39% Escape Rate is a Lie
Edit: This is NOT a discussion of SoloQ vs SWF. Enough other threads arguing that.
Statistics are the best liars.
I don't doubt 39% is the actual mean of escape rate for all survivors from all matches globally. However, that is masking a very big lie through omission with selective statistical presentation. Pre Patch BHVR happily released perk usage rate by MMR tier to justify itself. Note how they did NOT break down escape rate by MMR this time. The reason is quite simple.
High MMR SWF survivors probably have an escape rate of 50-75% going off data I can find and I think most true high mmr killers would concur. Those high mmr survivors also play a lot of games. So you have a small subset of your playerbase, maybe 10-20% who play lots of games that are skewing the number higher to 39% (which already seems painfully low).
So this means our low and medium mmr survivors are likely seeing escape rates of 15-25%, 30% tops. My own data from just under 400 games since 6.x patch pegs it around 15-20% (yes I'm probably low tier mmr now) and I've seen similar data sets from others.
What we need is escape rate broken down into MMR ranges and SoloQ vs SWF. Would also be interesting to average every players individual escape rate, then take a global mean and median of all players. This would remove impact of a pro survivor playing 1000 games in a week vs casual playing 10 games in a week. If there is a huge difference between the mean and mode, that would heavily indicate stats being skewed by a few highly skilled survivors. Same thing is done in economics, looking at the average/mean income isn't helpful because you have a few outliers earning tens of millions of dollars while the median shows your more true "average".
I'm also unclear how they intend an elo based mmr system to function if they aren't aiming for a 50/50 win rate. Seems like they were really close to good pre patch with 47/53 (although even that was getting skewed by high escape rate survivors). If most of your survivor player base sees under 30% win rate and killers get over 70% win rate, your mmr distributions are going to look like two inverse curves. Fear my mspaint skills to see the issue. How are you going to make good matches when each player base is lumped together on each end? With two different pools you have to match between, you would ideally want them to be similar shapes. This is probably why high MMR killers had bad queue times early on before they widened the spread.
Edit Update - Adding this here because I posted it farther in.
Because people don't seem to understand how easy it is to produce a 39% global escape rate while at the same time having terrible escape rates at lower mmr tiers here is some example data of the kind I would love to see from BHVR.
In this case, look how easy it is to get a global 39% rate while having vastly different escape rates for different mmr groups. This does not indicate a healthy balance and 39% hides a very ugly reality.
Now that first example is assuming equal play rates. As no surprise to anyone, pros play a lot more than beginners.
Now in this case I'm assuming the top 10% of players play up to 5 times (2nd top tier playing 4 times as much) as many games as the bottom 25% of players (big assumption on my part although conservative I suspect). I have also towned WAY back the escape rates on the upper end and mid tier, yet I still end up with just 39% global escape rate they are reporting. If we had kept my first assumptions on the top tier escape rates that would require even worse escape rates for the bottom tiers to reach the 39% when weight for game frequency. The other scary aspect is only 25% of the player base is seeing an escape rate over 50% while 75% see escape rates of 25% or worse in this scenario.
My ultimate point is that reporting a 39% escape rate is masking a huge issue in the lower 50% (or maybe 75%) of survivors. Public data on the lower 50% of players is non existent outside our own records. High tier play stats is more accessible and we can weigh in the smaller population size into the data.
Yes, my data is all hypothesis. I'm making a point that we need more data from BHVR because what they have provided is a extremely misleading and deceptive in attempt to mask a severe balance issue for a large percentage of the player base.
Comments
-
SoloQ should defo start with a "handicap" (lower mmr) than SWFs, and i also think that SWFs shoud be pushed up in MMR for each extra player in the group.
6 -
According to devs the escape rate variation for SWF and SoloQ is in the vast majority a matter of a couple of %. The highest mmr SWF will have an escape rate superior by 15 points of % at the very most.
So for common mortals, SWF and SoloQ does not make much of a difference.
That being said, I agree with the rest of your arguments.
(I'm getting big economic student vibes from this post and I'm loving it, as one myself)
4 -
After reading the o/p I am getting 'distrobution' vibes on my end, coupled with some 'probably', 'likely', etc vibes while trying to make a direct statement/correlation based on personal assumptions.
8 -
Oh I didn't say all of the numbers were necessarily correct. I'm saying the effort is there to make it official looking, and assuming their numbers are correct (which they may not be), the following logic is sound.
1 -
I do not get that at all, but I respect your viewpoint good Hag!
I'll agree that people die more starting out at lower/basic mmr, though - that's not really disputable.
2 -
Link to devs saying that? I don't doubt the difference between SWF and SoloQ is overstated or missing the actual causation. IE: Good players take the time to form SWF while poor and mediocre players don't. Poor players in SWF don't magically become pro. It would rule out a factor if they posted the %s. My issue more is High MMR SoloQ+SWF skewing the data heavily for everyone else.
Yes, I typoed in mspaint. Whoops.
Also, Engineer with Computer Science degree although college is long behind me.
Of course my data is suspect and incomplete. The whole point of the post is to ask BHVR to release more accurate data because what they released is deceptive lying with statistics. The issues I pointed out do not depend on my assumptions, they are a hypothesis at this point in need of research. It would be foolish to think the entire player base got 39% escape rate exactly.
4 -
Pobody's nerfect, friendo :3
3 -
Here you go ! (screenshotted for clarity)
12 -
Everyone stop trying, the funniest answer has been written xD
3 -
Unless your at very top of killer-level like top 5%, All of your matches are balanced and perfectly winnable. its only at very high-level where swf vs solo balance exists. i.e if your top-level solo, you'll have lesser escape rate(15% less) then that of high mmr 4 man swf. it has vague implication that soloq players generally have a lower max-mmr then 4 man swf's so for example, if your soloq player that is good at the game, your max mmr is likely to range between 1800-1900 where as top MMR swf would range between 2000-2200 or something along those lines.
for everyone else, it's basically same. a team that has 4 solo players survivors that are 1500 MMR vs 4 man swf that is 1500 MMR would have same/similar escape rate since their overall success-rate for both teams is suppose to be arbitrarily equal. both teams are of equal skill against a killer that is 1500 MMR range. as said, swf doesn't give you an advantage if you are not stellar at dbd in the first place.
2 -
I don't remember when DEVs revealed the actual soloQ escape rate numbers and they probably won't anymore, because they'd show how broken the game is for soloQ.
1 -
Yourself and OP's civil replies to our minor disagreements has earned my respect this day!
It's really weird to see replies defending stuff/disagreeing with me that don't resort to KILLER/SURVIVOR MAIN rhetoric/personal attacks.
2 -
It would be nice to see the numbers between solo and SWF. If you're following a recipe, words like "few" don't have any place in it.
0 -
At the end of the day, we're all people trying to enjoy this game, there's no sense in being aggresive over opinions :)
1 -
I agree that having more detailed stats would be lovely. I'd LOVE a good spreadsheet of DbD stats !
1 -
Agreed.
It's pretty amusing to see people turn into pretzels while I'm sitting here shrekked and passively typing.
2 -
1. I will not hug the hag.
2. The terminator survivor squads do not have a 50~% escape rate. It's likely closer to 80 or 90.
3 -
Terminator survivor squads would be at the utmost top of survivor mmr, which would be the top escape rate. They might very well be 100% escape rate. But what the devs mean is that at equal mmr, solos don't have a significantly lower escape rate. Solo Q at extremely high mmr also have a very high escape rate, though not exactly as high as SWF.
Fortunately, such squads are extremely rare and you're unlikely to come across multiple of them, unless you're also extremely high mmr as killer (in which case, congrats ! :D)
Also, you're missing out, she's a lovely lady when you get to know her. 10/10 will never, under any circonstances, eat your liver.
0 -
Ok, this has got really off track. My point is NOT SWF vs SoloQ (although I am curious). My point IS low and mid mmr vs high mmr escape rates. There is no way low and mid tier are seeing 39% which tells me high tier is skewing the number and the truth for majority of survivors is ugly.
I'm concerned about the longevity and health of this game post update as a fairly new player myself (4 months now). If you expect new players to stick with a game when their win rate is 15-25% at the low mmr tier...that is unlikely and you need new players to sustain long term.
1 -
where can i see my mmr?
0 -
Nirgendwo. You can estimate your MMR by observing your mates and your killer. I get a lot of survivor mates and killers who only have level 1 and level 2 perks. It is safe to say I should be low MMR. And I am happy about it.
0 -
Its probably the spleen. The spleen it the most well blood supplied organ in the body. Thats her kinda snack.
1 -
"I don't doubt 39% is the actual mean of escape rate for all survivors from all matches globally."
If you don't doubt that, then why you write a wall of text arguing that not global stats matter.
The game gets updated with features, or balance changes based on global stats they don't make a separate patch constructed towards every player's skill level, every region's pickrate, and so on. The game patches are globally applied so ONLY global stats matter even if you love imagining every single scenario in the book.
2 -
This is what I have been trying to say but you said it with an economic twist and in a more intelligent way.
Yes, BHVR is lying. I don't know why other than to make the game seem more fair than what it is (marketing). To be perfectly fair to BHVR, every single company lies. We as consumers have to be smart and make our own decisions and not believe every single thing that is said. It is clear this time that the consumers are not buying BHVR's propaganda of defending the killer of being at a massive advantage. Only killers are defending it because it supports a power fantasy over players in a pushed competitive game in what is really not a competitive game.
And 39% winrate is a perfect example. In layman terms, the game is taking the majority of all killers and says: "Dear Killer, I don't care how good you are, you will win a tremendous amount of time with less effort and skill than your 4 opponents". On the flipside, it says to survivor: "Dear Survivor, I don't care how good you are, you will lose a majority of time with way more effort and skill than your singular opponent.".
That is for the majority! I would say about 85 to 95%.
Here is where it gets complicated. On the far right side of your graph where you see a clumping of high MMR killers and an extremely small subset of survivors that are supposedly high MMR. This is because the game requires a tremendous amount of more skill and effort from survivors. I disagree that there is many solo survivors here because it is not possible. It takes four working in perfect harmony to over a killer that knows what they are doing and even then it doesn't always work. There is also a lot of hackers sitting at the top of MMR and that is influencing killrates as well. The overall survivor rate is less than 39% after you dismiss hackers.
The majority of survivors are losing all the time outside of their control while a very tiny population at the top and most of killers are playing in paradise.
1 -
I really feel like the devs should add a way to escape. Personally, I feel like if you've done 2 full gens alone, you deserve to escape. If you've done 4 gens together, both of you deserve to escape(so basically, finishing a gen together gives half a point, finishing a gen solo gives a full point and with 2 points you can escape).
You would give solo survivors the option to stick around to help teammates, or get out early when things go down south. And these escapes actually granting MMR to survivors and removing it from killers, not a "tie" like hatch.
I feel like this would solve Camping and Tunneling quite a bit(because if you're Bubba in a basement, at least 2 survivors can escape and the 3rd one could find the natural hatch.), it would give a reason to killers to not ignore people on gens and it would basically remove slugging for the 4k in a 2v1 scenario(because at least 1 survivor is able to escape early).
Yeah, you could heavily screw your teammates over, but on the other hand, you could also be spared a 4 minute slugfest.
It also gives a means for survivors to exit a game early on without needing to DC or suicide.
It would give reason for individual MMR for survivors, punishes killers who make games boring and gives reason for survivors to stick on gens if they can.
0 -
Honestly from my point of view i agree what they are saying is a lie, but is a lie also what are you saying, is already since the the new DLC of project W that i see "AN ABYSSAL" difference compared to the previous one, i don't wanna joke but "i think" my escape rate previously (more or less) was around 45% like every 12 games i escape atleast like 4-5 games...with this new chapter you know how many times i escaped "IN 70+ MATCHES?" ""LITTERALLY 11 (maybe 12-13)"" (and was only versus dredge, bubba, mayers and legion) which...i would say...they are enough weak...that's the reason...i don't know you bud but what you are saying is a lie aswell, the state of the game imho is a completelly slaughter...is litterally a nightmare play survivor
"STILL" there is a good things about all this story, i still got it iridescent 3 as survivor even after this clown killer fiesta that should be rivisited by the devs, i don't play at all SWF, i can understand that if you soloQ ofc has to be really hard to play "but not impossible"
1 -
Their MMR system is trying to place players into a position of kill/escape at 50/50.
Then, for whatever inexplicable reason, they want the kill/escape in an actual game to be 60/40.
Wow, I wonder what could go wrong, this sounds perfect.
2 -
Where did you get the info of 39% escape rate.
0 -
🤣 .
0 -
I was acknowledging my Title is slightly clickbaity, but also with a gem of truth. Literally, the 39% is probably true, BUT what they are trying to say with it is not true. ALL survivors are not seeing 39%. Some high mmr survivors are seeing much higher than 39% while the rest of us are well below that.
You are acknowledging a game balance issue that hits every game, I'm not arguing that one way or another. However, I am arguing against BHVR balancing around the pro subset of top 10% of players and creating a miserable playing experience for everyone else. That does not endear a game to last. Game devs need to aim for a balanced experience for your middle 50-75% of players and it feels like that isn't being done and they released stats to try to deceive people about that problem.
Perhaps. I don't feel like they have changed anything as drastic and dramatic in the Project W to differentiate it from the massive 6.X changes, but individual experiences will vary. I personally went from 45-50% escape and kill rate pre patch to 15-20% escape rate with 75% on my killer win rate when they changed the perks and gen times and that has stayed consistent pre and post Project W. Project W just added a new killer and mostly irrelevant survivor perks (resilience sort of hinders face camping and hyper focus is a pro perk only).
Released by BHVR couple days ago: https://forum.deadbydaylight.com/en/discussion/346830/stats-6-1-0-in-review
Few other stats in there are deceptive too, like not accounting for most common perk usage by what is available. It skews representation of certain perks not because its balanced but because the stronger perks are pay walled in some cases. However, not the focus on this thread.
0 -
You know your snacks well, mister !
1 -
you cannot see your mmr. I am just trying explain it using arbitrary numbers. I think a lot of killer think they're high MMR, but in reality, they are low mmr but they end up facing players of higher-level then their skill, so they think its hard game, but its because they are not good at the game.
For example, if your 1200 MMR normally as killer, and you won a bunch of teams and your MMR hits 1300. You face 1400 team. You might think that this 1400 team is unbeatable but it is not. Its just that your skill-level with that killer has not reached a point where that team is beatable for you where another killer who is 1400-1500 MMR will find that team super-easy to beat.
Your hard games are not the same as everyone else hard-games. there is different levels of hard-games. your hard is not my hard game.
Most of dbd's imbalances happen at very very very highest level. not at middle or low-level. bad swf are not hard to beat if your good at killer. only good swf are challenging to beat.
2 -
Interesting post, I'd be very eager to see the survival/kill rate spread based on mmr and swf/solo q. Average of 39% means that the lower end of players, (which is usually a large portion of the playerbase) are probably dying an inordinate amount pf the time.
1 -
You also forgot to include how many people have been killing themselves on hook since the update.
Where dead head became a dead head
I don't think it's a lie I think it's just not correct
0 -
Once again dont go off the stats, they dont actually show any useful information as theres too many variables you dont see, a common one is a DC. While a DC is void the others are not meaning its stil a 1v1v1v1 (if you didnt know that's how behaviour treat escapes/kills, they use the 1v1v1v1v1v1 aspect). This means a killer is likely to get more kills and those survivors will die. Not to mention this rating includes all MMR brackets.
As for personal escape rates Im far higher than 39% escape rate as a solo Q. Would be even higher if some games wasnt a SBMM mess with randoms never doing a gen lol
0 -
As if no one EVER killed themselves on hook before this, apparently nerfing Dead Hard increased hook suicides exponentially.
Stop posting your redundant drivel.
0 -
I tend to only play survivor with 1 friend. All that gets me is slightly better odds I'll get unhooked. no voice chat. When I'm hooked and say leave me it's a trap they don't and we both tend to suffer for it. not all swf are equal
0 -
Pre-rework, a lot of us read between the lines and assumed solo queue would be the ones to help increase the kill rates to a level BHVR would be happy with. So yes, I would assume that high MMR SWFs are probably still doing alot better at escaping than lower solos (or just solo in general tbh) and skewering the results. Pre-rework my escape rate was around 1 in every 4-5 games. Now it's about 1 in every 8-10 games. But I expected this, so wasn't surprised.
0 -
I started my own excel to see for myself now.
Played 5 PS4 games so far today and escaped 2 times. So i am at 40% right now.
0 -
No no your partly right but I've never seen a post on the forums telling people to kill on first hook untill these changes were made
0 -
Exactly, even better if we could see how this changed with the patch.
This is false and self defeating. Common anti science argument. Something is complicated (to me), thus it is too hard for anyone to figure out. They likely have the ability to exclude all games that involved a dc, although that would be interesting to see as well. Your next argument would be suicide on hook which may or may not be easy for BHVR to rule out depending on how what data they store. Also, BHVR posted that after a dc the mmr impact to other survivors is greatly reduced.
That said, dcs and suicide on hooks have lately been about 1/4 or less of my games. Usually the game is clearly lost by that point (ie: 2-3 down at once and multiple hooks with 0 gens complete). Maybe 1 in 10 is someone quitting too early or even while winning. That's just out of my own collection of about 400 games I've logged since patch. It has gotten better than shortly after the big 6 patch. At the mmr level I play, Dead Hard and DS weren't big factors even before the patch (you can see that in BHVR's data), so most of these didn't notice that change. My theory is gen times has probably been the biggest factor. Low Mmr survivors just can't loop long enough so the killers keep finishing a chase and resuming gen pressure too fast to finish any.
0 -
They are not, but we have to adapt to the majority, and not the minority.
The majority of SWFs gain a huge compared to SoloQ survivors
0 -
FYI the average overall escape rate on Nightlight is 45% over the last 30 days. That site skews toward the above average players (since new players aren't likely to bother finding it and uploading matches to it.) If you figure that there are more people whose MMR are lower than the average Nightlight user than their escape rates are probably around, say, 35% and the weighted average over everybody ends up being that 39%.
0 -
I'm a fairly mid-low MMR solo-queue survivor. Tracking my own game stats on NiteLight, I've been seeing about a 40%-45% escape rate. That's for me personally and for my teammates. Pretty much lines up with BHVR's report.
0 -
It's a well founded assumption. We know the devs are too stubborn to balance for high level alone, so they're using statistics created from all different skill levels, skewed results, so whatever 'conclusion' they think they've come up with us gonna be wrong every time. You don't focus on top level, you're not getting the full picture that is this game's balance.
0 -
Actually, I'm making the opposite argument. You say they aren't balancing around top mmr, I'm arguing they ARE balancing around top mmr at the detriment of the lower 50% of players. They had a good escape/kill ratio pre patch, EXCEPT at the top MMR where survivors escaped all the time. So they obliterated the top survivor perks AND gave killers all around buffs (gen times, kicks, attack speed). It helped move toward balance at top mmr, but those general killer buffs destroyed any balance at the low tiers.
I suspect you are not as low as you think. The games I am in are completely unwinnable regardless of how much you carry due to 1-3 teammates that literally contribute nothing. If you are above 40% escape rate for 20-30+ games, then you are a above average player beating the global average of 39% (which is artificially high due to high escape rates of top mmr players). Ball park I'd wager 40% would put you in top 30-40% of players.
0 -
I would also like to see the stats broken down. My experience has not been 40% escape rate. Maybe I'm bad, who knows. But you can't tell me it's not a problem when the games matches you with people who just started playing at 0 prestige. And my lobbies sometimes have 6 prestige and 0 prestige in them. Why should this happen unless they are SWF?
It's not always people suiciding on purpose on hooks, sometimes it's newbies who don't know any better and hit second stage attempting to escape.
The only time I felt I was playing with people of my own skill level was before this month's reset when I reached iridiscent grade. And at times I still got complete newbies on my team. Of course my escape rate would be higher if matchmaking was more fair.
0 -
If a win for survivors is a 2 or more escape I agree, having a buffer of letting half your swf die and still secure a win would make the win rate look extremely unbalanced. I don't think we will ever see a high mmr survivor win rate % ever because the number would be very high.
0 -
....If high MMR Survivors are getting much higher escape rates than lower and medium MMR Survivors, doesn't that mean this is a Survivor-sided game and that a 39% escape rate is actually pretty reasonable considering what it is at high MMR, and that 47% escape rate was NEVER considered low to begin with?
0 -
I know this wasn't meant to be a solo-queue vs. SWF discussion, but I think that unfortunately it always boils down to that.
It's problematic for the game health/balance to be based on general statistics that group solo-queue and SWF together as 'survivors', without considering each of them are entirely different game experience's. If we exclude the tiny percentage of tournament-level SWF's, even casual SWF's have numerous competitive advantages: communication, coordination, and the fact they can all plan their perks/items before the match. In contrast, as a solo-queue player, you're going into a match completely blind with absolutely no idea about your teammates and their skill level, perks, etc. We can't underestimate the huge impact this can have on the outcome of the game, and equally, can make the difference between survivors getting escapes or killers getting kills. This is probably an unpopular opinion, but it's almost as if solo-queue and SWF need to be different game 'modes', as in there are changes to the game conditions.
This then skews MMR: I've been playing for 3 years and feel I'm a pretty decent/experienced survivor, however, if I have a few rough games and my MMR goes down, I then get stuck in this loop of getting matched with newbies and not being able to escape. This links into the OP's original point: in order to make meaningful and informative conclusions, we'd need to see the breakdown of escapes/kills based on MMR, and also, separate solo-queue/SWF.
0
