Why isn't winrate one of the recorded stats?
We kinda tend to mix killer win rates with kill rates although those can be quite different in either direction. If I play bunch of games with 3ks each time for example, thats a 100% win rate and a 75% killrate. On the flip side if my games are a mix of 4ks and 1ks, I have a 50% win rate and a 62.5% killrate. Why isn't winrate being recorded?
Comments
That would require defining what a win is, which varies per player.
I thought it was pretty well agreed upon at LEAST for the killer side that more than a 2k is a win, less than a 2k is a loss and 2k is a tie.
as far as i am aware that is BHVRs official stance
The community rarely if ever agrees on anything.
The 2k thing goes back when kills were tied to pips. As that is no longer the case, it cannot be the messurement.
What could be used and also what I've always seen as the messurement of win or loss, is the pip. Now players have issue with that because it's possible to have both sides "win" or a killer "win" without killing or a survivor "win" without escaping.
That’s still a community-determined metric.
Having a win rate stat would require them actually defining what a win is, which they have never done. I don't think they want to risk hurting a player's feelings by reveling to them a low win rate. They can't even tell us the Entity is displeased.
Where have they stated their official stance?
i could be imagining things but i feel like i remember several years ago a mod stating that the game considered wins to be above 2k, 2k a draw and less a loss. Perhaps for mmr reasons at the time.
The devs can decide the win condition on their own, they just refuse to.
As others have said, that's a community understanding / concept.
BHVR has said wins are totally up to the players. Whether DbD is even something that can be 'won' is a player concept.
From an MMR 'win' standpoint, long term term kill rate / escape rate is what matters (with hatch making it much more confusing). How individual games break down is not important.
The problem there is the game fundamentally isn't using matchmaking that way. The way matchmaking and MMR works in say, a game like chess is simple. If i win, my MMR goes up, if i lose my MMR goes down. Even in a team game like say League of Legends, that is the case even if i play well, if my team loses, my MMR goes down.
DBD is done ass backwards quite frankly. It is balanced and treated as if it is a team game, but then scored based on not a team game.
A match of DBD is not a 1v4 in terms of MMR, it is actually 4 1v1s.
So you queue up as killer, you are playing 4 1v1s at once. When you kill a survivor, their MMR goes down, and yours goes up. So if you kill 1 survivor you "won" that 1v1. If the other 3 survivors escape you "lost those 3 1v1s". Hatch counts as a draw.
But this creates really stupid scenarios where like, what if the survivor i kill is way higher MMR than everyone else, i could technically end a match gaining MMR by only killing 1 survivor. Or vice versa, where i kill 3 survivors, but since they are way lower MMR, i actually lose MMR from the overall match due to the 1 survivor that escaped.
Until they actually officially state a true win condition and actually have MMR account for it and actually balance the game around it, then you have to treat it as 4 1v1s. Which means even a single survivor escaping means "you lost (that 1v1)"
I don't recall ever hearing that. I do recall hearing that matches are just 4 1v1s as far as MMR is concerned.
Of course, I could be wrong, but I don't recall hearing anything refuting this.
But if this is how it works then BHVR has never needed to track win rate as an evaluation of how an entire match played out. This system can create games where a killer could get a 2K and technically "win" (gain MMR) where 2 survivors die, 1 escapes out a gate and 1 takes hatch or abandons but never would a 2K be a killer "loss" where the killer loses MMR. Assuming a fixed and equal value for kills and escapes.
If there is any accuracy to my memory it may have been during the previous mmr system before the one we have now. It was a long time ago, ive played off and on for 10 years.
But thats how i had based my internal win/loss as when i was playing anyways
The previous MMR system was the Emblem system and rank (now grades) were our actual ranks and not just used to calculate out monthly BP bonuses. Sorry but it seems like your memory is entirely inaccurate here.
The only wincon in this game should be 4k or 3k 1 hatch (before all gens are finished).
😒
Why would you define a 3K with a gate escape with all gens done as not a win for the killer and a loss for the survivor team?
Especially for a killer with no gen slowdowns and/or an Endgame build?
Maybe I’m wrong, but I feel like it’s pretty universally accepted that a win is escaping as survivor (with all gens done and gate open) or 3-4K as killer. Now keep in mind that “win” doesn’t necessarily translate to “skill”, which BHVR still doesn’t seem to understand. You can do nothing all match as survivor and still escape because the rest of your team was really good or you can 3-4K as killer with very little skill if the survivor team is just messing around or not very good.
So long as there are people that only view a 4K as a killer victory, it’s awkward to state a ‘winrate’.
As for survivor, maths was never my strong suit, but shouldn’t the escape rate equal the winrate, if your individual escape is a victory?
In the early years ( i think before hag/huntrrss release) the surv community did some serious gaslighting to make the killers accept a 3k as a win.
"3k is a win" "you won already, show mercy" etc.
Personally I think its also slowly happening with "2k is a draw", instead of a loss.
As for your question about survivor, an escape is a win. The binary results for survivor, namely having escaped (gate, hatch) or being dead (sacrificed, killed), makes it easier to judge.
You dont get points for the other 3 survivors escaping when youre dead. And you dont "lose" if youre the only one escaping. There is no team being judged in dbd.
Maybe I’m wrong, but I feel like it’s pretty universally accepted that a win is escaping as survivor (with all gens done and gate open) or 3-4K as killer.The first part is not universally accepted as all. Plenty of players would happily take a 3e where they died over a 1e where they were the only one to escape.
Now keep in mind that “win” doesn’t necessarily translate to “skill”, which BHVR still doesn’t seem to understand.This is a common misunderstanding of BHVR's argument.
In the long run, which is what MMR cares about, the exceptions wash out. Across a 100 or a 1000 games will people escape matches they shouldn't and die in ones they should? Absolutely. It's just not statistically relevant and generally applies equally to all players.
Yeah, a "win" as a surv is much more subjective. I for one would feel much better about making great plays that allowed 2-3 of my teammates to escape than putzing around and slinking out while my teammates died. It's not at all uncommon for the worst surv on a team to be the only one who gets out.
But other people have different ideas about this, and that's the point. Too hard to settle on one surv wincon.
Which is why I think that while not perfect, emblems/scoring is the best way to settle surv MMR. You can be bad but escape consistently if you play a certain way. It's not so easy to be bad and consistently rack up high scores.
Which is why I think that while not perfect, emblems/scoring is the best way to settle surv MMR. You can be bad but escape consistently if you play a certain way. It's not so easy to be bad and consistently rack up high scores.The problem with this, unless BHVR changes its stance on MMR being public, is that it doesn't really change anything about people playing.
At best I can see it changing who makes it into upper MMR, but unless matchmaking is also made stricter that seems unlikely, and also potentially just entrenches 'low MMR hell'.
I think the easier way they could have gone with MMR on the survivor side is just always make it group based.