Why are there no more killrate updates?
I would like to know from BHVR why there are no more killrate updates?
According to BHVR, these numbers are constantly monitored and should therefore be easy to share with the community.
So why are there no more updates for all killers and separated into high and low MMR?
Comments
-
We put out various stat data, especially when things change drastically. Right now the last kill rate data still applies - so there's no real need to reproduce the data.
14 -
so you want to tell me, that all the changes that made the last year have changed nothing? that would be highly, highly unusual.
therefore, I am making the claim that the new data could concern the playerbase and that is why it is being held back.
22 -
I believe the kill rates we received this year came after those changes. My only source of umbrage is that the kill rates don’t factor matches with a DC and I think hatch escapes.
4 -
lets say all the killer buffs did nothing to the killrate. that means we can expect more killer buffs to change something?
10 -
Hatch escapes are counted as escapes in the kill rates. The only place hatch is treated differently is with MMR hatch counts as a "draw/neutral" match result.
5 -
Do you think Kill Rates have declined? What are you seeking to validate that could be a concern to Playerbase? Do you even believe the Kill Rates from that time?
0 -
you can check the evidence by yourself on nightlight.gg. The numbers are off by over 5% for most killers.
if you dont trust or like nightlight.gg, like i do, you might ask BHVR to give out the official numbers, like i did.
Post edited by BoxGhost on8 -
Yes i believe the kill rates from that time.
If the numbers have remained stable, it means BHVR is on the right track to create a balanced game by their standards. That's why I'm surprised they don't want to show that.
2 -
Accusing someone of lying based off of Nightlight alone is nuts. It may acknowledge trends, but Nightlight's data is nowhere near exhaustive.
4 -
Does the third party website show the kill rates have changed a lot this year, or are you only checking the current stats?
Because although you might be able to argue there is a correlation between the official stats and the third party stats, you need to track if there are any offsets when comparing the data. For example, the third party kill rates might consistently be 3% lower than the official stats.
TD;LR when using third party data, you need to track the kill rate trend over time, instead of just looking at the current data.
0 -
Correct. Another reason to only compare the official data that only bvhr can give us.
Please stay on topic and do not try to derail the thread by things that you came up with and I never said. Please contact me personally if you want to discuss this any further.
7 -
The thing is the kill rates have definitely changed, its statistically impossible for them to be exaclty the same, so its strange Bhvr. would say that. In my own experience the game seems way more killer sided then it did almost a year ago when those stats were released and I escape way less now then I did then, so I dunno.…I am definitely skeptical.
13 -
Kill rates are not a good indicator of strength. That is partially because MMR aims to put killers in a place where they can maintain a 60% kill rate no matter how strong or weak their killer is.
This also explains why every time a killer is buffed or nerfed, there is an immediate difference in their kill rate that ceases after some time. Only when kill rates remain consistently high or low for an extended period of time, do you even have a chance to draw some conclusion on the killer's strength from it.
The second problem is, that there are killers that are objectively more powerful than others, yet maintain a lower kill rate. Like Nurse. Meaning, while kill rates and killer strength generally have some correlation it is not as linear as "This killer has a high kill rate, so they are overpowered." or "This killer has a low kill rate, so they are weak."
10 -
You can do all the mental gymnastics you want but.killer is objectively easy mode in the game. The ability to "go next" on hook is the only reason survivor players dont just stop playing the game at least in solo que.
7 -
I'd like to see kill rates broken down by MMR and number of players in parties, on a per-Killer basis. And I'd like to see disconnects factored in (since they absolutely count when it comes to player enjoyment).
In fact, I'd like to see disconnect rates broken down the same way. Also, tell us how many generators were completed per 3-4K.
I don't think anyone from BHVR is flat out lying, but I guarantee they won't share stats that make them look bad…
11 -
Nightlight is self-reported with a much smaller sample size. I question someone's data literacy if they put more stock in self-reported small sample size kill rates over aggregate kill rates.
5 -
How is nightlight in any way indicative of the vast majority of the playerbase, when only a small fraction use that site?
3 -
For a long time the bps incentive has been on the survivior role which would dictate that the survivor role is low on players. Also, I play majority soloq and majority of those matches the gates aren't powered. In fact the only reason I keep playing survivor I'd that there is a small chance that hatch is an option to escape otherwise 90% of my matches would be 4ks from the killer.
I genuinely think the devs need to address this and show data on how many matches are ending in 4ks so we can get a good picture of the balance of the game. From where I'm sitting it's not that balanced. Personally I think 4ks and 1ks should be rare and it should lie mostly in 2-3ks where gens are either powered or 1 left. Also my point being that just because gates are powered doesn't mean the game is over but majority of my matches I'm experiencing the gates don't get powered. (Please note that I'm solo q and trying to help my teammates by going for saves and heals as well as trying to do gens in the process)
4 -
How is any of what I said mental gymnastics? Nothing I said can even be put up for debate. It's simply facts.
They could release a powerless killer with 4.4 m/s movement speed tomorrow and without perks I would still be able to achieve a 60% kill rate eventually because I would still dominate beginners without any issue whatsoever.
This is what happens when someone with thousands of hours of experience plays against people that have no clue about the game. Would such a killer be considered strong? Definitely not.
5 -
Well then thats a flaw with the game. Also killer dominates at every level not just with beginners. You cant deny that killer has gotten easier and survivor harder over the years.
5 -
The devs themselves have stated that the incentives are unique to not only each region, but MMR bracket; as well as being affected by time of day you play. So your experience is not the same for everyone. Just because you are seeing mainly survivor incentives does not mean that there are not enough people playing survivor.
3 -
Not to mention just because the incentive is on one side it doesn't always mean you will get faster queues that side. Sometimes it is the opposite for some reason.
2 -
Because it's 100% unreliable data that misleads people.
4 -
You are missing the point. I explained that even with a killer that would be objectively too weak to be considered balanced an experienced killer could maintain a 60% kill rate.
Whether or not this is good design is a different debate. I'd be more than happy with 50% kill rates, if they can find a way to keep the killer experience pleasant. We had that before and 10+ minutes queue times for survivors weren't exactly rare. So it's safe to say that a lot of people didn't consider the killer role very fun back then.
2 -
I don't think it's fair to call it unreliable when it's literally as accurate as it can get.
5 -
Well thats fine but then I hope you dont complain about survivors "going next". (Im not saying you do, just hope youre not of the many on these forums that complain about it.)
2 -
You have to understand the difference between "not changed" and "not changed significantly".
0 -
It is so obvious (to many I hope) yet it looks at least some fail to understand the concept.
Too many variables to take into account?
I'm amazed. At least now I better understand some of the outlandish posts I see popping from time to time.
1 -
Its obvious anyway you look at it though that killer is overpowered and easy mode in the game, both anecdotally and statistically. And just becauae statistics dont always tell the whole story it does provide a good deal of evidence when combined with other (such as the amount of survivors giving up) that killer overall has become the easy assisted mode in the game ( yes even freddy.)
3 -
I never understood why nightlight got so much flack and hate. Yes, its selfreporting, so its only the part of the playerbase thats even interested in this kind of data AND someone who is some kind of fan of the game, because someone casual wouldn't even know about the site, but its also showing you the other survivors and killers. When I record my killer matches there, the four survivors I verse are 100% true occurences and not in any way tampered with.
Likewise whan I play survivor, I get a pretty good picture of what general perks the killers I verse are running. Its a small sample size, but its result have been remarkably consistent with the date that BHVR has released over the years. The top 10 perks for both sides were 7-8 of 10 right, with often matching percentages, which is quite remarkable.
So just because its not official or self reporting this doesn't mean that Nightlight can be completely and easily handwaved and dismissed. There are many datapoints that are absolutely validly taken and Nightlight gives a pretty accurate snapshopt of the current situation.
I would love, though, that BHVR would allow easy tracking via steam. Steam already has this capabilities and if they greenlight it, we could all pool our steam data tomorrow. Why its not allowed is everyones guess.
5 -
It's an interesting site, but it has problems from a sampling perspective.
- Self-reporting- Wins and losses can have an affect on whether someone chooses to report the result. There's no automatic data ingest.
- I don't see any language options on the site. Someone can correct me if I'm wrong. But if it's English only, you're naturally going to over-sample Western countries.
- It's going to heavily over-sample PC platforms i.e. Steam/Epic via the method in which they ingest data. They take the data via screenshots through the Web UI or desktop app. It's not impossible for a console player to upload, but they'd have to jump through hoops.
It's a cool site and it can be fun to look at, but the data has problems.
4 -
Thats all valid and good points. I mostly thought about it in terms of self-reporting an win/loss and hadn't given the two other points much thought.
Still, in the end, just like with asking random strangers, you don't need to ask everyone, 1000 people is mostly sufficient, and Nightlights data has always been pretty close to BHVRs officially released data. Because Nightlight is basically live data you also see the effects of buffs and nerfs or even bugs pretty quickly, which is always fascinating.
So while I wouldn't take nightlights data as gospel or use it as hard currency in a discussion, I think it offers quite a good snapshot of the current situation that we shouldn't just dismiss it.
4 -
I am certain any general buffs killers receive (and/or consequently nerfs survivors receive) will raise kill rates. What BHVR provides is already inaccurate as their data tosses matches where at least one DC has occurred. If 5% of matches have a survivor DC and 3% of those matches end up a 3/4K, then the killrate would technically be 3% higher than BHVR has stated. But it doesn’t really matter right now.
5 -
How come kill rates for events and other event stats arent posted?
0 -
I suspect they have probably increased. But i hold to the theory that kill rates are only as high as they are due to the "giving up" epidemic among survivors right now.
Seriously like 40% of my games a survivor gives up in the first 3 minutes.
7 -
The give up player ones generally arn't counted so its higher.
1 -
I don't believe that is true. They remove DCs from the data, but not giving up. Also giving up has significantly different definitions, its not just a "Hook suicide"
4 -
Exactly. It does offer some cool insights into killer and perk popularity. And I do upload from time to time, because it's very good and convenient as a personal data tracker.
0 -
They are included. The devs confirmed it with Skull Merchant. IIRC her killrate was 71% and when they were asked if that included people giving up, they said if they took away people who die on first hook then her killrate was 68%.
2 -
We've had changes on both sides. Now if something has moved a few percentage points I think that is statistically significant given the sample size, though BHVR doesn't seem to worry about things like that too much.
But if Mandy says that the data is pretty much the same, then the data is probably pretty much the same.
0 -
You brought a very good point, probably the most important point of all:
have the kill rates increased because of the "giving up" epidemic?
That would be the point where we would need the official data. So we can only speculate what it actually looks like. However, both possibilities are frightening and would concern the players.
The kill rates have increased due the giving up epidemic:
First of all, it would show that there is a serious problem on the part of the survivors. The reasons for the epidemic are too diverse to discuss here and every player has different points of views how to deal with it. At least we can say measures should be taken as soon as possible. But can we tell how the real kill rates would look like? How do you filter out for example the matches where the second person got hooked give up? I think you cant without personal influence of the person who edits the data.The kill rates have NOT increased due the giving up epidemic:
Then all the survivors feel wrong? the new maps with the dead zones have no positive influence on the kill rates? For killers who received a base kit change nothing has changed? On the contrary, this means that the real kill rates have actually decreased since one has to assume that certain percentages are due to giving up. In this case, how do you say what the actual kill rates are and which killers are too weak or just not fun to go against?For BHVR it presents a different problem. They no longer have reliable data to work with. They claim for themself they do mostly data driven changes. Now that we've established that we don't have any backed up data, it means that BHVR is making changes blind. Is that why we're seeing weird changes like the recent SM nerf? The huntress buff nobody asked for. Is that why they're trying to find out what the community thinks about the reversal of virtually all of the Trickster changes?
For me all of that is very concerning.
4 -
You can't understand the principle even with the explanations?
And it works for any player by the way.
I'm sorry I have no idea how to dumb it down even more than what was already done.
4 -
I hate the give up epidemic. probably 20-40% of my games there's someone who gives up immediately which means those games arent counted on there data but people are still experiencing it so realistically survivors are losing 70% of their games on average
3 -
Again, what I wrote and survivors "going next" have nothing to do with each other. Nothing at all.
Most of the time survivors rage quit whenever even the tiniest inconvenience occured. It doesn't matter whether they were about to win or lose because it's readily available and any game will have these moments of minor frustration.
4 -
Posts like this are probably the exact reason why BHVR are so hesitant to show any stats.
There's way too many variables and behind the scenes factors that go into these stats that them releasing them often has no real benefit. They just get twisted to fit whatever narrative you prefer. They didn't even share them here yet everyone is still making up random conspiracies about it.
2 -
Probably because the stats do not mean anything, and are hilariously bad for determining anything about DBD
3 -
What are you even talking about? You're just making random pronouncements. You said that Survivors just have to accept that they will lose because the game is designed for killer to be easier and survivors have to accept that. Well then you have to accept survivors won't stay in this worthless game. If you want to say nothing matters, statistics, circumstancial evidence, nothing, then go play a pve game. Otherwise accept that survivors are actual humans and don't care about how you try to justify an unfair game. Survivors aren't going to stay in the game just to massage your ego so you can pretend you're good at a game even though you're playing with rules rigged in your favor.
3 -
Putting words in my mouth, are we? I did not say that survivors have to accept that they will die. I simply explained why kill rates are not a good indicator for a killer's strength and that this has nothing to do with survivors giving up.
If that was the case, then we'd have seen way more disconnects and hook suicides right after 6.1.0 because that was when killers were at their strongest, considering that killers had a realistic way to hold the game hostage.
If you really don't want to play the game, then don't do it! Is that so hard? Don't threaten to do something but actually follow through. Don't start the game and don't queue up, if you are going to leave anyway.
You clearly don't like the game and the rest of your team as well as the killer don't like to play a game that is over because someone decided to rage quit.
3 -
On top of this, there is also the issue that they exclusively use survivor-reported data for killrate stats, so there is both the bias of the kind of matches survivor players are more likely to be reporting, as well as that of the kind of players that would more likely be reporting on the site to begin with, skewing significantly toward more experienced players (given that they are interested in tracking their performance, aware of the site, and go through the trouble of submitting), that would also be more likely to play SWF than just any random average player. So you have stats that depict results from survivor players on the more experienced side, with a likely correlating overrepresentation of SWF, against random killers. Far from a random sample.
Given all that, it's honestly surprising how close they tend to get to BHVR's stats. But there are still notable differences. It's unfortunate that we can't look at nightlight's stats in much more detail either, it would be interesting to see the rates of all reports combined, to filter by killer-reported, by SWF, killer by map, killer by perk(s), killer by add-on(s), look at disconnect rates, at how many 3/4k matches had one survivor player score significantly lower than the rest of their group, and so on.
For the larger topic of killrates, it's definitely lame that they are only releasing any once in a blue moon. While I don't think too much will have changed since last time, there are some killers that would be interesting to see after their changes. Getting "high MMR" stats again would also be neat, as well as killrates on the respective maps and most commonly used perks, also for separate MMR brackets. At least we have recently gotten one new statistical insight: Contrary to common claims, people killing themselves on hook is far from a relevant factor in the killrates, as even for Skull Merchant it barely made a difference.
More than look at global average killrates however, I would hope BHVR at least internally actually make use of their ability to look at stats in much more detail, such as killers (and their add-ons) that are most represented in the "players with 70-80+% killrate" pool, or survivor players (and their perks) that manage comparable escape rates, to identify some of the most problematic killers/add-ons and perks stacked in SWF. It unfortunately doesn't appear they really do however, as most of their balance changes seem aimed at making killers coalesce around that 60% global average target, not really addressing blatant issues that anyone experienced at the game would be familiar with (in both directions: some killers having things that are problematically strong, others weak, despite not having concerningly low or high global average rates). They seemingly don't even put too much stock in "high MMR" stats, and that bracket is also still much too broad to see some of the most egregious, extreme issues, that enable players to dominate and win most if not all of the time. Granted, the primary issue is obviously the matchmaking that creates these much too broad player pools, but that if anything only makes it all the more compelling to address extreme outlier issues, as things are already easy enough in the vast majority of matches as a good killer player or a good survivor player in an SWF group with other good survivor players. The respective problematic killer characters/add-ons and stacked survivor perks enable these players then to have ridiculous winrates, basically never losing. Those things are simply not needed. You can win the vast majority of your killer matches as perkless Legion if you are good enough at the game - you don't need Nurse with good add-ons, she can get nerfed; and you can win the vast majority of your survivor matches if you are good enough and in a good-enough SWF - you don't need to be able to stack 4 BNP, 4 SB, 4 DS, and whatnot.
2 -
are there anyone who cares about killrates at all? they have been proven wrong and inconsistent over the years, nurse having the lowest killrate and killers like freddy, sadako and skull merchant having the highest
0