New Stats! (Feb 15th 2024)
There are already some threads sharing their findings from The Stats, but none that actually had all of them listed, so here goes:
Developer Update | Stats! - BHVR
Developer Update | Stats!
By PeanitsLast Updated: 2:55 pm
Since we’ve started regularly sharing stats last year, we’ve received various requests for different kinds of data you would like to see. On top of the usual statistics we share, we’ve selected a few of the most requested topics to share with you today.
Before we dive in, we would like to remind everyone that while data may be fun to talk about, it does not paint a full picture. The numbers we’ll be sharing are very broad, covering millions of matches played all around the world, so the numbers may not reflect your personal experiences. We encourage you to have fun discussing this info, but please try not to draw any conclusions from data alone! It’s important to dig deeper understand the context behind the numbers you see; even we don’t make decisions based on data on its own.
Popular Perks
Starting things off, here’s an update to the 10 most popular Perks for each role! The percentage listed next to each Perk indicates their usage rate- in other words, the likelihood that someone will have it in their loadout. Arrows indicate if the Perk has moved up or down the list since we last shared stats, while a line shows that they remained in the same spot.
Popular Killers
Like Perks, we’ve also tallied up the 10 most popular Killers over the past month. The number below represent the percentage of all matches where that given Killer was played, with the arrows and lines again indicating how they have changed in popularity since we last shared this data.
Beyond the top 10, the pick rates are as follows (in descending popularity). For the sake of simplicity, we have rounded these numbers to the nearest whole percent.
- Doctor: 4%
- Deathslinger: 3%
- Spirit: 3%
- Nemesis: 3%
- Trickster: 3%
- Clown: 3%
- Oni: 3%
- Xenomorph: 3%
- Hillbilly: 2%
- Plague: 2%
- Onryo: 2%
- Executioner: 2%
- Demogorgon: 2%
- Cannibal: 2%
- Cenobite: 2%
- Skull Merchant: 2%
- Pig: 2%
- Dredge: 2%
- Artist: 1%
- Hag: 1%
- Nightmare: 1%
- Singularity: 1%
- Twins: 1%
Deadliest Killers
Who spilled the most blood last month? Many of you wanted to know, so we’ve gathered the data to share with you this time around. The numbers below are the percentage of all Survivors who are killed when facing that Killer. For example, a 50% kill rate would mean they kill two Survivors per match on average. We try to keep Killers near a 60% kill rate on average to keep matches relatively even and support the horror theme of the game, where the Killer is a force to be reckoned with and the survival is not guaranteed.
We’d like to remind you again that this data covers millions of matches across all skill levels. Some Killers may be stronger when mastered, but less powerful in the hands of someone less experienced. (Yes, a good Nurse is much scarier!)
Kill rates do not include matches where a disconnect takes place.
Beyond the top 10, the standing look like this:
- Executioner: 60%
- Hag: 60%
- Artist: 60%
- Xenomorph: 59%
- Blight: 59%
- Wraith: 59%
- Nemesis: 59%
- Legion: 58%
- Good Guy: 58%
- Twins: 58%
- Oni: 58%
- Cannibal: 58%
- Clown: 58%
- Deathslinger: 57%
- Trapper: 57%
- Trickster: 57%
- Demogorgon: 57%
- Singularity: 56%
- Huntress: 56%
- Ghost Face: 56%
- Nurse: 55%
- Hillbilly: 54%
- Doctor: 51%
- Overall Average: 58.50%
Survival Rate in Groups
Last but certainly not least, many of you were curious about a Survivor’s odds of escaping depending on if they’re flying solo or playing with friends. In this case, a higher survival rate would mean that a Survivor is escaping more. We’ve also included the survival rates for high MMR Survivors as well for those who are curious.
One last time, we’d like to remind you that these numbers do not paint a full picture. For instance, a group of friends may be more coordinated, but they might also be more willing to sacrifice themselves in an attempt to save their friends.
That’s all for this time! We’d like to continue sharing statistics like these with you in the future. If there’s something else that you’re curious about, be sure to let us know!
Until next time…
The Dead by Daylight team
Comments
I maintain my MMR by being cute. Gotta get those boops in.
A 2k isn't a win. 3k is.
Very consistent with the 2022 stats and NightLight. Nothing surprising, really.
I think there's a difference between the two.
Killer has been "4K or bust" for years now. Survivors have been altruistic since the start. Unfortunately, I think it's too late to change attitudes.
An MMR change to reflect hooks WOULD be good, objectively. It wouldn't decrease tunneling though.
Yeah, survivors have ways of feeling like they did well, even if they die. Contributing to gens, rescuing, healing and even dying for their teammates. Altruism is just part of playing survivor. Killers don't really have that.
What's chuckies kill rate? 80%?
58%
Oh god i do hope this "2K = win" does not become a common sentiment among survivors.
As always it's funny to see people see average 60% killrate and think it means win 60% of the time.
That'd work only on survivors for whom escape-chance/rate really equals win-chance/rate.
You've been listening to Batusalen too much.
You're getting the outcome of an average match confused with winrates. They're not the same thing. Nightlight statistics show the same thing: outcome of the average match is a 2K with a small trend towards 3K, but the actual rate of 3K+ is almost 50%. Killrate =/= winrate, because the kill distribution isn't neatly equal between all outcomes.
Nobody plays a perfectly average match. Killers are winning more than they are losing.
Maybe that's why we get stats info maybe once or twice per year because everyone will just took what fits their allready set opinion on any side/killer/survivor and use it againts others because "oh look its in the stats!" that is also totaly out of context and includes GLOBAL all regions and ALL MMR.
I don’t count hatch as a win(that’s a mercy out bc the other side beat you so badly),It’s only a W if the gates get powered. Whether I die or not before leaving through them doesn’t matter. If the gates got powered and at least 1 person makes it through them that’s a win bc it took a team effort (in a 4v1) to get that far.
Damn, so even 4 man SWF matches are barely reaching that 50% escape rate huh
I think you keep´re keeping him from getting a rework. Devs look at their stats and think: "wow, he´s a beast! better not touch him for the next decade"
Nurse buff when?
*chuckles*
I wonder why doctor has such a low kill rate
guess it shows that doctor might be that weak I guess
Gonna be honest here, I have no idea who Batusalen is. Enlighten me?
SoonTM
Thats how serious i take those stats. Skill Merchant has the highest kill rate because she obviously is OP and not because everyone suicides on her. Nurse has among the lowest kill rate because she´s the strongest killer.
Balance decisions based on those stats really frighten me.
I mean they left the killer stats vague, so surely they can filter the data for killers based around skilled players performing at higher mmr; as they did for survivor. So I wouldn’t worry too much. Also depending on how well their data operations team functions and what tools they have(software application capabilities) can also sift(filter) around dcs and second stages on same hook compared to other killers. Consider the data and opt whether they would like to include it into their decision making.
so I guess the question to ask is can they apply the same filter process to killers based off mmr(most likely) and balance around the top performance for said killer, and buff those with low performance on the high side as to maintain viability for the sake of variety, and nerf over performers, and can they filter hook stages or purposefully missing second stage checks on hook to look at things like SM through a clear lens, which would depend heavily on the application they use to sort data(idk if an application that complex exists, but would be super useful if someone with an advanced (masters and up) degree in programming could make it, bc something like that would be incredibly advanced to create due to searching for deeper variables).
This info is fun to see, but an important aspect that doesn't seem to be mentioned much is that this data only covers a month's records.
It would be fantastic if they could release this data on a more regular basis; give a clearer picture of what's happening so we can better understand some of the decisions they make.
It is. Looks like you never play survivor. It’s so bad, especially since killers only tunnel and use only slow down perks, which is extremely unfair to go against. I‘m not playing a game to play like a robot, who isn‘t allowed to make any mistakes. I shouldn‘t start a match knowing I will die from the start. When playing killer it is also not fun, because it‘s too easy for me. Very rarely I get a round, which is a bit challenging. It‘s just not fun to stomp most people.
He basically made the same mistake you did: Taking the kill rate, inferring the outcome of the average match, and then considering that the average win-rate.
Weren't you in the last topic he tried to field that theory?
Anyway, a quick trick to realise that 'outcome of average match =/= average win rate' is by taking a sub-50% kill rate. If you take a kill rate of 45%, the average match outcome would be a 2K, with a 20% chance of it being a 1K. This would imply that the winrate would be 0%, which is realistically virtually impossible, as it'd require there to be not a single 3K+ anywhere in the dataset.
Loses one game and drops his kill rate to 50%
Heavy is the head that wears the crown.
They don't make balance changes based just on those stats - Peanits addressed this exact same topic last time they released stats lol
Why shouldn't killing most of the survivors be difficult to achieve?
I don't understand this argument. You want one side of the game to be hard but the other side easy? Why not aim to make both a challenge? If I'm default getting a minimum of 3k every game while barely trying, the game is too easy
It is meaningless stat while the strongest killer got almost lowest kill rate.
Obviously they need to exclude low MMR match to make it meaningful.
That exclusion makes no sense. Low MMR makes up a greater share of the population than high MMR. It seems like excluding low MMR would corrupt the data—and it still wouldn’t mean that killers are somehow weak, as they’re still around 58-60% kill rate at high MMR based on the survivors’ escape rate at high MMR.
It's likely due to the fact that a 60/40 split would mean that survivors on average would lose more MMR than killers.
In order to keep "survivors that escape more than 40% of the time" pairing with killers who "kill more than 60%", surviors MMR would have to be weighted higher.
I think low and high MMR are the smallest percentage groups. Majority of players are in the average MMR bracket.
Peanits specifically said high MMR makes up an extremely small portion of the player base. He did not say that about low MMR.
Nurse has been "the strongest killer with the lowest kill rate" for years. So I don't know why anyone should worry about balance decisions based on these stats.
We all know they base balance decisions on their horoscopes.
it would be a normal distribution that the low and high MMR are always the smallest percentage groups.
If you're getting a minimum 3K every game your kill rate is above 75%. That's not what these stats show.
60% kill rate is still a 2K average on a per-game basis, which is a loss for the killer. To translate to an average 3K, the kill rate would need to be above 62.5%.
Win rate =/= Kill rate. This isn't a symmetrical game where if one side loses the other side must have won. In this asymmetrical game, it is possible for both sides to lose, and both sides to win. A 3K is a win for the killer and a survivor, a 2K is a loss for for 2 survivors and the killer. There are 5 players per game, without an even number of players you can't have an even win rate.
A win for the killer translates to 75% kill rate or more per gamen or 62.5% average. A win for survivor is that particular survivor escaping, or 50% average. It's not comparable.
In order to keep an even win rate, the kill rate has to be higher than 50%, and lower than 62.5%, purely based on number of players per game.
The problem is that because of this asymmetrical format, both sides lose more than they win. Killers are losing more than they win if their kill rate is below 62.5%. Survivors are losing more than they win when their escape rate is below 50%. There is a 12.5% overlap where nobody is happy, and that is the reason why both sides think the game is biased against them. That's why as players, we all need to accept that this is not a game you're supposed to win a lot. Winning is a challenge. For both sides.
This is precisely why I'm happy playing survivor, and playing solo survivor, even when I'm not escaping most of the time. I'm not expecting to escape 50% or more, if I was that would imply the game was far too survivor sided.
It seems like so few people get this. Everyone saying the game is grossly killer sided, or grossly survivor sided, or that kill rates should be 50/50, simply do not understand the game they are playing.
You should be fighting to survive, against the odds, not expecting that you will.
You can't really drop out of the high MMR bracket as long as you're playing regularily. - Also losing consecutive matches doesn't make you lose as much MMR as wins make you gain. If you got a lucky streak over the course of 3-4 matches it will take you more than losing 3-4 matches to go back down again. I think it's meant to keep people from intentionally deranking to get easier matches.
On surv, only when I hardly played for nearly two months (idk, 2-3 hours a week or so; however long it takes you to complete the surv new challenges and rituals) and pretty much only soloQ (I escaped, idk, probably 2/10 matches?) I noticed I eventually dropped out of that bracket (not that I ever wanted to be there but with SWF where everyone has 2k+ hours it's kinda where you automatically end up if you play a lot - which I used to do). - I'd assume that most people up there do at least sometimes play SWF, making it virtually impossible to drop out of that bracket without taking a major break (and then promptly losing a string of matches).
Thats why i chuckled in my first message.
I mean... Billy got buffed after having an underwhelming kill rate. Meanwhile, Freddy won´t see a rework until the next decade.
Hahahaha.
If he wasn't already set for a change, then this could actually be true.
With great power (or in Freddy's case the lack of one), comes great responsibility.
Well it's not 50% it's 40%. That's what the devs aim for. Good SWF is overperforming by that standard.
The games asym so they don't want a 50/50 they want a 60/40.
My numbers are based off of the stats BHVR provided. Unless you're implying they are mistaken.
And the kill rate is literally what defines a win in a match. If the killer gets 3k+, it's a win. If he does not, he does not win. Very black and white. The average statement isn't about every match. It's the average.
As for the 62.5%, that translates to 2.5 kills per match. That means IF killers all had a 62.5% kill rate on average, then on average, they would be winning every other match, as the 2 isn't enough to win, but the .5 carries over into other matches (averaging, remember) which would then be a 3k for a win. This doesn't mean every match is like this. Just on average, a 62.5% would have killers winning roughly half of their matches.
I'm just pointing out that the majority of the killer's kill rates per the official stats are below BHVR's own 60% kill rate expectation. Too many people are demanding "true balance" be forced on the killers by giving them a 50% kill rate. If on average a killer has a 50% kill rate, then that literally translates to on average only getting 2 kills and not winning the match. If they truly were wanting a true balance, the kill rate would be 62.5%, as killers would win every other match. Now, do we WANT a true 50% win rate? I personally don't think so considering the nature of an asymetric game, but im trying to clear up the misunderstanding that somehow a 50% kill rate means 50% win rate, which it does not. That is on averaging never winning.
The upcoming change might be a hard nerf. Something like "people only fall asleep, after Freddy sings his Lullaby 3 times"
*requires microphone to be played
Definitive proof that 4 man SWF breaks the balance they are aiming for. That’s basically 50% when they say they want 40.
…at the absolute top of the playerbase.
Let’s not kid ourselves, that statistic does not reflect the vast majority of the playerbase. You’ll have to excuse me if I find it hard to believe any of us are among those levels.
Wouldn't these kill rates be inflated due to the large amount of self sacrificing going on in the game i.e. Skull Merchant?
It'd be a lot more accurate to exempt those matches as you do the DC matches since both are effectively the same in terms of outcome.
Only if you exempt the matches where killers are memeing or purposely letting everyone go, and all the numerous matches where they willing let someone go. Tbh it probably balances itself out when I really think about the throwing on both sides.
sorry didn’t see someone else responded almost the exact same way, didn’t mean to double down on ya bud.
1 survivor self sacrificing ends up in the other 3 also being doomed causing collateral kills.
A killer letting someone go is only 1 escape. Assuming hatch escapes count.
Survivors self sacrifice a lot more than killers let someone go afaik.
I don't see it evening out, the kill rates would still be inflated.
Some killers only let one survivor go, some let them all go.
Some self-sacrifices doom the other survivors, but they can also be the 3rd survivor letting the 4th get hatch.
I'm not going to try and argue what is more likely than the other as it's impossible to say. But in both cases, it is impossible to discern where these have occurred without watching video footage of every single game ever played.
You could track escape attempts and not attempting 2nd stage struggle at 4-5 gens. Assuming they have the tools to do so.
It doesn't really seem that hard.
Do you count all cases, or just ones near the start of the game? What about when all survivors are slugged? Does self-sacrificing as the penultimate survivor to let the 4th survivor get hatch count?
Heh feel a LITTLE bit less terrible for maining nurse when I did.
I swapped it for maining plague though so idk.
Here's why stats, at least the ones that were released, aren't reliable for really anything.