DBD is slightly killer sided (and that's great!)
Comments
-
Hey now, you r insulting my name, and inner healing is honestly really good atm imo
2 -
Let's take your example of 100 matches again. Killer plays 100 matches, gets a 3k in 80 of them and a 0k in the remaining 20. This comes up to 3 * 80 = 240 survivors killed, just as in your example. However, the killer win rate (by the metric you chose) is 80%, far off from the 60% kill rate.
Now, a second example of 100 matches. Killer gets a 4k in 40 matches, 3k in 10 matches and 1k in the remaining 50. In this case, the killer has again gotten 240 kills across 100 matches for a 60% kill rate, but has only a 50% win rate.
The point here is, if youre going to use a metric like "3k+ is a win, 2k is a tie, 1k- is a loss" you can't use kill rate stats to figure out how often killers manage to accomplish that. You would need to look directly at the stats for how common each result is. And what do you know, Nightlight has that stat:
Killers win 46% of the time and lose 40% of the time - their win% is 5 percentage points higher than their loss%. Survivor win% is 5 percentage points below their loss%.
4 -
Do you understand what an average is? It means that some killers would have higher kill rate than 60%, and other would have it lower.
That is why it is an average, and what you said don't contradicts the fact that 60% kill rate means 40% win rate for killers and 40% win rates for survivors on average. There would be also some survivors that escape and win more times than other survivors do, but those didn't fit your narrative there.
And yes, you can use the MMR metric to calculate this as it is the reason of the MMR working that way and devs wanting a 60% kill rate.
5 -
Precisely because of what you are describing there using flat statistical averages percentages is the only way to do it. Because even if some killer players are able to do a 3-4K every game and others would only gen 0-1K every game, the average would be around 2 survivors killed per game still, so, around 50%.
That's why it is an average, and that is why it represents better how many times each side wins.
1 -
I'm not sure what you mean by "you can use the MMR metric to calculate this." MMR has literally not come up in this discussion until you mentioned it just now
in any case, i'd love you to explain how increasing kill rates from 53% to 60% would decrease killer winrates from what they are at 46% right now to just 40%
Post edited by ratcoffee on0 -
the killrate is not the winrate for killers. it is the rate at which survivors do not escape. if the average chance to escape is 40%, this just means the average killrate is 100% - 40% = 60%
The actual chance for a killer to win is the sum of the chances to 3k and 4k. which is 47.52% (4*0.4*0.6^3+0.6^4 = 0.4752)
The killer's chance to win is actually just 1.188 times higher than any survivor's chance to escape.
3 -
You say you dont care about made up win conditions but come up with one yourself.
0 -
Yes, but your conclusion was wrong.
53% is the average, but it doesn't mean nearly every game is a 2k and killers are never winning any games. The conclusion is what I was contesting.
I don't disagree that we don't have the data to know exactly what the distribution of reach outcome 0k -4k is, but you can't conclude that every game is either a 2k or 3k based on just the average. You can't work the numbers backwards that way.
1 -
i still see 2k's as losses because it means that survivors powered exit gate. on average, survivors power the exit gate 53% of the games. this number is raw value but you have to consider how many survivor power exit gate then also die in end game randomly and give killer's 3k's or 4k's. I would not be surprised if survivor power exit gates for over 60% of their games.
if looking at 50% kill-rate killers. let's look at freddy who is only 0.7 off from perfect 50%.
if you add 4 escapes, 3 escapes and 2 escapes, survivors power exit gate around 59% of the time. So a 50% kill-rate killer is winning games at around 41% frequency.
there are barely any killer's at 60% but closest one is this killer right here who wins around ~57% of the time on 3k and 4k's and survivor power exit gate roughly 43% of the time.
killer player want killer to be at around 60% ratio while survivor want 50% kill-rate killers. at the same time, kill-rates are statistics and they do not factor in the player's skill or micro elements of the game. it is all averages.
0 -
Even the stated goal of 60% kill : escape by BHVR is survivor-sided. A TRULY balanced game, by BHVR's own terms, is 50% 3K+ games, 50% 2K- games, which would equal 62.5%.
0 -
This is simplified though, and assumes draws make up an equal share. But out of those 100 matches, you can also get 60% kill rate with 40 0Ks and 60 4Ks, in which case the killer has a 60% winrate, in line with the kill rate.
We don't know how high the contribution of 2Ks is to the kill rate, so it doesn't really work as evidence of survivor-sidedness.
3 -
Well I see 2k as a draw so agree to disagree I guess
0 -
Why do people act like the “average” survivor player in this game is some massive potato running in straight lines, not looking behind them and not capable of completing skill checks on generators? This explains some of the awful balance ideas people have.
There’s definitely a weird dichotomy, or maybe it’s just outright gaslighting, where the average killer player is viewed as decently skilled but the average survivor is considered trash. Newsflash: at a certain point most survivors are capable of looping to some extent and if they’re not they’re at least using windows for autopilot mode and are able to waste a decent enough amount of the killers time. Anyone with half a brain is now able to monitor the survivor HUD and use that information to their benefit. It’s really not difficult. Why are people making out that only high MMR (whatever that means) survivors know how to do this stuff?
2 -
I mean that it is the MMR what dictates what is a win or a lost for both killers and survivors. A killer only wins with a 3K.
And for your explanation, starting with Nightlight not being a reliable source, as it depends on their users uploading their data to the page and as right now it only counts 22637 matches... again: Averages. Averages and balancing.
In a game like DbD, if you want every killer to have the same chance to win between them and the other side, first you need to make sure that the mathematical equality is achieved. In other words, you first need to get the average to 60% kill rate, which would mean 2.4 survivors per match killed, and then balance individually those killers than can do more or less than that. That's why the devs wanted the kill rate to be 60%, and that's how you get a 40% for everybody. There would still be players that can get more or less wins on both sides as skill is obviously another factor for this, but on average everybody would still have the same chance to win on that ideal scenario.
But the best part of this is that you yourself have proved than in any case, according to Nightlight, the game is slightly survivor sided and not killer sided, as survivors are winning 47.3% of their games, and killers only 46.04% on average. In other words, you are still wrong.
2 -
actually it was like that untill MFT came out and fast vaults got broken alongside the cage patch IMO.
3 -
I mean that it is the MMR what dictates what is a win or a lost for both killers and survivors. A killer only wins with a 3K.
Is this confirmed? Because from what I recall of the devs' explanation, the MMR calculation actually also factors in who you get a kill on, and how many kills you've already had. It may actually be possible to get an MMR increase off of a 2K if it's not a flat value per kill.
In a game like DbD, if you want every killer to have the same chance to win between them and the other side, first you need to make sure that the mathematical equality is achieved. In other words, you first need to get the average to 60% kill rate, which would mean 2.4 survivors per match killed, and then balance individually those killers than can do more or less than that. That's why the devs wanted the kill rate to be 60%, and that's how you get a 40% for everybody. There would still be players that can get more or less wins on both sides as skill is obviously another factor for this, but on average everybody would still have the same chance to win on that ideal scenario.
I think you're missing @ratcoffee 's point though. Mathematical equality does not stem directly from the kill rate. 60% kill rate -could- lead to an equal 40%-40% winrate between killer and survivor (As a team), but it could also consist of a bunch of 2Ks and a few 3Ks/4Ks. In which case survivors (As a team) would have 0% winrate because they'd never get a 3E or 4E.
Your calculation of the 40% winrate only works with perfect distribution of the match outcomes, but DbD does not have that. The distribution is wonky and will vary per killer, too.
But the best part of this is that you yourself have proved than in any case, according to Nightlight, the game is slightly survivor sided and not killer sided, as survivors are winning 47.3% of their games, and killers only 46.04% on average. In other words, you are still wrong.
The problem here is that there's two metrics for survivors and you're using them interchangeably. Survivors exist both as individuals and as a team. Survivors, as a team, have an average win rate of 40.86% vs. Killer's win rate of 46.06%, while individual survivors would have a 47.3% win-rate versus Killer's 52.7% kill rate.
But if you pull it out of numbers and into practice: Do you think anyone familiar with DbD would consider a survivor to have won if they are the only person to escape, while the killer walks away with a 3K? Or would they say the survivors lost and the killer won?
At the end of the day, according to Nightlight's statistics, killers are winning more than they are losing, and they are killing more than survivors are escaping. Pitting the correct statistics against one another, there isn't really any particular metric by which survivors come out on top.
3 -
22k samples is more than enough to form a decent conclusion. As you've been very insistent on talking about averages, I'm sure you're familiar with the Central Limit Theorem, which would back up using 22k matches as a representative sample size.
The survivor escape rates also include all the hatch escapes. If you insist on using the believed MMR win/loss conditions to gauge balance (which I think is flawed but that's beside the point) you have to adjust for the fact that the MMR system does not count the hatch escape as a win for the survivors but counts it as an escape in survivor stats
Looking at the Left for Dead achievement, only 8.2% of people have powered the exit gate as the sole survivor and escaped, ever. That number's going to be far less in one month. However, I'll be generous and assume that those numbers don't account for games where the gates were powered with 2 people left, but only one got to walk out. I'm willing to entertain the idea that as many as 16.4% of 3k games end that way - twice the percentage of people who have the Left for Dead achievement. Doing the math on the data we referenced earlier, it seems normalizing for hatch escapes brings the survivor winrate (according to your metric) down to around 44% (actually less than that, but I'm being generous with rounding)
Post edited by ratcoffee on2 -
Is this confirmed? Because from what I recall of the devs' explanation, the MMR calculation actually also factors in who you get a kill on
The only thing that affects MMR apart from pure win-draw-lose scenarios is the MMR of the involved players. For example, if you escape of a killer with higher MMR than you, you would win more points that escaping from someone with your same MMR. As far as I know, draws are not affected by this (and if you have any source of it, it would be nice if you post it).
The other thing that can affect MMR according to the Wiki (as the other official sources has being deleted) is when being in a SWF, where it would reduce or increase the awarded MMR points depending on how many members of the squad dies or escape. Again, reduce or increase, it doesn't change the fact that you would still gain or lose points if you escape or get killed.
I think you're missing @ratcoffee 's point though.
I didn't miss his point, and he asked me to explain how a 60% kill rate could lower the win rate and I did explain how. So, I already answered you: Individual killer balance. But for you to do that, first you have to get to the base for it precisely because what you are saying, this can vary from killer to killer so you need a base to start balancing each one of them.
In other type of games this could be done differently, like in fighting games where they basically pick one character with averages stats (usually, the first one made for the game) and balance the other fighters against that "average". You can't do that here for obvious reasons, so the only "average" in what you could base your balancing for is in kill rates (in other word, 2.4 survivors killed per match). Again, this is the reason of why BHVR wants (or wanted, who knows) a 60% kill rate on average across the board.
The problem here is that there's two metrics for survivors and you're using them interchangeably. Survivors exist both as individuals and as a team.
As I already said, this is false. The devs took their sweet time in the day while explaining MMR to make it clear that even if the killer is playing a 1v4 match, each survivors is playing their own match, so for them in reality is a 1v1/1/1/1. Even if being in a SWF squad affects the amount of MMR gained or lost, if you are the only one escaping then only you won your match against the killer, while the other 3 survivors lost theirs.
At the end of the day, according to Nightlight's statistics, killers are winning more than they are losing
Again, false. And we don't have anything to put into practice here as again, how the MMR works marks what is a win, a draw and a lose for both survivors and killers.
A survivor win only by escaping through the exit gates, draws if he escape through hatch, and lose if he get killed. Ergo, 52.72% kill rate = 47.28% of survivors winning.
A killer only wins if he do a 3 - 4K, draws with a 2K, and lose with 0-1K. So, killers only wins 46.07% of their games, as that's the percentage of matches they are doing 3Ks or more according to Nightlight.
In conclusion, survivors are winning 5.44% more matches on average than killers, and killers are losing 53.93% of their games. Simple as that.
3 -
Feel like this entire argument about "winrates" is off-whack.
Anything under 3k implies the survivors finished their 5 gens and were on their way out, a killer can force a 1-2k in endgame by just camping, so right off the bat, even though 2ks are considered "ties" they are, lets be real, a killer loss since the survivors popped all 5 gens and left.
With the previous statement in mind, killers get a win (3K +) about 46%~ of the time , which in turn means the survivor team win rate sits at 54%~
Survivor is a team role, with a goal of 60% killrate, some people are *expected to be sacrificed* , but that still means your team effectively won.
2 -
22k samples is more than enough to form a decent conclusion. As you've been very insistent on talking about averages, I'm sure you're familiar with the Central Limit Theorem
Your problem here is that you have 33 different killers with different abilities and usability to measure. 2785 of those 22k matches are just for Alien. 610 of those are for Trapper. I think we can agree that not only those two killers are at least a couple of worlds apart, but that those separated numbers are not representative of the total games played with any of those killers, so the average of those numbers is also not accurate.
Also, the data of those 22K matches was the last 28 days of matches uploaded by the page users. Assuming a player would play at least 2 matches before disconnecting, with average number of players only in Steam right now being 30,248.2 players per day in the last month, and assuming those players have a perfect ratio of 4 survivors : 1 killer, would mean that you are missing at least 316,779 matches of data in those 28 days only in Steam. 22k matches is not even close enough to have a glimpse of the big picture.
The survivor escape rates also include all the hatch escapes. If you insist on using the believed MMR win/loss conditions to gauge balance [...] you have to adjust for the fact that the MMR system does not count the hatch escape as a win for the survivors but counts it as an escape in survivor stats
You are technically right, but as far as I know, the hatch is still considered an escape on the stats because even if it is neutral MMR for the survivor, it is still an optional free escape survivors have that gives them all the same conditions in game as if they escaped through the exit gates. That's why it is counted inside that "40%" win rate, as technically they have still escaped.
Looking at the Left for Dead achievement, only 8.2% of people have powered the exit gate as the sole survivor and escaped, ever.
Ever, on Steam. Unless you have more info that I could not find of all the other platforms, that is not representative of anything. Also, you have to take in count that first, 2.4 survivors killed per match on average means, again, most games ends in a 2K, and just a few in 3K, so hatch won't be as much of an occurrence on average (even calculating it with the Nightlight data, only 33 of 400 survivors in 100 matches would have the hatch available).
Second, and again, the hatch is an optional free escape. You can still go for the exit gates if you want, but most people obviously prefer to get the easy escape. If you ask me, they should get rid of the hatch and change the EGC all along, but that's just me.
1 -
I get confused when people say killer is too easy these days. They must be playing the top tiers or something cuz' man, gens still fly like crazy unless you're playing your 120% against these cracked players.
And on survivor side, yes SoloQ is bad but I didn't play the game expecting to survive every match. Killer SHOULD be the intimidating role, which is why I dropped using meta and MFT/Haste combo because it trivialized the gameplay loop.
3 -
You can point to specific examples for both killer's and survivors.
What about made for this, what about ultimate weapon?
What about coordinated SWF, what about overturned killer adorns.
There are specific examples for both but overall the game is balanced.
0 -
The only thing that affects MMR apart from pure win-draw-lose scenarios is the MMR of the involved players.
I am pretty sure this is also incorrect though, as I recall the devs mentioning that there is MMR amelioration based on the current state of the game. IE: The first kill is worth more MMR than the fourth.
But also, I don't know why you're trying to argue about kill rates when you want to talk about MMR. The latter system is very opaque and not directly tied to kills. It's well possible that a killer can go up in MMR on a 2K, in which case 2Ks wouldn't all be draws but would be distributed amongst wins and losses, making the stats fail to support -any- argument.
I didn't miss his point, and he asked me to explain how a 60% kill rate could lower the win rate and I did explain how. So, I already answered you: Individual killer balance. But for you to do that, first you have to get to the base for it precisely because what you are saying, this can vary from killer to killer so you need a base to start balancing each one of them.
Increasing kill rates would not lower the win rates, though, would it?
Adding more kills into the mix of stats would increase the likelihood of 3Ks and 4Ks, thereby increasing killer winrates. If the current killer winrate is over the value you projected (40%), then increasing killrates would increase the winrate further.
You can't do that here for obvious reasons, so the only "average" in what you could base your balancing for is in kill rates (in other word, 2.4 survivors killed per match). Again, this is the reason of why BHVR wants (or wanted, who knows) a 60% kill rate on average across the board.
You have already been explained why this doesn't work though: DbD does not have a linear distribution of kills among games, which means kill rates don't directly translate into win and loss percentages. I know you want that 40% winrate to follow from the 60% killrate, but it doesn't. As proven by the Nightlight stats giving a higher win% at a lower kill%.
As I already said, this is false. The devs took their sweet time in the day while explaining MMR to make it clear that even if the killer is playing a 1v4 match, each survivors is playing their own match, so for them in reality is a 1v1/1/1/1. Even if being in a SWF squad affects the amount of MMR gained or lost, if you are the only one escaping then only you won your match against the killer, while the other 3 survivors lost theirs.
It feels like you're arguing three different things at the same time: Wins as team, wins individually, and wins by MMR.
We do not know wins by MMR, simply because the system is too opaque. So there's no argument to be had here.
We do know wins by team, and with killers scoring more 3K+ than 1K-, we know survivors aren't winning as a team.
We also know wins as solo, and with 47% survive rate, survivors are losing more than they are winning.
The only argument by which you could theoretically state that killers are at a disadvantage is that an individual survivor has a sliiightly higher chance of surviving than a killer has to get a 3K+. And I don't think anyone reasonable here is going to use this as grounds to call the game 'survivor-sided'. But, there's another problem that you yourself have now indicated:
A survivor win only by escaping through the exit gates, draws if he escape through hatch, and lose if he get killed.
And that's the hatch counting as a draw. Which means that we can no longer take the 53% kill rate to immediately translate to a 47% win rate for survivors, because a portion of that 47% will be hatch escapes.
These must necessarily come from the 3Ks. If even half of these 3Ks are hatch escapes (And I think we can safely assume that it'll be more than half), then the survive rate drops to 45%. Under the killer's winrate of 46%.
At which point the stats are telling us the following:
1: We don't know MMR.
2: Killers are more likely to win than survivors are, when survivors are counted as a team.
3: Killers are more likely to win than survivors are, when survivors are counted individually.
The stat distributions just don't support the argument that survivors are favoured.
1 -
I am pretty sure this is also incorrect though, as I recall the devs mentioning that there is MMR amelioration based on the current state of the game. IE: The first kill is worth more MMR than the fourth.
You remember wrong.
I don't know why you're trying to argue about kill rates when you want to talk about MMR. The latter system is very opaque and not directly tied to kills.
It is pretty much directly tied to kills, as is kills what determines a win, a draw or a loss.
It's well possible that a killer can go up in MMR on a 2K, in which case 2Ks wouldn't all be draws but would be distributed amongst wins and losses
No, it is not. A 2K is a draw for a killer, ergo, no gain or loss of MMR.
Increasing kill rates would not lower the win rates, though, would it?
I already explained in detail why and how the kill rate should be 60% to guarantee a 40% win rate for both sides. Re-read it if you still have any doubt about it.
You have already been explained why this doesn't work though: DbD does not have a linear distribution of kills among games, which means kill rates don't directly translate into win and loss percentages.
I explicitly explained how kill rates directly translate into win and loss percentages. Survivors only lose if they are killed, ergo, the kill rate need to be 60% so there is a 40% win rate in the survivor side. Then, you buff or nerf each individual killer that had an average of survivors killed per match higher or lower than 2.4 to guarantee a 40% win rate on the killer side among the 33+ killers. Easier said than done, if even possible, but because of the asymmetrical nature of the game that's the fairest and probably the only way to do it if you want equal chances for survivors and each of the killers of the game.
It feels like you're arguing three different things at the same time: Wins as team, wins individually, and wins by MMR.
There is no "win as team". Survivors, in MMR terms, are not a team. If you are killed and 3 of your "teammates" escape, you have still lost. And in reverse, if you are the only survivor that escaped, you won and the rest lost. Again, each survivor has it's own personal match against the killer, so even if you have 3 other survivors playing the match with you, in MMR terms is a 1v1. You either win over the killer by escaping, lose to the killer by getting killed by him, or draw by escaping through the hatch. That's why that way of working was called by the devs as a "1v1/1/1/1".
And that's the hatch counting as a draw. Which means that we can no longer take the 53% kill rate to immediately translate to a 47% win rate for survivors
As I already said 3 post above:
- You are technically right, but as far as I know, the hatch is still considered an escape on the stats because even if it is neutral MMR for the survivor, it is still an optional free escape survivors have that gives them all the same conditions in game as if they escaped through the exit gates. That's why it is counted inside that "40%" win rate, as technically they have still escaped.
At which point the stats are telling us the following:
1: We don't know MMR.
And this is something that I had being asking since I started playing the game, for them to show us our MMR and if the cries of people allows it, those of our teammates and our rivals at least at the end game. For multiple reasons.
2: Killers are more likely to win than survivors are, when survivors are counted as a team.
There is no "survivors team" to compare about, and killers are not more likely to win according to the data in Nightlight for the reasons already explained.
3: Killers are more likely to win than survivors are, when survivors are counted individually.
Read previous point.
The stat distributions just don't support the argument that survivors are favoured.
Read previous point.
1 -
You remember wrong.
The article is unsourced and also kicks your argument to the curb.
It is pretty much directly tied to kills, as is kills what determines a win, a draw or a loss.
If the final amount of kills is the ONLY determining factor for whether the killer gains or loses MMR, then that means that killers WIN more matches than they LOSE, going back to the 46% chance of a 3K+ and a 40% chance of a 1K-. So in terms of MMR, killers are still better off.
I already explained in detail why and how the kill rate should be 60% to guarantee a 40% win rate for both sides. Re-read it if you still have any doubt about it.
However, you have also been presented with factual proof that your assertion is incorrect. 60% kill rate does not, at all, -guarantee- a 40% win rate.
To quote @ratcoffee , whose point you most definitely missed:
Let's take your example of 100 matches again. Killer plays 100 matches, gets a 3k in 80 of them and a 0k in the remaining 20. This comes up to 3 * 80 = 240 survivors killed, just as in your example. However, the killer win rate (by the metric you chose) is 80%, far off from the 60% kill rate.
Explain how it is that a 60% kill rate can produce an 80% win-rate if it is, instead, supposed to -guarantee- a 40% win-rate. Because it clearly didn't, here. So please, show where the math is incorrect, because as far as I can see, Ratcoffee has drafted up a scenario where there IS a 60% kill rate, but there ISN'T a 40% winrate.
I explicitly explained how kill rates directly translate into win and loss percentages. Survivors only lose if they are killed, ergo, the kill rate need to be 60% so there is a 40% win rate in the survivor side. Then, you buff or nerf each individual killer that had an average of survivors killed per match higher or lower of 2.4 to guarantee a 40% win rate on the killer side among the 33+ killers. Easy to say than do, if even possible, but because of the asymmetrical nature of the game that's the fairest and probably the only way to do it if you want equal chances for survivors and each of the killers of the game.
And I and others have explained to you why it does not directly translate into win and loss percentages. So your entire plan for rebalancing killers for a specific kill% is proven to not necessarily facilitate a fairer win%. So if the current win% is already OVER your estimation for a fair win%, how is increasing killer power to drive kill% UP going to bring the win% DOWN?
There is no "win as team". Survivors, in MMR terms, are not a team. If you are killed and 3 of your "teammates" escape, you have still lost. And in reverse, if you are the only survivor that escaped, you won and the rest lost. Again, each survivor has it's own personal match against the killer, so even if you have 3 other survivors playing the match with you, in MMR terms is a 1v1. You either win over the killer by escaping, lose to the killer by getting killed by him, or draw by escaping through the hatch. That's why that way of working was called by the devs as a "1v1/1/1/1".
At this point, though, you are absolutely detaching yourself from the game's practice in an effort to try and create a mathematical equivalence between two sides of an asymmetrical multiplayer. They are not the same and they cannot be evaluated as the same, either.
You are technically right, but as far as I know, the hatch is still considered an escape on the stats because even if it is neutral MMR for the survivor, it is still an optional free escape survivors have that gives them all the same conditions in game as if they escaped through the exit gates. That's why it is counted inside that "40%" win rate, as technically they have still escaped.
This is where I'm going to state that you are being disingenuous in your argumentation. You've hung up your argumentation so far on the MMR system in order to facilitate the 'Actually, it's a 1v1' argument, as a way to dodge the fact that survivors lose more as a team than killers lose.
But now we get to the hatch, and the way it cuts into the projected win-rate for survivors in these specific 1v1s, and the same section on the same article that states that a 2K for the killer MUST result in 0MMR ALSO states that a hatch escape MUST result in 0MMR.
Since this is inconvenient to you, you then change the definition of a win to get OFF the MMR system and still declare the hatch a win.
You are arguing over three different subsections of stats: MMR, win% as team, and win% as individual. And whenever the stats don't agree with your assertions, you switch tracks and start arguing about a different branch.
and killers are not more likely to win according to the data in Nightlight for the reasons already explained.
And you are blatantly disregarding any stats that are shown. The data in Nightlight shows 46% to 40%. So yes, killers are winning more than survivor teams are. And they're also winning more than individual survivors by a metric that YOU brought into the discussion.
3 -
The article is unsourced and also kicks your argument to the curb.
Need any more source than that?
If the final amount of kills is the ONLY determining factor for whether the killer gains or loses MMR, then that means that killers WIN more matches than they LOSE
And? They still win less times on average than survivors.
However, you have also been presented with factual proof that your assertion is incorrect. 60% kill rate does not, at all, -guarantee- a 40% win rate.
But it is the only way to guarantee having a mathematical average chance of winning equal for everybody, survivors and each individual killer. It's not hard to understand, much less when explained in detail multiple times.
Explain how it is that a 60% kill rate can produce an 80% win-rate if it is, instead, supposed to -guarantee- a 40% win-rate.
Again, I did already, multiple times. Go and read them again.
And I and others have explained to you why it does not directly translate into win and loss percentages.
And you are all wrong because, again, I explained how and why it is the only way to get an equal win rate for everybody by setting win rates for survivors at 40% and make every killer individually have an average kill rate of 2.4 survivors per match which translate in another 40% win rate for every single one of them.
At this point, though, you are absolutely detaching yourself from the game's practice in an effort to try and create a mathematical equivalence between two sides of an asymmetrical multiplayer.
From the Official Wiki article I posted:
- In essence, each Survivor is in an individual mini-Trial with the Killer and vice versa, in which wins, losses, and draws can occur.
Also, the 60% kill rate is something that the devs want themselves. So if anybody is doing that in any case, are the devs. I'm just following the MMR rules and what they say.
This is where I'm going to state that you are being disingenuous in your argumentation. You've hung up your argumentation so far on the MMR system in order to facilitate the 'Actually, it's a 1v1' argument, as a way to dodge the fact that survivors lose more as a team than killers lose.
There is no "team" in MMR. Even in game, you don't get any more BP points or awards for more than one of you escaping. You don't get an "Survivor team wins!" anywhere. You either had escaped, or you had being killed. If you are going to ignore every fact and keep saying I'm being disingenuous, please specify where in all the game are you told that if any specific number of survivors escape then survivors have won, when it makes pretty clear, icons included, that you and only you either escaped or got killed.
Since this is inconvenient to you, you then change the definition of a win to get OFF the MMR system and still declare the hatch a win.
Because in game terms, it is an escape with all that comes with it: BP, Emblems, Experience points, retain your item. Everything. And it is a free, optional, easy way of getting an escape from the game that requires more luck than skill. It is not inconvenient, it is logical.
You are arguing over three different subsections of stats: MMR, win% as team, and win% as individual. And whenever the stats don't agree with your assertions, you switch tracks and start arguing about a different branch.
Again there is no win rates as a team because not even in game the survivors are a team, and I'm strictly guided by MMR which yes, have the exception on hatches because it is still an escape but as it is based on luck is counted as a draw in MMR terms, but an escape nonetheless. So I didn't switch tracks anywhere, it's you who is trying to make a completely new one grasping at it.
And you are blatantly disregarding any stats that are shown. The data in Nightlight shows 46% to 40%. So yes, killers are winning more than survivor teams are.
No, killers are not winning more than survivors (much less "teams", as kill rates obviously counts only survivors, not groups of 4) because survivors have right now an higher win rate on average than killers as already pointed and proved multiple times.
And they're also winning more than individual survivors by a metric that YOU brought into the discussion.
Not true, by all the things already more than explained and pointed. But whatever makes you sleep better at night.
2 -
1 -
But it is the only way to guarantee having a mathematical average chance of winning equal for everybody, survivors and each individual killer. It's not hard to understand, much less when explained in detail multiple times.
Except it's been proven that this is not the case. This -might- work if there was only one win condition and one loss condition, but there's two loss conditions, two win conditions, and a tie condition. And these mess up your 'Kill% -> win%' pipeline.
In your explanation, you assume that the average becomes the rule, but you fail to acknowledge that the game's kill distribution is weighted towards extremes. 2Ks are the rarest outcome, as everyone and Nightlight will tell you, but you bank on it in the one time you tried to explain your reasoning.
And you are all wrong because, again, I explained how and why it is the only way to get an equal win rate for everybody by setting win rates for survivors at 40% and make every killer individually have an average kill rate of 2.4 survivors per match which translate in another 40% win rate for every single one of them.
The problem lies in this section:
and make every killer individually have an average kill rate of 2.4 survivors per match which translate in another 40% win rate for every single one of them.
This is proven to be false. The killer's winrate does not follow neatly from the kill rate. As Ratcoffee demonstrated, a killrate of 60% can swing up as high as 80% winrate.
I -know- you're going to handwave this with 'averages', but Nightlight disproves the practical application of your theory as well: It shows a lower killrate than your projection, but also a higher winrate.
And no, you are not going to be able to figure out a way to buff killers to increase their kill rate, while reducing their winrate.
There is no "team" in MMR.
Except there is.
IF the game were to be ruled as a strict series of 1v1s, then each kill the killer makes is a win. This means a 3K is 3 wins for the killer, and a 4K is 4 wins for the killer.
This would mean the kill rate = win rate, in which case any kill rate over 50% = killer-sided.
But, you disagree with this, so you instead opt for MMR to judge the killer for his performance against the team of survivors as a whole, and you count both 3Ks and 4Ks as one win.
Which is fine, except you'd need to do the same thing from the other side. Meaning that only 3Es and 4Es are survivor wins. If the killer is pitted against the team of survivors, then the team of survivors is pitted against the killer.
But you're not okay with that, so you instead opt for the MMR system again, in which survivors are judged mostly individually.
So up until now, you've ruled that MMR is the main guideline for how a win or loss is determined, but then comes the hatch. That thing unfortunately brings the kill rate down, while also cutting into the survivors' winrate by introducing a draw. This is the point where you arbitrarily choose to abandon MMR as the guideline and instead switch to 'vibes'.
But the 'vibes' also dictate that only a 0-1K 'Displeases the Entity', while a 2K is a sign of a 'Brutal Killer', essentially switching to win terminology. So should we now start counting 2Ks as wins, because it has the 'vibes' of a win?
Either stick to MMR, stick to 1v4, or stick to 1v1v1v1v1.
So in short:
1v4 -> Killers win 46% of the time. Survivors win 40% of the time.
1v1v1v1v1 -> Killers win 53% of the time. Survivors win 47% of the time.
MMR -> Killers win 46% of the time. Survivors win somewhere south of 47% of the time. Indeterminate due to hatch.
Vibes -> Killers win 59% of the time. Survivors win 47% of the time.
I also want to point out that your absolute best effort, using arbitrary rulesets for what constitutes a win, only manages to pull off a 46% killer winrate versus a 47% survivor winrate, which is a level of balancing that would make League of Legends developers moist. So with your best argument, you can place the game as being negligibly survivor-sided.
Everything else points to the game being varying degrees of killer-sided.
3 -
Looks like this was updated since you posted it. I'm guessing the wiki people weren't happy with you trying to use it to discredit them.
As it says, the wiki's source is the game's own code, unless mentioned otherwise. Seems pretty credible to me.
2 -
The above official BHVR post, explains they want the kill rate to hover around 59% to 61%.
This means that if the kill rate is currently lower than 59% to 61%, then BHVR should be making game changes to raise the kill rate back up to 59% to 61%.
5 -
Well, if you think that article has being "inadequately changed or vandalized" report it. Until then, the point still stands. And I could not find the streams with the own devs explaining the system, but here is Otzdarva reacting to the devs when the system was still being developed explaining pretty much the same as me in terms of the 1v1/1/1/1, even if some things changed since then (the length of the match affecting how much MMR you get or the conditions to gain or lose MMR for the killer, for example):
So... what was you trying to point, exactly?
Post edited by Batusalen on0 -
Those stats are not correct. Devs released their data. Killrate is close to 60% and at the top 5% its above 60%.
Said that, i am all for ballance. So 50% it is for me. I know this will be extremly hard tho.
You argument might make sense for you, since you like to be the underdog. I like to be in an equal match.
Last but not least, the game is on no level survivor sided, not in comp, not in top 5% acording to stats
0 -
Except it's been proven that this is not the case.
Where...?
This -might- work if there was only one win condition and one loss condition, but there's two loss conditions, two win conditions, and a tie condition. And these mess up your 'Kill% -> win%' pipeline.
Explained it multiple times already, go and re-read it until you understand it.
This is proven to be false. The killer's winrate does not follow neatly from the kill rate.
Buff and nerfs have being proved false? Because I'm starting to think that you reaaaaly don't understand what have being explained multiple times with you.
[...] a killrate of 60% can swing up as high as 80% winrate.
That's why you should adjust the average survivor killed per game to 2.4, so that win rate stays at 40% meaning 40% win rate for survivors and every single killer. What is so hard to understand here?
but Nightlight disproves the practical application of your theory as well: It shows a lower killrate than your projection, but also a higher winrate.
Read the above point.
And no, you are not going to be able to figure out a way to buff killers to increase their kill rate, while reducing their winrate.
To be fair, what the game needs right now is more a nerf to survivor than a buff to killers. But yet again, that's why you need to get to the 60% kill rate base, so you can buff or nerf every killer individually to meet the average. I think I'm going to start counting the times that I had explained it to you already. This is at least the 4º time.
Except there is.
No, there is not. 3º time.
IF the game were to be ruled as a strict series of 1v1s, then each kill the killer makes is a win. This means a 3K is 3 wins for the killer, and a 4K is 4 wins for the killer.
Hey!!!! See? It's not so hard to understand, right? A killer needs to win 3 times in his individual match against each survivors for him to have a win. And if he manage to do 4 wins, then is an extra win, and he gains more MMR.
This would mean the kill rate = win rate, in which case any kill rate over 50% = killer-sided.
And back to it again. Was nice for a moment.
Again, more than 3K is a win for the killer. Or, if you prefer it, 3 wins over survivors is an overall win for the killer. Ergo no, 50% kill rate in any case would means killers not losing or winning and survivors winning 50% of the time on average.
But, you disagree with this, so you instead opt for MMR to judge the killer for his performance against the team of survivors as a whole, and you count both 3Ks and 4Ks as one win.
No, the devs, the code of the game, the Wiki, and everyone that have read or hear the devs talking about or explaining how the MMR system works disagree with it. You are the only one that is trying to change the fabric of reality by denying the empirical fact.
Which is fine, except you'd need to do the same thing from the other side. Meaning that only 3Es and 4Es are survivor wins. If the killer is pitted against the team of survivors, then the team of survivors is pitted against the killer.
No, because that's not how the system works. Each survivors it's by his own. A survivor win or lose depending if he escapes or not, and doesn't matter what other survivors do. I-don't-even-know-how-manyº times already explained.
[...] but then comes the hatch. That thing unfortunately brings the kill rate down, while also cutting into the survivors' winrate by introducing a draw. This is the point where you arbitrarily choose to abandon MMR as the guideline and instead switch to 'vibes'.
The MMR measure "skill". Having a free optional escape spawn just below your feet just because you are the last man standing that rewards you the exact same as if you completed the gens and go through the exit gates doesn't require a lot. That's why it is still considered an escape, but you don't gain MMR with it. It's not vibes, it's totally logical and understandable. But as it fits your narrative, you are grasping at this straw to try to make a point, not me.
Either stick to MMR, stick to 1v4, or stick to 1v1v1v1v1.
It's what I'm doing, that why I'm sticking to the 1v4 for killers and 1v1/1/1/1 for survivors as that's how it works.
1v4 -> Killers win 46% of the time. Survivors win 40% of the time.
1v1v1v1v1 -> Killers win 53% of the time. Survivors win 47% of the time.
MMR -> Killers win 46% of the time. Survivors win somewhere south of 47% of the time. Indeterminate due to hatch.
Vibes -> Killers win 59% of the time. Survivors win 47% of the time.
"Killers win 46% of the time. Survivors win somewhere south of 47% of the time" That's the only truth. Everything else it's you distorting things to accommodate your point of view and your bias.
I also want to point out that your absolute best effort, using arbitrary rulesets for what constitutes a win
It's not my "arbitrary rulesets", they are the rule sets programmed in the game working at every match you do. So no effort at all using them, as it as simple as understanding them and pointing them out.
[...] only manages to pull off a 46% killer winrate versus a 47% survivor winrate, which is a level of balancing that would make League of Legends developers moist.
But balancing was not the point of this post, this post said that the data proved the game was "slightly killer sided" when it proves totally the contrary.
The best part of it is that you really think you can compare LoL with DbD, when that only shows that you don't have any idea of what you are talking about in terms of balancing. No wonder it is so hard for you to understand simple things even if explained to you a billion times already.
So with your best argument, you can place the game as being negligibly survivor-sided.
Everything else points to the game being varying degrees of killer-sided.
"Slightly" survivor sided. And yet again, no, nothing not even in Nightlight points that the game is killer sided in any degree, but the contrary for all the reason explained to you a billion times already.
And before you make your next post, I will answer to it already: I already explained it to you. You are wrong.
2 -
60% is ballanced?
Adam Ries is turning upside down in his grave :(
1 -
Multiple post explaining why 60% kill rate is balanced in this thread, go read them.
3 -
Because of what? Survivor DC? Multible posts explained how this is bs and the real kill rate is actually even higher then that. Go read them.
2 -
So, instead of reading to the pointed explanations you are going to simply ignore them and keep saying whatever you feel like. Ok.
2 -
I am sorry, i have not found any of those so called explanations, besides the killrate is fair because survivor DC more often then killer. If you got somethign better then that, i am willing to read it of course.
0 -
Yes, 60% is balanced, and above is the official BHVR post telling us that 59% to 61% is where they want the kill rate.
3 -
And the short explanation is it is the only way to get an equal 40% win rate on average for survivors and every single killer. If you want the math, why and how, again, they are in this thread.
1 -
Where...?
By Ratcoffee, when they constructed an 80% winrate out of a 60% killrate.
Explained it multiple times already, go and re-read it until you understand it.
You have explained what you think, but I have explained multiple times that you are incorrect.
Buff and nerfs have being proved false? Because I'm starting to think that you reaaaaly don't understand what have being explained multiple times with you.
You need to understand that right now, you are doing the equivalent of proclaiming that 1+1=3, and that me disagreeing with you is just me 'not understanding'.
You are mathematically incorrect.
That's why you should adjust the average survivor killed per game to 2.4, so that win rate stays at 40% meaning 40% win rate for survivors and every single killer. What is so hard to understand here?
That 60% killrate would be an average of 2.4 survivors killed per game. If you can win 80% of your matches with an average of 2.4 kills per game, how can you guarantee that 2.4 kills per game will result in 40% winrate?
You keep insisting that you have explained this: You have not. You have asserted it, but you have not explained it. You have not explained how 60% killrate translates to 40% winrate. You have not explained how you are going to guarantee that the 240 kills you are aiming for over a hundred matches will neatly distribute themselves to produce 40 wins, as opposed to what Nightlight is showing, where fewer kills are weighting themselves towards a higher winrate.
How do you intend to make killers that kill more than they currently do, lose more than they currently do?
Hey!!!! See? It's not so hard to understand, right? A killer needs to win 3 times in his individual match against each survivors for him to have a win. And if he manage to do 4 wins, then is an extra win, and he gains more MMR.
You may want to get off your high horse here because if THIS was the metric you wanted to go for, killers have a 53% winrate, and survivors are nowhere near that no matter how badly you're trying to spin this. Which would prove you wrong and demonstrate the game to be killer-sided.
Again, more than 3K is a win for the killer. Or, if you prefer it, 3 wins over survivors is an overall win for the killer. Ergo no, 50% kill rate in any case would means killers not losing or winning and survivors winning 50% of the time on average.
So your remark to the tune of me 'finally getting it' is immediately undone by you switching gears and going for a different metric. You are following along with what I'm saying, just trying to reframe it within your own perception.
IF the game were ruled as 1v1s, killers would have a 53% winrate.
You say that that's wrong, but what you -mean- to say is: "The game isn't ruled as 1v1s"
So we get to the 1v4s, after a little detour via MMR:
No, the devs, the code of the game, the Wiki, and everyone that have read or hear the devs talking about or explaining how the MMR system works disagree with it. You are the only one that is trying to change the fabric of reality by denying the empirical fact.
I want you to keep this one in your mind.
The MMR is -your- reason for ruling that the killer's winrate is not determined by the rate at which they kill survivors but by the rate at which they secure 3 or more kills in one trial.
We will come back to this.
No, because that's not how the system works. Each survivors it's by his own.
The system, here, being MMR. That is the 'system' in place. That is what you are referring to.
The MMR measure "skill". Having a free optional escape spawn just below your feet just because you are the last man standing that rewards you the exact same as if you completed the gens and go through the exit gates doesn't require a lot. That's why it is still considered an escape, but you don't gain MMR with it.
And this is where you -drop- the MMR, because here it becomes an inconvenience.
It is considered an escape. It is NOT considered a win. It is considered a draw.
The MMR does not refer to the hatch escape as a win. You stated that the killer's winrate isn't determined by 1v1s because the MMR says so. You stated that the survivors' winrate isn't determined by 1v4s because the MMR says so.
But when the MMR tells you that the hatch is NOT a win, you ignore it.
It's not my "arbitrary rulesets", they are the rule sets programmed in the game working at every match you do. So no effort at all using them, as it as simple as understanding them and pointing them out.
These are not rulesets programmed in the game. The game itself does not track 'wins'. It tracks Bloodpoints and MMR. It does not govern what is a win and what is not. It simply tells you the statistical outcome of the match.
The closest thing you can offer that is actually programmed in the game is the MMR system, but that system disagrees with you about the hatch.
What YOU, personally, classify as a win, is arbitrary and is not backed up by any element programmed in the game.
But balancing was not the point of this post, this post said that the data proved the game was "slightly killer sided" when it proves totally the contrary.
What do you think 'Slightly killer sided' means? It's a statement on the game's balance.
The best part of it is that you really think you can compare LoL with DbD
Reading comprehension is not your forte, is it?
5 -
We can't assume the number of kills follows a normal distribution. There is zero evidence it follows a normal distribution, and yes, I reviewed your math, and it's wrong.
The same problem is happening on a different thread on these forums, involving the average number of pallets on a map, where someone wrongfully assumed the number of pallets follows an even distribution, and therefore wrongfully decided the average number of pallets = (min number of pallets + max number of pallets) / 2.
3 -
A) DC automatically void that trial for the stats for killrates.
B) Lets do some math experiment.
You got 4 coins to toss. For every coin ther is a player associated with it, lets call those survivors. They win if the coin lands on heads. Theres a fifth person who instead wins if 3 or 4 coins land on tails, lets randomly choose to call that one camper.
Now the chances for each survivor to win is 50%, as they only need for their own coin to land on head. The camper though has a chance of 4 * 0.5³ * 0.5 + 0.5⁴ to win. It's 31.25%. Every of the 4 survivors has a 1.6times higher chance to win than the camper.
Now as for the so dreaded 60% killchance, the survivors chance to escape/win becomes 40%, which results in a chance of 47.52% (4 * 0.4³ * 0.6 + 0.6⁴). The camper chance to win is 1.188times higher than each survivors.
In the scneario with a 60% killchance, the chances to win are simply closer together, which is fairer.
Now obviously in the real game, there isnt a fixed chance for an outcome, each survivor performs differently, but people take the "average killrate" as gospel so i think my model seems useable.
The provlem is that too many people dont seem to know what they want the average killers winrate (3k-4k) to be equalized with. Equalizing it with the average individual survivor escape rate ( the average escape chance or whatever) makes "team wins (0k-1k)" rather rare.
But equalising killer win and team average win chance... See my 50% escape chance example above? The team also gets the same winchance as the killer , and thus makes it on average far more likely for a survivor than a killer win.
60% maybe isnt perfect, but waaaay better than 50%.
0 -
There are no "team" escapes. There has being never team escapes in this game. You don't get or gain anything for any survivor other than you escaping. That's what people seems to not understand. You don't win or lose as a team, you either escape or get killed.
2 -
The game shouldn't be sided to either. It should be a 50/50. Yes lorewise killers are supposed to be the power role and all but this is a game and not a show or movie so there needs to be enough balance on both sides to handle it.
2 -
Your team's performance does affect your gain/loss of mmr tough. As the dev's pointed out if you are the only one to die you barely lose mmr or you get more mmr if all escape, for example. You do in fact gain something from your team mates escaping.
This news article was posted on November 3, 2022.
2 -
That post was deleted from the developers's logs, and it's the only source of that. According to the wiki, what 6.4.0 introduced was for SWF, meaning that if you are in a SWF you would have those conditions, not on SoloQ as I already said. In fact, the patch notes for 6.4.0 don't even mention any changes to MMR at all, so either it was completely scratched and the Wiki is wrong on that point, or the Wiki got it from data mining the game's code.
Post edited by Batusalen on0 -
My entity, you guys trying to prove math wrong is exhausting.
The problem: We need to get all killers and survivors have the same chance of winning. How we do that?
The solution: Lets start with the mathematical fact:
The mathematical average for that will be 2.4 survivors killed, meaning 40% of survivors escaped on average, and killers are only getting 3K 40% of the time on average, while the rest they would only do a 2K, meaning they are not winning. If it was 50%, that would be 2 survivors killed per game on average, meaning killers are never winning while survivors are winning 50% of their games. If it was 70%, that would be 2.8 survivors killed per match on average, which would mean survivors are winning 30% of their games, while killers are winning almost every game they play (as they would be doing 3K in most of their games).
So, the only way to get mathematical equality in win rates is by having a 60% average kill rate and every one of the 33 killers doing a 2.4K per match on average so everybody has a 40%. In other words, the closer we get to this numbers, the more balanced is the game.
Now, how we get to that in practice:
60% kill rate means 40% win rate for survivors. We get all killers across the board to a point where that kill rate on average is achieved by either general buffs or nerfs to killers and/or survivors (something that they already achieved the last time they officially announced kill rates) and done. 40% win rate for survivors side achieved, which will means that some survivors are escaping more than 40% of the time, and other are escaping less. Or in other words, skilled survivors would be winning more games, as every single one of them has the same chances of winning than everybody else.
Now, how we get so every single killer do a 2.4K per match? By buffing, or nerfing, each killer individually until they are close to that 2.4K on average. Which will have the same implications as with survivors, meaning that some people with the same killer would be able to kill more survivors than other.
With that done, as Thanos would say, "perfect balance". Everybody has the same chance to win and if you don't is either because you are not good enough or your rival outsmarted you, not because the game favors any side or specific killer. This is the sole reason of BHVR wanting to have a 60% kill rate and if I have to guess, why the MMR system works how it works in the first place.
Now, any of you either bring empirical proof of how the math or the practical method to achieve what the math says is wrong, or accept reality for good already.
Post edited by Batusalen on1 -
None of your math is relevant at all. BHVR had an official post stating that 59% to 61% killrate is ideal. That's it. That's all the evidence you need for what the ideal kill rate is. You don't need to do any math at all. You can just directly quote BHVR.
1 -
My entity, you guys trying to prove math wrong is exhausting.
Not trying. Succeeding.
The mathematical average for that will be 2.4 survivors killed, meaning 40% of survivors escaped on average, and killers are only getting 3K 40% of the time on average, while the rest they would only do a 2K, meaning they are not winning.
Alright, let's try and break it down.
Your claim here, is that a 60% kill rate MUST necessarily produce a 40% winrate. Your claim is therefor that there is NO way that 240 kills can be distributed over 100 matches without producing ANY killer winrate that is NOT 40%.
Your math is as follows:
100 matches = 400 survivors
60% killrate = 240 kills
40 x 3 = 120 kills
60 x 2 = 120 kills
Therefor, with 60 2Ks and 40 3Ks, you get 60% kill rate.
This is your proof for why a 60% kill rate MUST result in a 40% winrate.
Here's mathematical proof that there are other ways to distribute a 60% kill rate that do NOT produce a 40% winrate:
100 matches = 400 survivors
60% killrate = 240 kills
80 x 3 = 240 kills
20 x 0 = 0 kills
Therefor, with 80 3Ks and 20 0Ks, you get 60% kill rate.
This is proof that a 60% kill rate CAN result in an 80% winrate.
Dispute this, because this is the cornerstone of your argumentation. Do not brush it aside again and simply repeat your own math.
Now, how we get to that in practice:
In practice, the killer winrate is too high for your projection. If you are aiming for a killer winrate of 40%, then killers will need to be nerfed or survivors will need to be buffed.
Please explain to me how you intend to raise the kill rate via killer nerfs or survivor buffs.
Or conversely
Please explain to me how you intend to lower the killer win rate via killer buffs or survivor nerfs.
And no, 'Setting their kill rate to 60%' is not an explanation, as it's already been proven that this does not necessarily facilitate a 40% win rate. You have to prove how you intend to do -both- 60% kill rate, AND 40% win rate.
Please think about this question in isolation, just try to put all the mathematics out of your head for a second and think about this question:
What buffs do you propose for killers that would bring their win rate down?
3
