Rare W for BHVR and the matchmaking tests
I thought I was reading a bait post for a second because this is just so correct? I'm not used to matchmaking updates that have factual information including statements that have clearly been gathered from the community.
idk who managed the MM tests and who gathered feedback but they've done a hell of a good job understanding what the community experienced and in turn what the community wants.
I mean, I can let the hockey analogy slide now tbh.
Comments
-
Idk I just don't think it solves anything respectfully sure they believe they found a balance but the one thing I see missing from this equation is people who go out of their way to make the game unfun, for either side, this hasn't changed nor been addressed. It's indeed the elephant in the room. All this does is burn out even more players, so now it's not only, High MMR who are tired of MMR in the game.
8 -
It changes nothing.
Matches will still be awful for anyone outside of mid tier and they are still using the most basic of basic to determine mmr where nothing you do matters no matter the outcome if you die and encourages bad gameplay from both sides.
Pump gens and avoid the killer on the survivor side.
Get kills no matter what on the killer side. (camp, tunnel, play nurse.)
10 -
That's true but I think that is more to do with in game mechanics that need adjusting rather than MMR. Hopefully a more relaxed MMR system incentivises people to bring more chill perks and we can see less double range nurses or 2 CoH + 4 DH.
2 -
I'd hope so but something tells me that's just not gonna happen I fear even the mid skill players will convert to the meta tbh.
0 -
They didn't even show the kill rates from test 2 after admitting they dropped.
0 -
I'm curious if they just gave up on bidirectional matchmaking. I looked back at my comments and it seems I had the most fairly balanced matches on days when bidirectional matchmaking was enabled. It seems like having bidirectional matchmaking be forced on for some time until it times out would help prevent someone being pulled into an unbalanced match when they haven't waited a long time.
0 -
I had a distinct feeling on some days that I was being shipped 'up the curve' to slake the appetites of the 10,000k hour SWF squad type players who, without upper-middle folks like me, would never see matches. Those matches aren't really my favorite, but if lowering the MMR ceiling and creating a sort of "Thunderdome" area in the MMR will give them more matches it's less likely those matches will be against me, so that's a win!
I'm pretty happy with these changes overall. I think the most egregious thing they needed to train the system to avoid is pairing new players vs. people with 5x-10x their playtime. Getting viciously stomped early in your DBD career is an easy way to never play again, and we need to make the game welcoming to them!
TBD after I play with the changes, but if it truly does allow for more variety I'll be overjoyed to bring sillier tools and less meta killers without feeling doomed to only getting players who don't share my sense of humor.
1 -
The big winners of those changes are low mmr players, who will only be matched with each other, which is totally fair, and high mmr killers, who will get to terrorize mid mmr players so they can have "laid back" games. The losers are mid mmr players, who will have to go into unwinnable trials against high mmr players.
7 -
Actually not. If you read the feedback for every single day, or better yet, have played during the MMR test days, you'll notice that the changes made the SBMM system even more of a wretched torture than it already was. Now they are just going to permanently muck it up.
1 -
This post was already running long as is, so we couldn't add full details on everything (a lot of the stats aren't super noteworthy). But that said, I don't believe we talked about kill rates at all in test 2, unless I'm mistaken. Do you mean test 1? If so, the top end climb to about 70% to give you an idea. We saw a difference of as much as 10% depending on the skill range.
4 -
Is there something I can clear up here? The folks in the middle should actually see an improvement with the maximum range being reduced; the worst possible cases won't match you with Killers nearly as far away anymore. The worst case scenario is fairly unlikely, though, as the majority of people are in the middle of the pack, meaning that they're unlikely to end up waiting a long time for a match (unless they're playing in off-peak hours).
To give some perspective, on average, matches will have far below 50 points of difference. Even at 50 points, the chances of winning/losing don't change significantly.
7 -
L + Ratio
2 -
For legal reasons this is a joke 🙂
0 -
Maybe you can. From my understanding, lowering the cap means that the absolute best dbd player ever is considered to be in the same mmr group as an average player that just barely reaches the cap. Is that correct?
0 -
Not quite: That would be the case if we lowered the cap too far, but in reality, it's still not anywhere near the middle. We nudged it down a little bit. This basically pairs the insanely good players with the really good players, not with the average folks.
10 -
Oh, that's even more exciting for me! I might occasionally play consistently well enough to crawl up to that "really good players" point, but with the reduced range I may even be middling enough to fly under the radar of those folks - no malice towards them, but woof! I will never be their equal.
2 -
That's good to hear. Thank you for clearing this up.
1 -
Would it be possible to get a clarification on what insanely good and really good players are, and then where we are in relation to them. I know that there are people who are not keen on the idea of showing what player's ranking are, but I honestly think it would help me out a lot. I will have a game where I dominate, and then the next game get dominated, I would really like to be able to compare the rankings of the people I went against. I would also like to see my own ranking, and be able to know if I was actually getting better or if I am getting worse, but also being paired with worse player, if that makes sense.
3 -
Are there statistics for the high end survivors? You told us the top end of killers climb to about 70%, but what about the top end of survivors? Were they at a 70% escape rate?
1 -
The escape rates are the opposite of kill rates. For example, if the average kill rate is 70% in the absolute highest end, the escape rate is 30%.
4 -
While I’m still not a fan of MMR as a whole the fact that they are willing to do tests like this is encouraging to the future of such a system
2 -
It's odd to me that they found two very distinct groups of people....yet the solution isn't doing what many other Multiplayer games do and just have a Ranked and Non-Ranked matchmaking Q?
Even the more competitive streamer people don't always wanna go hard 100% of the time, sometime you just want to meme about and use fun builds. So having two different places to q would be good? And then you run into a SWF on comms with all meta perks that aren't having any of it. (or vice-versa with some fun survs who run into a competative killer).
Like....why hasn't a ranked/unranked system been tried? Give rewards/incentives/higher BP rewards for the ranked system to reward people for serious play and reduced for just messing about to possibly discourage non-ranked bullying. And yes....I know I KNOW before I get a bunch of replies that there will still be people who "sweat" or do non-fun things in an unranked environment to get easier games....but it should significantly reduce the amount of times people trying to have fun run into competitive people compared to the current system..not eliminate it entirely.
And if this has been discussed at length elsewhere, and I should go look there, I apologize. Just my thought on the current subject.
0 -
Sorry about that; I didn’t realize this was talking about average rates across an MMR range. I thought it was more like “the killers with the highest kill rates averaged to 70%”.
The reason I asked was because that day I saw some killer streamers that had a super inflated kill rate, but it also seemed like some of the SWF streamers had a super inflated escape rate.
0 -
I don't feel the conclusion really touches on the core of the problem.
As survivor in particular, the current system thinks I'm very low-skilled as I don't play selfish. I have spent quite a bit of time stuck in low-MMR-hell, because I die a lot with my playstyle - I take chases, I try to save in EGC, try to cleanse NOED and save the last person, even if everyone else has left etc. Sometimes it pays off but I also die often, especially so as I get paired with people that do not know to bodyblock when rescued with BT near an open exit gate. 😭 I also find that I have been dying on first hook more than I used to before the skill-based system was introduced as I'm basically stuck in a loop where I'm the only one running the killer but eventually they will catch up with me, my teammates don't pop gens and are too scared to come and rescue. So my MMR goes down.
Now the system will supposedly be tighter, so I will be further down that spiral and I don't see how this conclusion will improve the experience for people that play altruistic in solo queue and hence die/MMR goes down.
2 -
To put it simply, a ranked & unranked mode is not a magical solution, but it is something that has been discussed numerous times.
For starters, like you've mentioned, you're putting faith in people using them responsibly. If someone who's really good at the game just wants to flex on newer players and get easy matches, nothing's stopping them from playing unranked.
Second, the game needs to be able to support both modes. It could absolutely handle two queues, but splitting players between two queues also limits what the matchmaker has to work with. Say you're playing ranked because you want close matches, but some of the people around your rating are now playing the casual mode. Now the matchmaker has to look further to find you a match. You'd still get a match, but it wouldn't be as close, so the point of the ranked mode becomes less effective.
Third, for offering incentives to people for playing ranked, I'm not completely sold on this. If we're talking things like bloodpoints, then the newer players (who would probably want to play casual) would arguably need those more since they're just starting out, leading them to play ranked when they really wouldn't want to, nor should. If we're talking cosmetic rewards, that's less of a concern, though some people will undoubtedly suffer through it anyway for the rewards, as you often see in other games.
And last but not least, the dirty secret of unranked modes is that they are usually secretly ranked. Most games that have an unranked mode are still tracking your skill in some way, they just don't show it to you. This is why your super casual friend can play a game without any issue, but suddenly when they play with you, they have a really rough time. They're getting pulled up to your (secretly) tougher matches.
But all that's not to say it's an awful idea. Even just giving people the option to separate themselves based on their mindset (casual, just wanna mess around vs. competitive, I want a close match) can be a good thing. It just wouldn't necessarily solve everything and could create other issues.
10 -
Nothing will change basically as long as it stays in the game
3 -
People say this often but the system doesn't evaluate skill. It simply measures survival. It should be Survival based matchmaking.
0 -
When are the changes coming in?
0 -
Survivor skill is a tricky subject for a lot of people. Some would say chase is the only survivor skill there is. And then you see survivors at the absolute highest levels of play who hold W and pre-drop every pallet. Their positioning and game sense is immaculate, but they don't really "loop" anything unless the situation calls for it.
So then the problem becomes: how are you grading survivor skill if not by escapes? If you weight chase too heavily, you end up devaluing extremely important things like objective efficiency and game sense. BP? That would heavily favor altruism for the sake of altruism. But a consistently high escape rate is usually going to involve a mix of good chases, good decision-making, and efficiency. It's simple, but it works in most cases.
1 -
The only thing left to do now is to stop incentevising face camping for killers :) Literally had 5(!) games where I was placed on the hook and the killer never moved further than a few meters from me. How am I supposed to shoot goals in hockey if the goalkeeper is literally sitting on top of me?
2 -
I am glad to hear that grades in the end game scoreboard will be removed. Hopefully then players will start understanding they don't directly corollate to skill or your SBMMR.
and
"These include making Survivor ratings team based rather than personal (so dying heroically to save everyone else isn’t as punishing) and improvements to backfilling lobbies to prevent players from outside of the acceptable range from being pulled in."
I hope this line means that SBMMR will start to be determined based on how well you perform in game, rather than just Kills/Escapes.
3 -
Oh, yes, you are absolutely right. However, that not what BHVR are calling it and as long as it's supposed to be skill-based, the escape factor is the major flaw in the system. That's what I was trying to illustrate. I also don't feel like I'm alone in being a "semi-skilled survivor" (not saying I'm a god looper) stuck in MMR hell. I seem to see quite a few comments like this and I remember ScottJund even made a video on this and it really reflected my experiences so well.
Ironically, the current system does work better for me on the killer-side and I am going to assume my killer MMR is significantly higher than my survivor MMR, whilst try to go for 12 hooks/11 hooks+hatch (ideal-case scenario). I'm better on some killers but I seemed to average two kills across the board when I recorded it for a while in the beginning of this year.
0 -
I honestly think you guys get your hopes up way too quickly. I can't really blame you though as we are starved of transparency and information most of the year.
2 -
These are my games in a nutshell as I too, do not have a selfish playstyle.
0 -
All fair points!
Thank you very very much for taking the time to reply to that! <3
Either way, I hope the solutions y'all come up with work out! I know SBMM is the big hated on subject at the moment and many have touted as the reason behind the large hemorrhage of player base. So the fact that y'all are testing, changing, and taking the time to talk to us is a BIG sign of hope for me at least. Love this game and I wish the utmost success, even when I get frustrated with it! Lol. Good luck, and see you in the fog.
0 -
Lowering the soft cap was the opposite of what should be done, I don't know why you're happy about this unless you enjoy stacked 4man SWF games or meta Nurses.
The goal seems clear, protect brand new players and make sure the bully squads don't have to wait in queue too long.
0 -
One thing I did not see mentioned was how the MMR tests handled groups. Did groups experience a higher than normal wait time on some tests but not others? Did games with SWF members in them experience a significantly higher-than-normal victory rate on some tests more than others? Were SWF groups more common in the upper end of the MMR curve than the lower?
I ask because I'm not entirely sure how SBMM actually works for groups. If one person has a top-end score and their partner a bottom-end score, are those just averaged? Is the highest value used? Are any secondary factors considered, like the number of people in the SWF team?
0 -
As soon as you brought up hockey , you guaranteed that you would not get a response from BHVR. lol
0 -
Eh, whenever logic is based on a fallacy, I become very skeptical about it. But honestly, it doesn't matter to me, I was never one to whine about mmr. And I didn't really complain about how rigged the game was for new players when I joined and got matched against red rank swfs either.
0 -
You could've, I promise we don't mind it being longer if we get to see actual statistics
0 -
I dont really think it'll change anything. The problem is people in the high end that all use meta perks and try their best to win. That's why there's no variety in matches, it's not just getting different people but wanting to see different perks and the meta is so much stronger then everything else that you feel the need to use them to have any chance at an enjoyable experience
1 -
But meta always will be there, if you change anything the next strongest takes its place. Meta exists no matter what
1 -
Yes a meta will always be there but the problem is the meta is too good in this game. That's why most of the same perks have been meta for years at this point and why there is so little variation. With each side having maybe 1 or 2 perks that can be swapped out for depending on the killer you are playing or the map you send yourself too. Everything else just makes you feel like your intentionally crippling yourself just to play something different. Plus having the same meta for so long is another problem and makes it feel all the more stale. It's okay and good for a games lifespan to absolutely murder some perks and buff other perks to the moon to intentionally change the meta every once in awhile but bhvr is too scared to do it for some reason.
0 -
I wouldnt like the change, I love dbd the way it is now
0 -
Nice to see you without the bars 😏
0 -
I loved the post BHVR did today. This is what I'd like to see more of moving forward. Personally, I see good things coming out of these tests and I hope they continue making tests to make this game more fun for newer players. As a veteran, I'm already locked in the game, but they need to retain new players so making it more fair for them (both sides) will hopefully increase player retention.
One thing I hope they address soon is the grind. That's one of the things new players complain about and rightfully so. It's a huge time sink.
1 -
Okay well then we've come to an impasse cause I literally can't play the game anymore.
0 -
So, the question is, will I be able to actually play the game as Killer and Solo Q Survivor?
I don't much enjoy feeling stressed out of my mind as Killer, nor do I enjoy being tunneled/face camped as Survivor.
0 -
No
0 -
Then I continue as planned.
0
