf, since one of the reasons it was nerfed was because new players couldnt hit great skill checks.).
2 -
That's one of the issues. SWF will often have more escapes because they play for each other. This is one of the reasons as someone who plays solo, why I dodge obvious SWF in my team, because a lot of times they will just escape and leave the solo behind. Does it mean they were better during the match and more skillful? No, it can mean they played more selfishly. Which again, is one of the reasons I hate that you can be a really good during the match, but if you die and the blendette who did nothing all game escapes, the game sees that claudette as more skillful than you.
The system only counting escapes as skill, just once again boosts SWF players. SBMM should look at everything done during the game, not just at the outcome.
0 -
Hey sure, I have nothing against that.
I think killer's way of getting MMR by kills is best we can get, because that is hard with so many killers and different playstyles.
But MMR was survivors could be improved for sure.
0 -
It’s more accurate to say the MMR goes up or down based on escapes or kills, but how much it changes depends on other factors like the relative ratings of the opponents and the order the players died (i.e. the first player killed changes MMR more than the second because once one player dies it increases the likelihood of more players dying.)
Anyway, I don’t expect them to program a neural net for matchmaking if they don’t want to, I’m just saying it’s possible in principle.
0 -
from my own experience with these kind of matches, the newbie gets tunneled out in 2 minutes and the killer wins. The only option you have is to hide the newbie as good as you can and take aggro and hope he is doing gens...
1 -
Average MMR is not the fairest for the Killer, SWF gamed the last system you had in this way and they'll continue to do it with the new system. You are forcing the Killer to play against higher MMR survivors, when it should be the low MMR Survivor volunteering themselves to play in a high MMR match.
1 -
SBMM is never going to function the way that we all want it to. Perfect example: Otz was training a new player on how to play killer, and within just four games this new player was facing off against people that had thousands of hours. It's ridiculous.
1 -
I disagree. A leatherface who facecamps can get a 4k, while a leatherface who doesnt camp can hook everyone twice and not get a single kill. Which one would you consider more skillful?
I think how a killer plays during the match should also matter more than the end result, and hooks should count more than kills.
3 -
This is my experience as well, that's why I said having a 10 hour survivor on the team is like being in a 3v1.
0 -
So if I managed to slug everyone as Oni with my power, I should lose MMR for it? Even it's not really easy to do, or am I supposed to be forced by system to farm them?
Kill is same for every killer. Doesn't matter how you got it. Unless you want to make specific way of how to calculate MMR for each killer, then you would always have killers that would be punished by your system.
The Bubba that got everyone on deadhook, chosed to do it, or he didn't remember how many times he hooked which survivors, which is not really high skill...
0 -
Sadly, this is an issue in most games of this nature and it's not something that can really be fixed. Eventually the MMR of the 'high' player will decay.
It would have been the exact same under ranks.
1 -
I could say if you were able to slug all 4 survivors at 5 gens, they were probably below average. Is killing below average survivors skillful?
Because I'm pretty sure you dont see that kind of thing happening with better survivors.
1 -
Yeah, but if your system were based on hooks, I would stay there and were able to do it over and over again...
And also if you have Bubba that camps from first gen, what do you think is better result:
give him lower MMR, so he stomps noobs again forever, or give him higher MMR so he eventually gets against survivors, who can handle it?
This has nothing to do with how that player got those kills, he got them and if you lower his MMR it's going to work for him forever.
0 -
A facecamping bubba would stop stomping noobs and then just kill veterans instead (you cant save against face camping bubba) and make their MMR get even lower so they can play with noobs instead? No thanks
0 -
Since when it is impossible to win against facecamping bubba? It's 100% possible for decent SWF groups without weak links. 2 escapes most of the times, 3 still possible.
But because you don't want to face them, you would just let them masturbate on new players. wow...
And so what if he gains MMR by it? In both cases you will not have to face him, so good for you and games will get harder instead of easier for him, how is that bad result?
0 -
@MandyTalk I think decaying MMR is just invitation to purposely lower MMR to face easier players. I can definately see players playing dbd for 1 week than not playing for 4 weeks so that their MMR goes lower. It was like that before with old rank system resets. decaying MMR silly concept because your skill doesn't degrade that quickly for one and MMR itself always punts you back down if the matches are too difficult for the player or brings you up rapidly If you beat higher MMR opponenents than you. Its self-correcting system. Decay is just another way to mess with matchmaking. MMR did not exist before 2021 so I am not sure why 2016 is being used as beach mark year.
As for hour played. It ultimately depends if the player is competitive or casual. Casual players have a lot of hours in the game but do not actively look to improve. they just play the game for fun. competitive players use the hours to get better at the game, so they're typically at highest rank because they play everyday to up their game. hour played =/ skill, Its just correlating factor to towards skill partly because hour played increases knowledge of the player and knowledge can help improve the player's game.
0 -
The core of the problem is, the game is 100% not intended to be played as a killer can only have to commit to 1 person and 1 hook to then be able to take advantage of 2-4 kills.
The game is supposed to be designed and played to involve a killer chasing survivors not getting only 1 hook out of 12 or 4 out of 12.
It is just bad killer/mechanic design which totally bypasses the point of hook states and loops and resources and a map to use.
The problem is, bubba exists, face camping at 5 gens exists and there isn't an easy answer or solution for them to come up with to deal with it.
I'm not arguing, valid strategies, whether it's fair or anything - it's irrelevant. If the game is meant to be played that way, just make 1 hook = sacrifice. The other survivors are not playing any form of game anyway sitting without being chased doing gens only with 0 interaction.
The gameplay style is however totally unhealthy for the game, for killer and survivors skills to be improved. - Just like many map RNG's are unhealthy - you shouldn't be getting 2 jungle gyms into shack and things like that either - bad design.
0 -
Not everyone plays SWF, and "2 escapes most of the time" isn't a win if you're the one who died on the hook. And I dont see how pressing m1 all game when the killer is afk in front of the hook, either, but I guess that's your preferable "masturbating" method.
0 -
And what do you want them to do? It's either that or putting a 10 hour killer against a SWF with thousands of hours. If you don't want to deal with that, you can always check profiles and lobby dodge.
0 -
I have never said I am doing it, but it's funny to see how you react when you run out of arguments...
0 -
Did you miss the part of the argument where I replied and said "not everyone is swf"? Probably. Because if all your arguments are about SWF.... you're leaving everyone else who isnt behind.
0 -
I shouldnt have to dodge lobbies just because the SBMM system isnt great, I already dodge them, still dont think it improves anything.
0 -
Your post started about SWF, no ######### I am going to talk about them.
Your only argument was crying that you don't like to play against it (camping), or crying that you don't like SWF playing with new players.
0 -
If you see from my post aswell, you'd notice I am not SWF. So, why do you care about the SWF experience, on a post from a solo survivor? I'm talking about solo survivors experience.
Can you quote me where i am "crying" or are you just having a bad day and things seem more frustrating to you than they really are?
0 -
That's just what I feel from your post and comments.
No arguments, no ideas how to fix it, you just complain, which I consider crying -> just nothing constructive...
0 -
I literally mentioned multiple times ways to improve SBMM, like not making it about escaping/kills, making it about what you've done during the whole match, not the outcome.
You havent added any solutions either, so how are you being constructive? And no, using your "feelings" isn't being constructive.
0
