I come to the conclusion than any killer that is lower than B tier are 'fun' to face

I saw recent statistics about survivors and killers. According to survivors, weak killers are considered fun to play against. Lmao. That just shows that survivors want their matches as easy as possible.

https://www.reddit.com/r/deadbydaylight/comments/18r7rbl/data_on_how_much_survivors_like_and_dislike/?utm_source=share&utm_medium=web2x&context=3

Comments

  • GeneralV
    GeneralV Member Posts: 11,072

    That isn't exactly true.

    For a very, very long time Hillbilly was considered the most fun killer to go against and he was far above the B tier.

  • GentlemanFridge
    GentlemanFridge Member Posts: 5,677

    There's a few rather striking outliers there.

    Clown and Freddy are hardly top tiers, yet are among the most disliked.

  • GeneralV
    GeneralV Member Posts: 11,072

    I don't like to see it, but I don't blame people for disliking current Freddy.

    Extremely boring power, after all.

  • Smoe
    Smoe Member Posts: 2,890
    edited December 2023

    I wouldn't put too much stock into these stats as they don't really mean anything significant since they are only covering the opinions of 1000 Survivors and nothing more.

  • HerInfernalMajesty
    HerInfernalMajesty Member Posts: 1,782

    I can’t say that they don’t mean anything significant. It’s not perfect data (which is near impossible to obtain) but 1,000 more than accounts for statistical deviations. 100 is actually the minimum number.

  • Smoe
    Smoe Member Posts: 2,890
    edited December 2023

    In comparison to the entire survivor playerbase or even just a quarter of the entire survivor playerbase? 1000 survivors doesn't really tell enough stats-wise.

  • HerInfernalMajesty
    HerInfernalMajesty Member Posts: 1,782

    For sure. But many stats derived from the population hardly ever sample even close to the whole population. Usually it’s just a fraction. Random sampling, (while not perfect) still accounts for deviations in data. It’s not the perfect picture but it often times tells us that something is going on which may or may not relate to the hypothesis.

  • HerInfernalMajesty
    HerInfernalMajesty Member Posts: 1,782

    The idea is that diversity of data is mostly evenly interspersed. So whether you get 100, 1,000, or even 10,000+ the proportion of statistical data outliers should mostly be the same. Greater data gives greater precision.

    Let’s say out of 10,000 people, 1,000 enjoy Apples over Bananas. In 1,000 people the number should be 100. In 100 people that number should be 10. Now of course there will be slight variations but overall the deviations that stick out should be proportional to the whole.

    Again 100 is usually the minimum accepted number. There are exceptions but I think the in this case 1,000 while not the most ideal, is still definitely good enough to show the outliers.

    It’s a question of why are there outliers.

  • Smoe
    Smoe Member Posts: 2,890
    edited December 2023

    Now of course there will be slight variations but overall the deviations that stick out should be proportional to the whole.

    If you compare 100,000 to 1000, it's going to be more than just a slight variation and the outliers could very much change entirely, especially since fun is subjective.

    100 might be the minimal acceptable number, but it still only covers an incredibly small portion of an otherwise much larger playerbase and the more people you cover, the more accurate the stats will become, after all you said it yourself ''Greater data gives greater precision.''

    To me, 1000 people only is simply too little an amount to be taken as anything concrete for any talks regarding changes to any killers.

    Also any stats done by the community should be taken with a grain of salt no matter what.

  • HerInfernalMajesty
    HerInfernalMajesty Member Posts: 1,782

    The thing is that outlier percentage should be mostly the same despite the group size.

  • Smoe
    Smoe Member Posts: 2,890

    So you say, but without comparing it to a larger number, we won't know that for certain.

  • HerInfernalMajesty
    HerInfernalMajesty Member Posts: 1,782

    We do have a smaller number to compare it to actually. In the Polls section if I recall the final two most fun Killers to go against were Hillbilly and Demogorgan. So despite a smaller sample size the proportions stayed relatively similar.

  • lav3
    lav3 Member Posts: 766

    Billy having 23% just proves most survivors want easy M1 toy killer.

    Also Trapper and Doctor more than 80% shows people hate killers even if they aren't good.

    Anyway it doesn't matter whether people like or hate which killer, killer mains will consider playing killers they like and enjoy even if they are hated ones from survivors.

  • NerfDHalready
    NerfDHalready Member Posts: 1,749

    um, no. peep oni, blight and wesker. even saying huntress, nemesis and demo wouldn't be a stretch.

  • C3Tooth
    C3Tooth Member Posts: 8,266

    Its actually all about Nurse making mistakes, thing is that makes you feel you juked her good, and feel fun to play against.

  • solarjin1
    solarjin1 Member Posts: 2,128
    edited December 2023

    billy a giga chad. The only killer that survivors will sometimes intentionally not drop pallets against! just to see can he curve it.

    so beautiful

  • ratcoffee
    ratcoffee Member Posts: 1,448

    The top 3 hated killers are Trickster, Clown, and Skull Merchant. The top 3 liked killers are Billy, Demo and Oni. Among other decently highly enjoyed killers are Wesker, Blight and Huntress.

    I'm curious whose tier list you're using to determine the power of those killers

  • MikaelaWantsYourBoon
    MikaelaWantsYourBoon Member Posts: 6,564

    Well that's my personal list. I don't care how strong or weak they are but my tier list here


  • DrDucky
    DrDucky Member Posts: 675

    I feel this is just not a correct conclusion, Wesker is up there along with Oni etc. Both are strong killers regarded as pretty fun according to those stats. Yet a killer like clown is widely disliked.

    Conclusions like this are not healthy since it just encourages more negativity where it is not needed.

  • Krazzik
    Krazzik Member Posts: 2,475

    I guess if you ONLY look at the 2 most popular killers to face, Billy and Demo, then yeah, they're both weakish killers who are easy to play around. However, Oni is a close third and is very strong, and Huntress and Wesker are pretty popular and both solidly strong too.

  • RentheKitty
    RentheKitty Member Posts: 5

    I think most of us survivors hear the music and have some kind of personal reaction. Then when we see what tier THEIR gameplay is... that's when we make our determination of likeness to dislike ratio. Haha. But I will always stand firm that no matter what, I'll hate a Clown. However, I can't REALLY hate on any pick because that's the contract reality you have when you search for a match. It's the killer's choice and I can't get upset. You just adjust your playstyle.

  • GeneralV
    GeneralV Member Posts: 11,072

    Ah, if only I had the power to change that...

    Freddy would have been fixed a long time ago.

  • Ayodam
    Ayodam Member Posts: 3,062

    They’re not actually wrong. Technically larger samples are preferable because such mitigate the impact of outliers. What groundbreaking research has used a sample size of n=20 & extrapolated its results to a population of millions? It sounds like it would be super unscientific and unethical.

  • squbax
    squbax Member Posts: 1,483

    Directly to point 1 yes larger samples mitigate the impact of outliers but it will not always do so, as adding more points to a distribution does not guarantee those points are gonna make the median representative as more data can add more outliers or data that randomly skews the distribution, hence why for example in population genomics we use an n= 20 because addin more data essentially just adds irrelevant data that does not contribute to the estimate in a significant way but makes the computational analisis way more unoractical.

    Now to point two, mostly if not all animal physiological neurology principles were determined by using even one single animal as contrary to what you mught believe is more ethical than opening the brains of the whole population of animals, studies with n=7 can be read regarding insect phisiology and if Im not wrong electric fish neurology was fone with like 14 fish, now extrapolating this data has been usefull even to date as the samples were correctly handled.

    Furthermore if you want an example in humans, when Pasteur developed his rabies vaccine it was used in a single person to test if it worked, and afterwards went on to save thousands of lives of children bitten by rabid animals, literally an n=1 that saved countless lives by extrapolating the result.

    However I myself am against extrapolating data in the medical field, but given we are discussing dbd, a sample of 1000 if its of random players is more than enough to give a reasonable estimate.

  • Ayodam
    Ayodam Member Posts: 3,062

    While larger samplings vis-a-vis population do not guarantee a change in outlier, they are still more likely to prevent data skew (outlier impact). That appears to be Smoe’s point—and he is right on that. 1,000 out of 700,000 is not indicative of any sort of larger trend.

  • radiantHero23
    radiantHero23 Member Posts: 4,132

    That these people just dont know anything.

    Pig should be 0% dislikes.

  • radiantHero23
    radiantHero23 Member Posts: 4,132

    Pls....

    Put piggy one tier higher.

    Do it for the Boop 🐽

  • squbax
    squbax Member Posts: 1,483

    Why is it not indicative, if its randomly sampled and the estimate is close to the real populations one even a sample of 700 or so is enough.

    I will make a final example that is actually used in real life, in order to check food safety, specially microbiologically wise you need out of a shipment to evaluate a certain number of samples, acording to the FDA and other microbiological analisis wanna guess out of shipments of food products that are in the hundred of thousands of units how many do you have to check to ensure the product is safe in the cases of seafood, which is specially delicate and has to be checked with more samples, you need to check 5 units out of each shipment.

    Because contrary to what you may believe optimizing sampling is important in order to make studys practical, you keep saying that you need more data, that while true doesnt mean that the current data we have is not enough or not representative of the population, it might be skewed dont get me wrong but you have given 0 arguments on why it is not representative.

  • SunaIIanu
    SunaIIanu Member Posts: 825

    The number ist fine, but the sample might not be representative for the whole dbd survivor community because of the way it was collected.

    It seems like Killa Whale collected answers over the YouTube community tab and the survey was in english, so I would assume that most participants are probably based in North America or Europe.

    Also, people who play dbd very casually might not interact with dbd content, so the sample might be biased towards more serious dbd players.

    I obviously don't know if this matters, but as far as I know, playstyle and Meta builds can vary between regions, so I think it is not implausible that opinions about Killers might vary aswell

  • I'm getting real tired of these posts.

  • Eleghost
    Eleghost Member Posts: 1,190

    Probably depends. You can do some nasty stuff with some killers that you can't with others. Like Sadakos iri tape makes her super unfun to deal with, pigs slow search scream builds are horrendous Myers is kind of bad with gravestone piece etc. The ones i mostly hate dealing with are Blight or ghostface, but it's usually other survivors that are worse to deal with than the killers.

  • Bartolomeo87
    Bartolomeo87 Member Posts: 35

    Yeah you’re essentially correct. Any killer that can put up a fight against the higher level of player is almost immediately considered “unfun” to most survivors. They want to be able to hold w and pre-drop god/strong pallets for an easy W. Hell this even applies to lower tier killers when you play them well. A survivor will love playing against your killer up until you actually do well vs them and all of a sudden you weren’t fun to play against lol