Visit the Kill Switch Master List for more information on this and other current known issues: https://forums.bhvr.com/dead-by-daylight/kb/articles/299-kill-switch-master-list
Get all the details on our forums: https://forums.bhvr.com/dead-by-daylight/discussion/436478/sign-up-now-to-receive-a-recap-of-your-2024-dead-by-daylight-stats/p1?new=1
Killrate, Winrate and game balance
So. Because certain topics went into the course of killrates and winrates i came into conclusion that new topic is needed.
This is not a suggestion tho it will contain my suggestion of how to balance things. Main point of this topic is to discuss over what feels fair in terms of team balance. And as much as balanced players opinions are most valuable, because they know each side more less the same, all opinions are valid since its not a suggestion thread but a discussion one. So killer players, survivor players and even entitled players can be heard.
So. Current devs aim is around 60% killrate. Some people may call it being killer sided but truth is that its not as simple.
Statistics is about big number, and statistically 60% killrate means that every one survivor has 40% chance of escape, and 2.4 survs in every match are sacrificed. But this also means that killer wins 40% of the time with rest 60% being draws/loses. Idk how devs calculate hatch escape for killer, i would consider it as 4k for better statistics, and aim more for 57-59% killrate, but 40% winrate for every player is quite fair tbh.
But this is because dbd is not team game. Survivors are competing against the killer but also with themselves - even if killer kill only 1 survivor and lost a match, this still means that at least one survivor lost along with the killer.
For more balanced and good feelings, i would AIM for 50% winrate for each TEAM but this would require survivors being a team, yet since every survivor can loose before others fate is decided its impossible. We would need to remove one of core dbd gameplay features - hook sacrifices.
I have an idea of making trials having timer.
Killer need to hook, mori, hit, and do whatever they can, to interrupt survivors objectives.
Of course objectives would need a balance pass.
Hooks would last a minute with ability for being unhooked, but after a minute, hooked surv would be removed from hook and placed at least 36m away from killer.
Mories would put surv in ghost mode, where they can move freely, but cannot take conspicious actions, nor have collision, and may or may not be visible by other survivors. Mori lasts for 1.5 minutes.
If survs manage to do their objective their whole team won and escaped.
If survs fail to do their objectives, they are sacrificed to the Entity, and they are dead (did i mentioned that trials should be at night, and timer should count down to dawn?) by daylight.
Comments
-
Its actually simple. 5 matches, 12 kills, 8 escape.
You can list all the possibility. The win rate is ALWAYS higher than lose rate.
3 -
and statistically 60% killrate means that every one survivor has 40% chance of escapeSure, that part is true. Over many games each person playing survivor should escape 4 out of 10 games.
But this also means that killer wins 40% of the time with rest 60% being draws/losesAnd this part is not true.
This is only true if you presume each possible outcome (0k, 1k, 2k, 3k, 4k) have an equal chance of occurring. But they don't. DbD's outcomes are not evenly balanced, they aren't even a bell curve, they are skewed toward the extreme edges (an inverted bell curve would be a closer model that is then dragged to the killer side). 2ks are especially unlikely meaning that the higher kill rate also has to translate into a higher win rate.
Everything else in the post is about game direction, but that's a different discussion. The presumptions on the equation you are constructing are incorrect.
4 -
I think the game struggles with prioritizing the destination over the journey. If I'm at the target escape rate on survivor but half of my matches feel utterly pointless, the kill rate target or matchmaking needs to be examined.
A 40% escape rate in the random free for all matchmaking we have feels a lot worse than a 40% escape rate in a more competitive queue would feel. Like, it's cool that you have this kill rate goal, but half of my matches feel like a complete waste of time to get there
5 -
In 3, there should be 40% win, 40% loose :)
Do the same for 50% killrate :D0 -
Here we go again…
Statistics is about big number, and statistically 60% killrate means that every one survivor has 40% chance of escape, and 2.4 survs in every match are sacrificed. But this also means that killer wins 40% of the time with rest 60% being draws/loses.
This model is inaccurate. It does not account for the distinction between 0Ks and 1Ks, nor does it consider 3Ks and 4Ks.
If you take this model and dip the killrate down to 50%, you'd see this immediately, as the model predicts a 0% winrate on 50% kill rate. Meaning that the model thinks it is impossible to have two matches, back to back, where one is a 4K and the other a 0K.
This model is not possible in theory, and is, in practice, disproven by Nightlight. (And before anyone interjects that Nightlight is not reliable: That is not relevant here. Unless Nightlight's data obeys entirely different rules and is recording outcomes that are impossible, such as one trial having a 3K and 2E at once, the model is supposed to work on it.)
4 -
We have massive amount of games where Survivors throw the game at the end and Killer instead of 0 kills can get 2 or 3 and more…
Typical Average Bubba will get 2 Hooks for entire game, but at the end he is gonna get 2 Kills or even more - just an example.
Kill Rate isn't a good statistic for balancing the game. With better End Game mechanics / some rework, maybe it would be better, but at current state it doesn't show anything.
1 -
You don’t understand probabilities and statistics enough to make this statement. Most everyone doesn’t,.
My suggestion is use nightlight.gg to get your numbers instead of trying to generalize the statistics and probabilities, because that doesn’t work at all.
If you use the stats from nightlight.gg, you will see killers generally have a 60% win rate as well as a 60% kill rate.5 -
60% kill rate does not mean 40% winrate for killers. A killer can win anywhere between 20-80% of matches, they can draw anywhere in the range of 0-80% of matches, and they can lose anywhere between 0-40% of matches with a 60% kill rate. The range of possible outcomes is massive. The fact that such a wide range of possible outcomes exists is the exact reason why you can't assume the mathematical average of those numbers applies to individual match results
4 -
I don't think that is too often expect for bubba but killers throw games too. And killer can simply play well in endgame too and with noed winning 3vs1 is not too hard. Everything should be taken account ofcourse.
0
