The math behind DbD's MMR is just a single "if" statement (900h vs 11,000h lobby analysis

You are arguing from the perspective of a theoretical vacuum, completely ignoring how Dead by Daylight actually plays out in reality.

1. "Activity vs. Skill"
   You claim that measuring chases measures "activity, not skill." What other mechanical skills are there in DbD? Are we even playing the same game? The entire mechanical depth of this game is built around looping (windows/pallets), managing totems, and item efficiency (flashlights, toolboxes, firecrackers, med-kits). If a player successfully loops a Killer for 3 minutes using precise mechanical inputs while their team repairs generators, that isn't just "activity"—that is the highest expression of skill the game has to offer. And what are the main goals in DBD matches? Looping the killer and repairing generators = escape. Without generators, you won't escape, and without looping the killer, you won't be able to repair generators. What else do you want to measure as a goal? People optimize their actions anyway, whether it's for generators (genrush) or killer looping. This has happened, is happening, and will continue to happen regardless of the system. Why? Because this game features perks and the ability to build different combinations. When I test a new perk for looping, I create a full build for looping and optimize my actions for looping. If this is no longer happening, then why are people using perks like lithe and windows? Precisely for looping. Already, depending on which killer they encounter, people are pre-dropping pallet after pallet, doing nothing but destroying pallets, so I don't know why you're pretending that this is only going to happen when it's already happening. And if they optimize their own skills for looping (for example, training on custom maps), what's the problem? Isn't the point to keep getting better? Where's the problem here?
2. The "Optimization" Myth
   You argue that if the game tracked chases, players would "optimize for those metrics instead of the objective." Have you looked at the current meta? Players already optimize for chases! The vast majority of loadouts are entirely built around chase perks, exhaustion perks, flashlights, and looping add-ons. The community already plays this way. The only difference is that the current 1/0 MMR system is completely blind to it.
3. You Misunderstood the Math
   I am not suggesting that a long chase should equal a massive MMR spike. I am suggesting a slight, weighted bump. If you loop well, your MMR should receive a fractional adjustment to reflect that mechanical competence, rather than the system treating you exactly like a beginner just because you sacrificed yourself on the final hook.
4. The 1/0 System Measures Luck, Not Skill
   Currently, the binary outcome doesn't measure a "reliable signal." It measures who had the best combination of map RNG, meta perks, team coordination (or lack thereof), opponent mistakes, and sheer luck.

If you truly believe that looping a Killer and buying your team the time to finish the objective has no bearing on actual skill, I have to ask: what do you actually do in your matches? Do you just sit in the corner of the map doing generators and never engage in a chase?

I didn't ignore your question; you just ignored the answer. I explicitly wrote: "nobody is saying 1 hour equals exactly 1 MMR point." Asking me to calculate your exact hidden MMR score based on your 7,200 hours is asking me to reverse-engineer a proprietary black box that BHVR actively hides from its player base. Playtime is a heuristic proxy—a statistical approximation we use because the actual MMR system is completely corrupted by the locker-hiding scenario you literally just provided. And I don't see you playing realistically. I can't give you your exact MMR number, but basic data science dictates that your 7,200 hours put your mechanical baseline in a completely different galaxy than a 700-hour player.

As for your second point: "It's easy to criticize than make a system that has no flaws."

That is a textbook Nirvana Fallacy. Nobody is asking BHVR to write a "flawless" system. We are asking them to write a functional one.
I DON'T WORK AT BHVR, I DON'T GET PAID FOR A SYSTEM THAT DOESN'T WORK. That's one thing. Another thing is that just because I don't create the system in DBD myself, it doesn't mean I can't criticize it. No one here is talking about a perfect system, but about a system that is fair, even if only a little, and that works, even if only a little. If I had created such a system at university or at work, neither my professor nor my boss would have accepted such an explanation. I pay for the game and for the characters - I have the right to demand. If we are going to trivialize everything in an ignorant way and say, "If you don't have any ideas, don't criticize!!! It's impossible because we don't want to!!!“, then all discussions about characters, nerfs, buffs, balance, or the ”holy trinity“ in DBD are pointless, they have no right to exist because ”You're not doing it? You don't have any ideas? Then don't criticize!!! You're so smart? Then give us the perfect solution instead of criticizing!!!". No one is talking about ideals, just about making the basics work, so I don't know where you're getting that from. Don't ignore the whole thing.

Industry-standard matchmaking systems in games like CS2, Valorant, or League of Legends are not "flawless," but they track multiple variables, use weighted averages, and accurately separate beginners from veterans. DbD's system relies on a primitive if (escaped) boolean from 2016. Criticizing a multi-million dollar studio for relying on lazy, one-dimensional spaghetti code instead of implementing industry-standard architecture isn't just "easy"—it's necessary if we ever want the game to improve.

Your "Q.E.D." is premature because your entire argument rests on a logical fallacy: that asymmetry justifies data blindness. In reality, asymmetrical systems require more granular data to achieve balance, not less.

1. The "Asymmetry" Excuse
   You claim comparing DbD to CS2 or Valorant is a "false equivalence." From a data science perspective, that’s incorrect. Whether a game is 5v5 or 1v4, the goal of an MMR system is the same: to minimize the variance between player skill levels in a lobby. Asymmetry doesn't make math impossible; it just makes it more demanding. The fact that DbD is 1v4 is exactly why relying on a single binary outcome (escaped/died) is such a massive failure—it completely ignores the complexity of the team dynamic.
2. The "Context & Perks" Fallacy
   You asked specifically: "What if you have Deliverance? Babysitter? Unbreakable?" and claimed a system can't understand this nuance.

In software engineering, perks are not mysterious human nuances—they are explicit, known variables. The server knows your exact loadout before the match begins.

If you use Deliverance, the system already knows you performed a safe unhook earlier to activate it. It logs a conditional self-unhook event. If it didn't register, Dliverance would never have been activated. It's not magic.

If you use Babysitter, it applies a modifier to the "safe unhook" metric because you gave the unhooked player haste and hidden scratch marks.

If you use Unbreakable, the system logs the time spent slugged and the recovery state.

A multi-variable system doesn't need to "think like a 10k hour player." It just applies mathematical weights to actions based on active modifiers. Ironically, the current 1/0 system is the one that gets fooled by perks. If a terrible player loses a 5-second chase, gets slugged, uses Unbreakable at the exit gate, and crawls out, the current system rewards them with a "Win." A granular system would see the actual telemetry (short chase, slugged, perk carry) and adjust the rating properly.

1. The Time-Space Trade-off
   You claim the issue is just "personal preference" between queue times and match quality. But you ignore the fact that the "search band" in DbD doesn't just expand—it collapses almost instantly. The system doesn't try to find a fair match for 2 minutes and then expand; it hits a "panic button" after 30 seconds and pulls a 900-hour player to face an 11k-hour squad. That isn't a "trade-off"; that’s an architectural surrender.
2. The Hypocrisy of "Nuance"
   Lastly, you conveniently ignored the point about Camping, Tunneling, and Slugging. The community constantly demands complex, context-heavy "anti-tactic" code to stop these behaviors (like the anti-facecamp meter). Yet, you argue that calculating a slightly more complex MMR system is "impossible" due to "nuance." You can't have it both ways.

. You wrote “while knowing nothing about the current MMR system.” But what do we know about this whole “mythical” MMR? We know that it collects data in a binary manner. You escaped - MMR up. You died - MMR down. 4K - MMR goes up. 3K MMR goes up. 2K - MMR stays the same. 1k and 0k - MMR goes down. And that's it. I don't know anything about the actual state of my own MMR because there's no way to check it, we have nothing. So tell me honestly: what are windows, pallets, and generators for? If the system only counts the end result - escape/kill, then I can hide the entire match and according to the system I'm better. And according to the same system, my MMR is high enough, so I can face a better player. And the fact that I can't loop, I can't hit a skill check on a generator, doesn't matter, right? Right? And tell me, honestly, if that's the case, why do so many survivors sweat, take meta builds for looping and generators, when all you have to do is hide in the bushes to be better? Or the other way around: why have anti-slug, anti-camp, and anti-tunnel systems if, when I want to play against weaker players, I can camp 1 survivor - 1 kill - MMR down and goal achieved? Or why use anti-slug if I knock down 4 survs and do a full slug because they made 1 mistake, I'll do 3k or 4k even though I'm blocking myself with grass, I don't know how to mind game or loop, and according to the system I'm better, right? Right? So what's the point of these systems? Well, exactly... The answers to these questions are also answers to the fact that THE CURRENT MMR SYSTEM DOES NOT WORK AND NEEDS TO BE CHANGED! IT DOESN'T HAVE TO BE PERFECT, BUT IT HAS TO WORK! That's it.

If BHVR can't handle the "nuance" of tracking an Unbreakable play, how can they ever hope to balance the actual gameplay? We aren't asking for a sentient AI; we are demanding that a multi-million dollar game stops using a 2016 boolean to manage its 2026 player base.

Once again, you are putting words in my mouth that I did not say. NO ONE, ABSOLUTELY NO ONE, DEMANDS A PERFECT SYSTEM! ALL THAT IS DEMANDED IS A SYSTEM THAT IS A LITTLE BIT FAIR AND A LITTLE BIT EFFECTIVE!
That too much to ask of bhvr studio? If so, how did they manage to create a game like dbd? How are they still able to create new DLC? People, read carefully...

The 5v5 lobby is the complete opposite of asymmetrical DBD. But this system could be improved. Not perfectly, but better.

Exactly. I have outlined methods that would improve the system. Not create a perfect system, because that is simply impossible, but one that is better than the current one. But bhvr's laziness, their greed for money, and the community's stupid attitude (here's a perfect example - “they don't collect this data during the game, it's too difficult”). And yet, every match they get points for blinding the killer, the anti-face camp system works xd), nothing will change.

Your argument runs into a fundamental contradiction.

You’re saying a survivor who runs the killer for several minutes but dies should still gain rating because they showed “mechanical skill.” But ranking systems aren’t designed to reward isolated mechanics, they’re designed to rank players based on their ability to produce wins.

If a system starts awarding rating to players who lose simply because they performed one part of the match well, the ranking immediately stops reflecting win probability. At that point the optimal strategy isn’t winning, it’s maximizing whatever metric gives rating.

That’s why competitive matchmaking systems anchor ratings to outcomes. Individual stats are far too context-dependent in team games. A long chase might come from survivor skill, but it might also come from killer mistakes, strong map tiles, teammate interference, or the killer committing too long.

Once you detach rating from outcomes and start rewarding intermediate metrics, you create a system where losing players can climb. That doesn’t measure skill, it breaks the ranking entirely.
So what? After our discussion, I can tunnel, slug, and camp with impunity? Well, if it doesn't matter what I do during the match, what skills I demonstrate while looping, because only the final result counts, then I have to strive for that final result. So the anti-face camp, anti-slug, and anti-tunnel systems are pointless, unnecessary, and completely destroy the current MMR system. At least according to your logic, because in reality it's different.

You are conflating "total accounts created" with the "active matchmaking pool," and using the "casual game" excuse to defend broken architecture. Let's break down why this logic fails:

1. The "Median Player" Fallacy
   Yes, if you count every person who downloaded the game on Game Pass, played two matches, and uninstalled it, the median playtime is low. But matchmaking algorithms don't sort inactive accounts; they sort the active, daily player base. In a 10-year-old live-service game, 900 hours is absolutely mid-core .

You misunderstood me. By “a perfectly average, statistically balanced player” with my nearly 900 hours, I mean that I am an average player IN THIS SPECIFIC CONTEXT, in this specific game. Not about the entire gaming world This is a forum about DBD, a post about DBD, we are talking about DBD. It's like in football - I would be an average player. Not the worst on the school team, but not the best in the world or even like Ronaldo. Saying that I'm an average soccer player doesn't mean that I'm an average player in every game, in every sport. We're not talking about being an average player in games in general, but in DBD.

But let's assume you are right and 900 hours makes me a top-tier veteran. If that is true, why does the matchmaking system frequently put me against squads with 11,000 hours? Why does the system sometimes match me with players who have 20 hours of playtime? Since I'm a veteran and the system works, this shouldn't happen. If the system was functional, it would recognize that massive gap. The fact that it throws us together proves the soft-cap is broken.

1. The "Casual Game" Excuse
   You claim: "since when was DBD ranked or even meant to be Balanced? ... This is a casual game."

If Dead by Daylight is just a casual party game, why did BHVR spend massive resources to implement strict Skill-Based Matchmaking (SBMM)? They explicitly removed the old casual Rank system to force a hidden, competitive MMR rating onto the entire player base.

You cannot excuse a broken competitive algorithm by claiming the game is "casual." If it's truly a casual game, BHVR should remove the MMR system entirely and let matchmaking be random. But they don't. They insist on using an MMR system, which means as paying customers, we have the right to demand that the system actually does the math correctly.

If it's “just a casual game,” then why are so many players sweating it out and trying so hard? Why have this forum, why make these balance changes if it's “just a casual game”? The excuse “it's just a casual game” in discussions about balance and player selection for this game is really not a valid argument. In fact, it's proof of ignorance and disregard for serious problems. Following this logic, one could say: why bother with anti-slugging, anti-camping, and anti-tunneling if it's “just a casual game” and people will play a few rounds and then leave? At the same time, people who play DBD a little more want these changes. Why? Why bother? Who are these changes for? I'll tell you: for people who play it a little longer.

1. The Park Chess Analogy
   Your chess analogy actually proves my point. If I go to a park and a Grandmaster sits across from me, I can visibly see that I am outmatched, shake his hand, and politely decline to play.
   If the game was never meant to be balanced, then why the balance changes? Why nerfs and buffs? Do you even know what's going on in these patches, or are you only interested in new characters? DBD isn't “ranked” because it doesn't have ranks, and that's also a problem. It does have MMR, though. A poor MMR, but it has MMR. However, complaints about MMR and “chess ELO” should be directed at bhvr, not me.

DbD's matchmaking system blindfolds you, glues you to the chair across from the Grandmaster, and hits you with a temporary matchmaking ban if you try to stand up and leave. Because the game forces us into these automated lobbies and punishes us for leaving, a functional automated sorting system (MMR) is absolutely mandatory to protect the player experience.

True. Most of the whiners at matches have TTV in their nicknames, and they're the ones whining on Twitter. And bhvr listens to them.

1. The DodgeWhether I introduced the 5-gen looper scenario in my last post, the one before it, or yesterday is irrelevant semantics. You are bringing it up only to still dodge the actual question. I will ask it a third time:If a player loops for 5 gens, secures 3 escapes, but dies on a facecamp, who produced that win?You refuse to answer because acknowledging it destroys your theory. Under your 1/0 model, the MVP who generated the win gets punished (MMR DOWN), and the three players who safely held M1 on generators get rewarded (MMR UP). That does not measure "win probability." It measures survival by proxy.
2. The "Variables" Excuse and Basic StatisticsYou claim telemetry can't determine if a chase is skill, map RNG, or killer mistakes. Welcome to the Law of Large Numbers. Over a sample size of 50 or 100 matches, map RNG and lucky tile spawns average out. If a survivor consistently logs 2-3 minute chases across dozens of games, that is statistical, measurable mechanical skill. Furthermore, forcing a killer to make a mistake and capitalizing on it is skill. A functional algorithm smooths out variance over time; the 1/0 system just ignores it entirely.

You write about these random events, about map RNG, tiles resp, or teammates getting in the way, as if that weren't already the case. It already does. And how does it affect the MMR result? Well, by demonstrating your skills, you take advantage of it. If Blight happens to be on a good open map, and the scouts make basic mistakes, Blight takes advantage of it. It already happens. Your problem is that you are trying to give examples of situations where the improved system would have a problem, would have to deal with them (but the current weak system already does), as if they only appeared with the new system, but they already exist. Your examples are already present.

1. The Emblem Myth and "Gaming the System"You claim the Emblem system was replaced because it was "easy to game." That is historically partially false. BHVR implemented a hidden MMR to enforce a synthetic kill rate of around 50% and soften matchmaking parameters to reduce queue times, not to ensure fairness in gameplay. But fairness in gameplay isn't okay here either. You want to talk about a system being "easy to game"? Under your beloved outcome-based MMR, a Survivor can hide in a locker for 15 minutes, touch zero generators, get zero unhooks, wait for their entire team to die, jump into the Hatch, and gain MMR because they "survived." Meanwhile, the teammate who did 3 generators and unhooked twice dies and loses MMR. A system that rewards locker-hiding and punishes playmaking is functionally broken.
2. The "Losing Performance" ParadoxYou state: "The moment a ranking system allows players to climb while losing matches, it stops measuring who actually wins."Dead by Daylight is a 1v4 asymmetrical game. Individual survival $\neq$ team win. If my death guarantees 3 people escape, my performance directly engineered that victory. In modern, highly competitive games (like CS:GO, Valorant, or Apex), underlying personal performance metrics actively mitigate your ELO loss—or even allow you to climb slightly—if you play exceptionally well but the match is a technical loss.Punishing the playmaker breaks the ladder. Recognizing their telemetry fixes it.