Yesterday’s Q&A Livestream with the Dead by Daylight dev team gave us quite a bit of info on upcoming changes to the game, including nerfs to the soon-to-be-released Boon perks and possible modifications to the Prestige system. Additionally, Lead Game Designer Patrick Harris finally explained exactly how the newly implemented Skill Based Matchmaking System tracks player skill. Debates based on anecdotal evidence have been commonplace in online DBD communities since the SBMM was first turned on, so it was nice to get some concrete info — even if that info didn’t exactly make everyone happy.
As it turns out, the SBMM system is fairly simple: As a Survivor, if you escape the Trial, you win. If you are sacrificed, you lose. As Killer, each sacrifice counts as a win, while each escaped Survivor counts as a loss. How much these wins and losses affect your hidden skill rating apparently involves some “fancy math”, especially for Killer. Patrick explained that Survivors that die 2nd-4th have their losses counted not as a full loss, but reduced based on the order in which they died, with the final Survivor’s death counting the least.
Interestingly, escaping through the Hatch does not count as a win. Instead, the SBMM system treats a Hatch escape neutrally — neither a win nor a loss. Patrick said on the livestream that this is because Hatch escapes don’t seem to correlate statistically with skill. It was also revealed that SBMM ratings are not capped, although there is a soft cap in terms of how matchmaking works. Players with outlier MMR will have their skill rating rounded down to prevent them from never being able to get a game.
When asked about SBMM as a whole, Patrick stated that matches are much fairer with the new system, and also said that the granularity of information that they can now get from matches has improved their ability to balance the game. Assuming his claims about more matches being fair are true, this makes a lot of sense: if you have more balanced trials, you can make much more confident assessments about the performance of Items, Perks, and Killers.
To say the reactions on Reddit to learning how the SBMM system worked were negative would be an understatement. Players are not happy with the fact that someone can hide in a locker all game and get the same MMR increase as someone who did all the gens — or even worse, that a hiding player can earn themselves a win while someone who loops the Killer all game and is then camped to death gets a loss. These are fair concerns, and it certainly is a possibility that players will begin taking advantage of the SBMM system to win without helping their team.
However, if the system works as intended, players should find themselves escaping in half their matches, and being sacrificed in the other half. A player who hides all game and then escapes isn’t contributing to their team’s success, and it seems clear that a player who never does gens or never learns to loop will end up with a significantly lower SBMM rating than a player who actually helps their team win. The fact that MMR is hidden also means there isn’t really any motivation for players to intentionally game the system.
Matchmaking, especially in a game that sits awkwardly between casual and competitive, is unquestionably challenging. Only time will tell whether or not the system ends up improving the game, but it’s great to see the DBD devs being more transparent about their systems and their reasoning behind them.
There was a lot more besides SBMM covered in the Q&A, and Reddit user BenMQ created a fantastic summary — if you’re looking for more details on what else we learned from the DBD dev team yesterday, go check it out!
What do you think about the logic behind the Skill Based Matchmaking System? Have your DBD games felt better or worse since the system was implemented? Let us know in the comments!
Share this article:
Unabashed FromSoftware fanboy still learning to take his time with games (and everything else, really). The time he doesn't spend on games is spent on music, books, or occasionally going outside.