They can't be accurate and broad, it's one or the other. If they are accurate then everything should be accounted for, if they are broad with many factors that weren't checked, then they are inaccurate. The fact that the stats are mixed between console and PC alone make it incredibly inaccurate because it's two very different playstyles. Old Nurse was the best on PC, but the worst on console, if you mix those two stats together it would be a very inaccurate and not tell the full story.
This is a direct quote from you on a Demogorgon post:
You'll have to forgive me since I don't have time to read through the whole thing right now, but:
I only thought I'd address this now that I finally have the PROOF that Demo is not a powerful killer. As per the stats released, at red ranks, Demo fairs very poorly, having the SECOND LOWEST kill rate in the game, just barely above The Clown.
We said it in the stats post and I'll say it again here, you should not draw conclusions from those stats. They are about as broad as they get. They don't show details and account for any amount of factors. To give an example, he's just barely below the Nurse and the Huntress, two fairly powerful killers. Are they bad too? I feel like most people would say no, they just have a learning curve and there's a lot of people who don't regularly play them will pull those numbers down.
When you even say "They don't show details and account for any amount of factors." then you can't say they are accurate. You can't have no details and it be accurate.
I will be happy when we get accurate stats. My post is simply saying that the stats are useless since it doesn't cover anything. I agree that it is hard too look at stats with so many factors, but too post them and say "They are accurate. The reason why you shouldn't draw conclusions from there is because they are very broad." it tells two different stories. You can't have accurate stats if you don't account for many, if not all the factors in said stats.
I am happy at some things the devs do, but the things I'm not I'm gonna talk about it, as people should. I am happy they are trying too post stats, but the stats are useless if it's not accurate.
The stats are accurate for what they actually represent. What makes you state they aren't? broad stats after all doesnt mean inaccurate.
Peanits simply said dont read too much into them as the be and end all of balancing as they take in many many more factors nothing more.
Of course they are broad and as I said all stats will always be that way for some as can never be what people want. The game is so varied on what can be used each match a stat on every single possible perk, addon, offering, killer, survivor, rank, skill level, saves, hooks, gens done, gates opened, totems done etc would be impossilbe. The devs have these and if we actually think about it its why things can take so long.
This quote from Peantis. they said "They don't show details and account for any amount of factors" which means they aren't accurate. You can't have broad and accurate stats, if you can't factor in everything in a stat/stats, then it is inaccurate.
They are both accurate and broad. I think they word you're looking for is precise. They are not looking a specific thing (e.g. Which is stronger, snares or fake pallets?), which is why you should not make a very specific conclusion with them.
I see people are still trying to find excuses until they get very deliberate data that supports their stance.
That's what I stated in my first post 😉.
The stats are accurate for what they represent.
The devs have said not to draw conclusions from them regarding balance as they basically don't show the whole picture.
That means anyone who does draw conclusions on either side are not thinking about anything and everything and are just using them to try and fit what they want to believe but not what is actually the case 😁.
1. Can you define what types of matches constituted as being 'Red Ranks' matches? (e.g. - a.) All 5 players started the match in Red Ranks, b.) Killer started the match in Red Rank but not necessarily survivors, c.) All 4 survivors started in Red Ranks but not necessarily the killer, d.) at least 1 player started the match in Red Rank, e.) majority of the players (3/5) started the match in Red Ranks.)
Just to clarify (since I've asked this several times without any answer) , what criteria did you use to determine that a match is a 'Red Rank match' to be used in the data?
2. Why were the Nurse's kill rate statistic only given for PC? Can you provide them for the other consoles?
Idk, I think that if they're aiming for 50/50 and we're not even seeing 50% survival rate on the 'best' maps I think there is probably a problem.
I dunno what it is, or what the solution is, but they're not hitting their goal.
Yes, and its funny to observe.
"The stats don't show what toilet paper killers/survivors use, so they are inaccurate and meaningless!".
If they are accurate that would mean killers are averaging like 70% or more win rate and that can't be right, you said it was closer to around 50% for all killers.
That is the problem people have with it, on the steam forums everyone is freaking out saying nerf EVERY killer because every killer so so high above that 50% marker according to that sheet.
That's the problem with making a conclusion with just the stats as they don't show any reason.
For all we know the majority of survivors just might not care about dying as much and throw themselves into situations which enable this to happen.
Then we have these stats which include the event and archive challenges which as we know the players have been farming instead of playing the game objective itself. Both these can heavily skew the stats.
I stated this before challenges came out as it was the same scenario in other games and it severely hinders your chances of escaping.
accurate and precise are synonyms, and if they are accurate, then why can we not draw conclusions from it? you are telling two different stories here. You can't have data that is both broad and accurate, it's simply not possible. If killers have around 50% win rate with everything factored in, but the data posted says they are in the 70% range, then it isn't accurate. It's simply not possible.
That's the problem, they aren't accurate. You can't have broad and accurate stats, it's simply not possible. If you exclude factors that would change the data from the data, the data is inaccurate. All it does is give us a very, very vague idea of where killers sit. Someone said that the actual killer win rate is 50%, but the data says it's up too 70%, which is causing people too say "nerf x killer." If this is true, the data is off by 20%, which is the farthest thing from accurate that is possible.
Of course they are as that is what a broad stat is.
These particular statistic show the kills over all ranks, platform along with popularity. That itself is still an accurate statistic for what they represent.
Whether they are accurate to use for overall balancing is the question (which we know they are not)
Thxs for the info, but my point still stands. Just cause the kill rate is around 60 to 70%, doesn't mean that play killer is fine.
It can't be accurate if there are factors that change the result. And it doesn't even show kills well because they combined Console and PC together, which completely ruins the data. If they were separate then that would be one thing, but having both combined completely changes how the data looks.
A friend of mine recently posted a fairly active thread on the state of Demogorgon, and cited the fact that he's one of the least played Killers with the lowest win rates as one of the reasons Demo needs buffs.
Peanits comes along like "lol actually the stats aren't accurate, and the reason Demo isn't played is cause he's hard lol. Also way less people play Hag lmao."
The devs specifically release stats to explain to us why they make balance changes but then completely ignore their own stats when balance changes are suggested. Again what's the point of releasing stats then? Why tell us that post-rework Nurse is the only Killer with a sub 50% kill rate just to tell us "she's hard lol that's why she isn't getting kills", or to tell us that Clown is one of the least played Killers in the game just to say "he's hard lol that's why people don't play him."
"They're hard lol" isn't a good reason for a character to be dogshit tier.
You are completely missing the point.
These are stats simply to show the average kills by killers across all platforms and who is used. They show exactly what they are intended too and for that reason they are 100% accurate.
What they include or don't is irrelevant as like they stated
"Please keep in mind that numbers are not everything; there are a whole bunch of factors that could influence these numbers."
Can they be used as a tool to balance? No
Should the players use them to prop up any arguments? No
Do they show what the devs chose to show? Yes
What they include or don't doesn't matter as they show what the devs wanted to show. Even numbers which may include dc's are 100% accurate in that regard. Not what they should use for balance but accurate for what they are none the less.
Agreed. Plus the devs keep saying "The stats are accurate, but we didn't remove any factors that could alter a killer's kill rate." which means it isn't accurate. The devs said that all killers are around 50% kill rate, but the stats posted go as high as 70%, so now people are calling for those killers too be nerfed.
Honestly they just need too remove the stats and give us stats that are actually accurate, not random stats that don't even work because they are combined from both console and PC.
Why? What reason would PC or console be separate? They aren't 2 separate games. I continue to ask this of people who declare that PC stats should be separate from console stats and I have yet to get a valid answer. The only difference between the two is framerate, which effects next to nothing in a game like this. Literally no difference.
Because they control in completely different ways. best example is Nurse. On PC Old Nurse was the best in the game, but on Console she's the worst. Combining her stats together completely ruins any valuable data that could be gathered since it's too different variables.
Nope, the target we aim for is 50%. The numbers they're actually pulling in are all over 60% right now. I can confirm that they are accurate.
Can you please gather the data for 4-swf at rank 1? That is probably the most accurate representation of "balance" we can gather from stats.
We have 4-swf at all ranks, which is irrelevant just as nurses killrate at all ranks or all platforms. And killrates at rank 1 for all group sizes. But most killers see the problem specifically in 4-swf at rank 1. Lets clear this up
So every killer is getting nerfed then.
That's the takeaway a lot of people are getting from this.
No one is calling them inaccurate they are saying with this and no other information provided it's creating more confusion than it's helping.
This is why historical we have not released stats that often; when we do, even if we suggest not drawing conclusions, some people will draw conclusions.
Why not simply give us stats that we CAN draw conclusions from? Why must the stats be so vague and inaccurate for when you make them public? When can we see stats that we can use for balance suggestions? When are we getting stats that are accurate and have all factors accounted for? Where are the survivor stats?
Also, if you're aiming for 50%, all killers are actually around 60%, but the stats posted have killers at 70%, there is still a massive problem.
I have never claimed that they are "actually around 60%". I've only stated that they are all at least 60%- which is true. Even the Clown, who is the lowest on those graphs, is over 60%.
Stats alone are not enough to make decisions on. Even we do not make decisions purely based on stats. We could go into very specific details, but there's always going to be a "what if". What if you counted this as well? What if you exclude that? What if you only look at matches with this perk?
Point being, the stats are there for fun and to spark conversation, not make conclusions.
"Nope, the target we aim for is 50%. The numbers they're actually pulling in are all over 60% right now. I can confirm that they are accurate."
This is a direct quote from you earlier in the thread, you literally say they are all over 60%, which is confirming that killers all are around 60% or higher, so if you are aiming for 50% and the lowest a killer is is 60%, with the highest being around 70% in the stats, then they aren't accurate.
Also, how can we start conversations when every time someone does a dev comes in and says "We said it in the stats post and I'll say it again here, you should not draw conclusions from those stats. They are about as broad as they get. They don't show details and account for any amount of factors."? What's the point of giving stats when it doesn't mean anything in the end and people can't use them for conversations? Will we ever get stats that are actually based for balancing so people can have conversations about them?
Some expected informations were confirmed: freddy and spirit are top tier killers that are easy to play