MMR Test experience summed up from a 6000hr player
Hi, I'm a super sweaty player and thought I'd come up and leave my take on the new MMR system.
After reading this please leave me your feelings and experiences, I'd love to hear them.
I think it's fair to say that things aren't quite accurate for the most part, however there were a few games that felt like they were correct and I wanted to talk about it.
I split my time 50/50 on killer and survivor to get a more balanced opinion.
In my survivor games, they started out as expected, sweaty Midwich nurses. All in all, really great and I was really happy to see good killers. I lost one by dying in the exit gate and won the other. However, after these games began the descent to madness. Constant rank 10s and I even got a dude trying to get his adept. I'm unsure as to why this happened and I'll talk about my ideas at the end.
In my killer games, it was just bad. I started with rank 1s but they weren't great, and then I started getting purple ranks after that. I basically ended my games instantly and the survivors got 5k pts. Really boring and made me really nervous about their system.
So my current idea, is that perhaps people with too high of MMR are being treated as outliers and being given whatever is available. Another option could be that low rank players who haven't played many games, but have won the few games they've played, are being treated like they have high win rates and being paired with higher MMR players.
So my suggestion overall, is that they need to disclose what the MMR scores of players are, so that we can see visually if the system is working correctly. I'm unsure if they're scoring us incorrectly or if it's matching us incorrectly, and we won't be able to tell if it's not disclosed.
I'm aware their opinion has been to not be transparent in fears of manipulation, but quite frankly people who derank will be an issue regardless of whatever matchmaking system we have. So rather than focus on the few people we can report, we should maybe focus on making the majority of players have accurate games that we can verify.
As verification of my claims I'm leaving my stats and the final screenshot of my last game.
Comments
-
I think it would be nice to see the MMR scores, at least compared to those of the same role so people can see where they stand against other players. I mean, that is what most rank games are about. Bragging rights to who has a higher MMR. I don't think it matters for the test though. I'm sure they have the ability to look at the matches that were created and played in the back end to tell what type of variance there is during the test.
My solo survivor games so far, I feel on average, I have been given better team mates, but the match to killers was usually at the extremes. Either my team all escaped or the killer destroyed us. Only 1 or 2 that were close and maybe could have gone a little different.
My killer games so far, I have been dominate in all of them. The only escape against me so far has been a hatch escape.
0 -
I honestly don't care about bragging, my main issue is that I can't tell if the system actually works. Like good on them if they can see, but if I can't see, it's going to make me mistrust the validity of the system, which is a problem for the game overall. Players should trust the ranking they're being given. I never doubt my ELO in games like R6 Siege and I shouldn't in DBD
0 -
As someone who has a lot of hours as well (2772 at the moment), the mmr system is easier to feel for me. I don't have enough hours/month to keep a super high killer rank and yet have been stuck killing noobs all the time. Today, I'm getting "mostly" rank 1-8 survivors instead of the normal 10-15. My guess is that MMR is more tuned to people like myself than the active die-hards. The system is probably scared of giving you too long a queue time as one of their most active players. It would be interesting to see 50 matchmaking results where the match takes less than 1 minute to find vs 50 where it takes over a minute.
We have no idea how matchmaking considered SWF groups as well. Is it an average, the highest, or the lowest?
Hoping they run the project for a good month or so as we're all getting impressions with too small a sample size.
0 -
I'm actually not sure if two days are enough. I think ideally it would need some time to ballance itself out and gauge the skills of survivors against certain killers, as someone who's good against Oni might struggle against e.g. Doctor or Nurse.
The killer games were mostly unbalanced where the game was usually lost to survivors after 2 minutes except for one game that had a strong team.
I think this kind of system is quite difficult as the deciding factor is not as much how well individual survivors can loop but if the team as a whole is organized and efficient with generators and altruism.
0
