Visit the Kill Switch Master List for more information on these and other current known issues: https://forums.bhvr.com/dead-by-daylight/kb/articles/299-kill-switch-master-list
We encourage you to be as honest as possible in letting us know how you feel about the game. The information and answers provided are anonymous, not shared with any third-party, and will not be used for purposes other than survey analysis.
Access the survey HERE!
Why the DBD ‘stats’ are useless garbage
I’m a data analyst. Just sayin…
These stats are non segmented. All player skill levels, experience levels, casual, competitive, all rolled up. In something where experience and skill matters, this vastly muddles the info for any good or better player who wants to know what good or above players do. It’s too tainted with data from new and/or bad players.
Also… huge drop in experienced survivors playing lately so this is based on a pool that is likely much more heavily weighted by weaker survivors.
Also…
Matchmaking is so broken that solo matches are basically all random at this point. I’m a (I guess what I’d have to call a ‘once high MMR / ‘red rank’’) survivor that has been soloing forever and constantly loses from bad teammates, I lose way more than I win because soloing is awful and this is true for everyone. So you have tons of ‘once high MMR’ survivors who have been mashed down into the depths of low to average survivor MMR and low to average MMR survivors artificially brought up a bit from playing on teams with the ‘once high MMR’ survivors. They probably have 80% of solo survivors all in roughly the same MMR range. So yeah… that terrible matchmaking is going to ruin your games and also produce ######### data on how games (supposedly correctly match made games) are going.
FInally… Realize this: Most solo teams are probably like 1 good to great player, 1 good player and 2 average or bad players. (ON AVERAGE). So figure this is an average team. A bad solo team will be tend to have more bad players and a good team will have more good players. A great solo team will have gotten lucky and managed to get a few or all great players. Because of how bad matchmaking is - Bad killers are beating teams average and below. Average killers are beating teams good and below. Good killers are beating great and below. Great killers are beating teams great and below like 99% of the time. And MOST solo teams are average. Meeting a randomly match made solo team of all great players is a holy grail. It ain't happening.
These stats are taking none of this into account. The announcement is their useless review of bad data. If I ran a company I wouldn’t release it. And I have to assume they can’t segment it out because of what I said about MMR… they have no useable, accurate MMR segments because they’re rendered garbage by the way they determine wins and losses.
Comments
-
They weren't making a statement by releasing the data. It wasn't meant to be useful or have something big taken out of it, it was just so people could see some stats, because people like seeing stats.
20 -
Sorry but is this defending them? “It wasn’t meant to be useful”. Releasing bad data is bad. Good companies don’t do it. Ever hear the phrase “a little knowledge is a dangerous thing”? It makes people think they have something to talk about when they actually don’t.
9 -
It was a message for all players and the stats reflected that.
0 -
You say it's bad data because it doesn't fill in the missing quotient of your grandiloquent equation of the True™ Statistics®.
I say it's good data because I think it's funny how much Self Care is being run.
Maybe you need to step back from the chalk board.
17 -
The fact that self care is being run a lot seems to suggest the data is filled with inexperienced players. Yet, back in the day when we had ‘good’ data we learned red rank players were running self care more often that the average player would think. So… what does this data really mean? Who tf knows.
6 -
Who knows?
Who cares?
It's fun to see and speculate about. It's not gonna shake up the way we see the game, it's just interesting. Even if it's useless.
11 -
I have no idea what's got your tizzy tumbling, but you're fun to talk to.
Have a good rest of your night.
Post edited by Rizzo on8 -
I don't understand the number of people that feel the need to either twist these stats into something the company never claimed they said, or people who just outright claim the stats are a lie (lol). Why do so many of you play a game that drives you this crazy?
8 -
You just defined the forum.
1 -
I quit dbd a while back because the forum is actually more entertaining.
5 -
Because they don't bother showing us something useful.
0 -
There is no such thing as bad data, only bad interpretation of data.
4 -
I think the OP is accurate. It's not particularly useful data.
I also think @Grandpa_Crack_Pipe is accurate. It's fun to talk about it. DbD is entertainment after all.
I'm not even sure if BHVR even pays attention to good data. Like which killers get DC'd against the most.
We already know they base their perk balancing decisions on usage rather than actual balance.
Self Care will always be a new player favorite and see above average usage. It's available for free early on and new players see it and think "Hey I can heal myself without a med kit, that's good right?" and they will run it until they figure out or are told it's bad.
4 -
haha what?
0 -
You're right from a standpoint of being professional with statistics.
And you're wrong at the same time, and you know it. If someone wants to make a decision and needs an argumentation aid, he can cut statistics in such a way that they support the decision.
I'm sure that if BHVR broke down the numbers into ranges that there would be an outcry in the community. That's why they don't publish them. And that way they can decide what they want without regard to facts.
And you know what? It's just a game, this is not a matter of life and death. Statistics falsification in areas where human lives are at stake, I find much worse.
1 -
Good intentions but as you know if you were on the forums that day it just lead to a bunch of people feeling uber justified in themselves and complaining using these stats as "evidence" for their claims. Almost as bad as people calling the BP bonus mainly being on survivors "evidence"
0 -
I agree that those stats don't say much. They are much too agregated.
The devs have actually said that they do have extremely detailed interactive stats spreadsheets, adjusted by mmr and character and all things nice, but that all that would be way too complicated to release in an easy to read format such as their typical communication.
Hence : the agregated mess :)
1 -
Eh. I'll go out on a limb and say 99% of those people already held those opinions, they just wanted something to point to to be loud about it.
It didn't really change many minds. It just made a couple people more annoying.
5 -
Also… huge drop in experienced survivors playing lately so this is based on a pool that is likely much more heavily weighted by weaker survivors.
There’s actually no real evidence there’s a huge drop in experienced survivors, at least not to the extent you’re implying where data is “much more heavily weighted by weak survivors”.
Matchmaking is so broken that solo matches are basically all random at this point. I’m a (I guess what I’d have to call a ‘once high MMR / ‘red rank’’) survivor that has been soloing forever and constantly loses from bad teammates, I lose way more than I win because soloing is awful and this is true for everyone. …
Matchmaking isn’t as broken as you’re claiming. The MMRs themselves seem to be fairly good estimators in that when the MMRs are close the match is close. This was evidenced for instance in the tests they did where when the matchmaking was strictest on MMR pairings the matches were “sweatiest” and when the MMR was totally ignored or the system used Grades the matchmaking was much more random than it is normally.
Also most players probably are in the middle section of what would be a bell curve for the MMR ratings, that’s entirely normal. When we took some rough guesses the best fit was about 6% of players are over the 1600 soft cap based on a normal distribution with mean 1000 and standard deviation 400. Peanits has also commented that the percentage of people over the soft cap is small, although he didn’t know the exact number it’s not anywhere close to the 20% figure some people are throwing around (which seems to stem from some bad math.)
These stats are taking none of this into account. The announcement is their useless review of bad data. If I ran a company I wouldn’t release it. And I have to assume they can’t segment it out because of what I said about MMR… they have no useable, accurate MMR segments because they’re rendered garbage by the way they determine wins and losses.
The devs literally said the numbers are just rough overall snapshots of the entire playerbase and that they don’t use them for internal decisions but have much more detailed dashboards where they can drill into finely segmented cross-sections of data.
P.S. I’m also a data analyst, just sayin.
3 -
While thats certainly true, it also makes it harder to change minds, when they have incomplete data they can use as a tool to try to warp things to their perspective. Its kinda like enabling a bad habit.
0 -
If I'm gonna throw out a bit of a hot take-- almost no one here will change their mind to begin with. About miniscule things, maybe, but larger things are set in stone. Doesn't matter how much "evidence", good or otherwise, is presented.
It's probably a little too cynical to say, but people are on forums for validation, not discourse.
2 -
Its not necessarily wrong, and i wouldn't even consider it a hot take, but I have seen people both having civil and constructive discussions, as well as perspectives change about even core mechanics themselves. Even though there seems to be less and less direct feedback, i like to think that there are people on the dev team who also understand and appreciate when valid points are discussed rationally as well. The vast majority is as bad as you say though, to be sure.
To this day I still think that many people are simply lacking perspective in regards to a lot of issues, especially the ones that don't affect them directly. Its always a shame when you see two people who are generally level headed get caught up on two details that aren't even mutually exclusive of one another.
1 -
The only objective of these stats is to show how much survivors are escaping and how much killers are killing. There's not much more to it. They're not "useless garbage" because you can't pull out a 5 page essay about game balance from it, Mr. Data Analyst lmao.
0 -
tbh my main reason for hating them is because of how many people keep using them as justification for their "boycotting" excuse for why they ragequit games and leave me trying to carry a 1v3 in half my games. They're incomplete data due to the amount of factors they ignore, but they become dangerous incomplete data when they fuel people's self-justifications for ruining the game for others and themselves. I'm sure quite a large number of people who agree with me do so for similar reasons.
0 -
If those stats are "useless garbage" for killer mains, then you will be ok if DEVs would strive for 80% escape rate and 20% kill rate. I'm fine with that too.
DEVs, give us 80% escape rate, killers don't care for the stats anyway.
0 -
Your assessment is extremely one sided so i can see why you'd imagine anyone that disagrees must be the opposite. If people are seeing games as hopeless that early in, they should probably not be playing the game to begin with, instead of ruining more games for others and, in turn, making them share their hopelessness.
The fact that you brought up a magic number in your post shows that you would rather use it for an agenda than explore its context, cause, or solution. You don't even care if that number is too high or low, just that you can use it to try to make an impact. The entire point of this topic is about how that data is missing context that is directly related to its result, which is something that the devs have to explain every time they provide them.
0 -
The fact that their data since the patch shows people are running self-care, shows that putting on a challenge that REQUIRES a single perk for you to escape with, will inflate that perk's usage lmfao
0 -
I still think most challenges are self-sandbagging in general, especially the "within one trial" ones. People are more likely to throw to accomplish them and then just give up because they got it, and the ones that add requiring an escape somehow make that even worse. I like the idea of challenges, but a lot of them are hurting your teammates as much (if not more) than you personally.
0 -
Spot on! If you pay attention to the difference in challenges between killers and survivors, you will notice that most survivor challenges are self/team-sabotaging.
Fall from a great height whilst in chase, get hit whilst vaulting a window, hide in killer radius for 100+ seconds, stun the killer..
It's all self-sabotaging because most players won't play seriously
0 -
A lot of killer ones require self sandbagging as well, but at least you're the only one who suffers for it (unless its one that requires scummy play to become reasonable, but thats a different topic entirely) and the other players often benefit, due to the 1v4 nature making that somewhat impossible to achieve parity. Some of the survivor ones are definitely more harsh though, which always led me to believe they were an experimental attempt at trying to get the community to "balance itself" so to speak.
0 -
They have already stated on other threads that regions OUTSIDE of the US use Self Care, like all the time. That is why the pick % for Self Care is so high.
1 -
I'm pretty sure i remember one of the devs saying on the forum that the stats are from prior to that challenge
0 -
It is so comedic you say US, as if everyone on the forums is from the US. I'm from Europe, I play in European regions, I hardly see self-care.
0 -
Just to clarify I’m pretty sure it’s players in Asia driving the Self Care number, for whatever reason it’s apparently really popular there.
0 -
Didn't he dev post say Self-Care saw more use in Asian regions and that's what gave it the number 1 spot?
My understanding was that was a factor in its high usage in the first place.
0 -
Thats alright, just found it odd to automatically assume its ppl everywhere but US
0
