Conclusions from data sheet
I LOVED the selection of data that made it to the dev post. I think we should be able to overall draw some conclusions from that.
- SWF groups are really, really, REALLY not as prevalent as most people here think.
- Even among the SWF groups, the escape rate doesn't change much. That goes to show that full groups are more annoying than they're strong, and most people who play SWF do so to fool around with friends, not to tryhard and make killers miserable and the role impossible to play.
- The overall survival rate goes to show what we all might've expected. The least experienced/skilled players are, the more killer-favored the game is. The only concerning point is the survival rate at red ranks, rank 1 more specifically. However, it's a sudden spike at rank 1 and it could be due to a small sample size. I'd say an escape rate anywhere between 45 and 55% would be ideal, but 40-60% is an okay range too.
- The killer popularity numbers are a bit more... concerning. If high rank players feel like they should pick Nurse so often, something's wrong.
- Survivor perks aren't as centralizing as I thought, except for low rank self care, but that's expected.
- Killer perks need to be looked into. I don't think Ruin is THAT strong, I just think it's a necessity depending on what killer you're playing (at high ranks, you ARE gonna use it if you're not Nurse, Billy or Spirit, and even those three use it sometimes). BBQ is absurdly popular too but that's probably partly because of the BPs. Maybe more interactive ways of slowing the game down through perks are necessary.
I'm not gonna touch on killer kill rates since it's not divided by ranks, only by platform.
Overall, the game seems to be in a more balanced place than expected. I think a lot of the complaints come from the fact that the game is an inherently frustrating experience when you do poorly, more so than other games. But the numbers look pretty good imo.
I think we should paste the data sheet link everytime someone asks for SWF to be nerfed. I still think some solo survivor buffs can be discussed, like making an iteration of Kindred basekit, but other than that, when you get better as survivor, you tend to be smarter around the map and more optimal, either you're playing alone or with a group.
Thanks a ton for the data, devs and @Peanits . I hope we can get more posts like this, especially about the future impact of EGC on balance and game times, and maybe some map-specific data, if there's enough of a sample size for that. A few days ago I made a post about how vital I think having game stats is for healthy and constructive discussion, and you guys delivered.
Comments
-
I'd like to see survival rates per killer, per group type (2, 3, 4man swf, solo). I'm genuinely curious if its the nurse and billy keeping their survival rates low, compared to not top tier killers
5 -
I don't agree with your take on SWF...
8 -
I however do agree on the take on SWF, not everyone tryhards, I can understand when a friend will be like "OH NO MYERS IN CHASING ME RUN!" Not making call outs to ruin the killer gameplay in general. I play SWF just to relax with my pals and get a game, I'll just follow them to a gen, and that's not wrong cause even solo players will do SWF things, like follow to a gen and work, or prioritize objectives while the killer chases.
13 -
I personally believe that alot of players just use the old SWF!!! as an easy excuse for a loss.. I guess it's easier than accepting their own mistakes.
I agree with your comments about swfs. Most I see like to meme around.
13 -
On xbox if I get a 3 man or 4 man lobby load it's usually a group of try hard bagging SWF you can tell which of the 3 are not on com. If you kill just one of them they message you something toxic. If they all escape they message you something toxic. They will even invite you to their chat party to trash talk you. Not all of them are like that of course but the majority of my end match communications are toxic and salty with swf groups. I main killer and play survivor sometimes. I was in a 3 man SWF playing the solo survivor hooked in the basement next to someone else. They unhooked their friend and left me on the hook no terror radius no urgency to run and literally left me hanging for the entity.
1 -
I don't think it's an excuse with most players, I think it's because those few SWF matches REALLY sting and stand out, which makes them feel a lot more common as opposed to the numerous forgettable mindless games.
I mean, what are you going to remember more? The five games where you went 2 kills and 2 escapes and it was just kinda shrug worthy? Or one where you were clearly against coordinated SWFs who unhook eachother the moment you turn your back, are doing gens at a blistering speed, and every move you do is being countered by three other guys?
I know as a new player, the first time I ran into that I'll never forget it. It was probably the worst experience I had on the game, but I also recognize it's not the majority of games, but it *feels* like it's more games then it should be just because it stands out so much. That one match felt like Ten, while the other matches felt like less than one.
Perspective I guess, you know? The logical side vs the lizard side of the brain.
...But I'm sure there's some killers using it as an excuse too, but, I wouldn't paint it with such a broad brush.
14 -
A very well worded and thought out post mate.
I agree with alot of what you have said. The bad games definitely stick out in the mind while the easy games tend to be forgotten or considered what should be the norm in the brain.
I merely based my assumption on the sheer number of posts and comments about SWFS!!!!!! on the forums. There is a tonne of them.
Going off that, one could be forgiven for thinking that almost all SWFS!!!!!!! are as coordinated and organised as the NAVY seals lol.
I guess its as you said of the painful games sticking out in the brain more than the fun/casual ones though.
It's just annoying to me to see so many people blame their losses on SWFS!!!!!!!!!!!
(The exclamation points were for comedic effect)
3 -
Yes, it's the same principle as remembering all the times MoM/Adrenaline/BT cost you a lot of time, pressure and kills and forgetting about all the times where they were minor annoyances.
That's why I love game stats and hope the devs do more of it.
4 -
this stats dont matter to me since at rank one in my region the only thing i face is the tryhard swf with all the crutch perks and items. same for killer players who only uses the same two killers with the most bs addons. no fun at red ranks my friend. and you never balance a game around ppl who are bad at the game btw. game is is ######### at high ranks because the balance is ######### period.
4 -
I think you forget that many SWF groups turn 0-1k games into 3-4k games with their extreme post-exit altruism.
I see that happen on a regular basis and in my games, it's usually the norm. If Survivors would just gtfo and sacrifice one of their buddies, the stats would be quite different.
6 -
It seriously took less than one page before someone tried the "but in my region..." excuse.
Stop lying. Your region isn't any different than the rest of the world. You are trying to blame your lack of ability on some SWF boogeyman in order to make yourself feel better. These stats are global, not "global except for your region of the world".
It wouldn't swing the stats that much, due to 4-man groups only being 3-5% of matches.
This data shows that the game isn't in nearly as unbalanced of a state as people on this forum try to make it out to be. It's amazing how every time the devs hand out data, it winds up dismantling a lot of the excuse making and ego stroking on these forums. This entire forum is purely confirmation bias and circle jerking, and this is exactly why you don't see a lot of the changes you want.
The changes you ask for are based on emotion, not data. It's knee-jerk reactions to having bad matches, while forgetting the multiple good matches that were had. The devs cannot take those opinions into account, as it would throw everything off. The only thing they can objectively use to balance things is hard data, not your soft ego.
15 -
This thread wasn't completely free of jabs and vitriol, but the forums have definitely changed. I quit them a few months back bc it was toxic up in here, and i returned to see these stats bc i saw many people talking about them on Twitch. I was pleasantly surprised to see some civil discussion. This is of course only based on one thread, but I want to venture forth and that is different.
3 -
I'm not just talking about 4man SWFs, the same holds true for 3man or 2man SWFs: they are suicidally altruistic most of the time and many Solo players are as well.
The statistics, as usual, lack context in how the data was obtained. Drawing conclusions from that is dangerous as they don't necessarily reflect how the game actually plays out IRL.
And another point: realise that with a 4man SWF, all slots are full, so the number of lobbies containing 4man SWFs will always be very low. Fact is that about 70 % of all in-game lobbies contain some form of SWF set-up.
2 -
I wouldve liked to see the swf per rank percentages.
Running into 4mans might be more common in red ranks as there are less people at that rank
1 -
I was really surprised with how low the amount of SWF actually is.
And Killer Perks need some love like right now.
1 -
"The killer popularity numbers are a bit more... concerning. If high rank players feel like they should pick Nurse so often, something's wrong."
Well yeah. I could tell you why just from looking at the killer roster.
3 -
First off, anecdotal evidence is, well, anecdotal. Your experiences are not hard data, and do not need to be taken into consideration when it comes to balance.
Second, if solo players are just as altruistic as SWF players, then your point is moot regarding survival rates. You said "I think you forget that many SWF groups turn 0-1k games into 3-4k games with their extreme post-exit altruism." So, which is it? Are we still blaming the SWF boogeyman for everything? Or is it just players being altruistic because of other factors such as emblems and bloodpoints? I'm pretty sure it's the latter, and not because of SWF.
Finally, all of the data combined shows that 4-man depip squads, as usual, are not anywhere near as prevalent as people here make them out to be, average game times are not nearly as short as people here make them out to be (all those 3 minute genrushes), and survival rates are sitting right around where the devs wanted it to be on average.
Stop with the 70% stuff. It obviously does not have nearly the impact that people on here try to make it out to have.
Every time the devs release data, killers have their house of cards crumble, and then they move the goalpost to try and save face. It's honestly really sad. There is hard data staring you all right in the face and you still try to angle it to suit your confirmation bias.
Seriously, you all need to finally admit that the game is not nearly in as bad of a state as you have been pushing on these forums consistently since the beginning. Is it perfect? No. But, it's not as bad as the forums try to make it look.
9 -
Let's play the devils advocate here...
If SWF isn't prevalent, then removing it shouldn't have a major impact on the game.
8 -
Nobody is making the argument it's not prevalent. People like playing games with their friends. This data is calling into question the level of impact that it has on matches, which seems to be considerably less than many people on these forums make it out to have.
But, if you look at the first data image, it's not 70% anymore. Solo players make up over 50% of the playerbase now, and all forms of SWF combined make up ~44%.
At the end of the day, the argument to remove it becomes even weaker, due to it not being as prevalent as it was once thought, and it does not have nearly the impact that these forums would lead one to believe.
The foundation of the anti-SWF argument is sand, and it's crumbling. There is no evidence that SWF needs to be removed outside of anecdotal one-offs by forum members who had a bad match or two and can't let them go.
8 -
"Overall, the game seems to be in a more balanced place than expected. I think a lot of the complaints come from the fact that the game is an inherently frustrating experience when you do poorly, more so than other games. But the numbers look pretty good imo."
You just nailed it.
3 -
My main conclusion from the survival rate is that the game is balanced for rank 8-11. There's an 80% survival rate at rank 1, which is even higher than I thought it was.
2 -
@Orion Yeah, that is a really strange anomaly that happens between rank 2 and rank 1, a 20% increase in survival in one rank is odd.
But, it does make sense that they would balance around the mid-ranks, as that is where the majority of the player base is, more than likely, playing at.
4 -
@245_Trioxin It's not odd. The new pipping changes made it extremely difficult to reach rank 1, so only the best of the best can get there. This shows exactly what everyone already knew: killers cannot win against opponents of equal skill.
And no, it makes no sense to balance around mid ranks. Every game balances around the best players, because that's what makes sense. Think about it, do you think it's fair to change the rules of chess based on how the average players fare against Kasparov?
5 -
@Orion Maybe competitive games do. But this isn't one of them. They need to ensure that the majority of players are enjoying this very casual, grindy, party game that was designed from the ground up to be viable for streamers and content creators.
It's not an esport. It was never designed to be. Trying to shoehorn balance standards from esport styled games would ruin it for the great majority of players.
6 -
Regarding the Nurse pick rate: While the Nurse is the most played Killer at ranks 1-4 on PC, the most she's ever played at any given rank is 22.39% at rank 1 (on PC). This is the highest pick rate for any Killer at any rank on any platform, but she is still only used in roughly 1 out of every 5 matches. Second would be Huntress at rank 20 on PC, at 20.18%, followed by Trapper at rank 20 on Xbox at 18.36%. The rest of the most used range between ~9.5% and 18%.
The point being, while she is the most picked, it's not a landslide victory, she's still picked in the minority of matches by far.
15 -
@245_Trioxin Every killer that leaves means another four survivors can't get into a trial. They need to balance the game and the rest will sort itself out.
8 -
actually 50% of players being solo's means that about 87.5% of games include SWF players
8 -
My point exactly. People not understanding statistics should not interpret them (looking at OP)
9 -
@Orion False. Every "killer main" that leaves will just be replaced by someone else who plays both killer and survivor.
Seriously, all of you who say this on the forums, put your money where your mouth is. Stop playing and let's see just how much the sky actually falls.
For real, do it. Stop playing. Take a week long break and let's see how much the game actually suffers.
You won't though, because deep down you know that statement is a lie.
These statements and ideas are ludicrous.
9 -
@245_Trioxin The last time there was an organized strike, survivor queue times spiked and player count dropped.
So no, not false. In fact, demonstrably true, because it's happened before.
You know what they say about those who do not know history.
6 -
@245_Trioxin It is not really strange at all.
Rank 1 doesn't have pips, you either earn a pip, black pip or lose 1. As soon as you lose, you become rank 2, and for the most part, you lose a pip when you die as a survivor (not all the time)
If you want to know real survival rates at rank 1, you should make rank 1 pipping system like every other rank. In this way, rank 1 doesn't become all or none.
7 -
@Orion Oh, the one in 2017? Can you show some actual data regarding exactly how much survivors suffered from it? I'm searching and I'm not finding much outside of anecdotal blurbs on old threads in various forums.
7 -
@245_Trioxin No data was given, but unless you think it's a massive coincidence that many players all suddenly started suffering from high queue times at roughly the same time as a large group of players stopped playing, then had their queue times returned to normal once that large group of players returned to the game, I don't see what more you'd need.
4 -
@Delfador That, I agree with completely. There should be a bit more play with staying in that rank. But, like I said in another comment. I think it reflects the issues with the current rank system. Killers do have an easier time getting to rank 1 than survivors, so the survivors that are at rank 1 are going to be the cream of the crop, whereas the killers may not be at quite that skill level. It would explain the jump in survival rate.
3 -
Well, right now it's just hearsay. I would have to just give you the benefit of the doubt.
2 -
@245_Trioxin I mean you can check the steam player count
3 -
@245_Trioxin It's either that, or you suggest that it's all a massive coincidence, including the player count.
2 -
OK, swf with VC is not OP,but not fair.
0 -
Is it safe to assume it was the 4 month period where DBD lost 50% of it's playerbase?
3 -
@NuclearBurrito Yup.
1 -
In that case. There is your hard data
3 -
The main problem of SWF is, that they are at the same time bad winners and bad losers.
If they are stronger than the killer, they start bullying (everyone who tried to learn nurse, knows what I mean).
If they are weaker, they start to disconnect. All at once.
In my opinion, they should give all survivors voice chat and then balance around that.
6 -
@Delfador You know what, Delfador? You raise a very good point that I hadn't thought of. However, even if we take the survival rate at rank 2 into account, it's still a 60-80% winrate (since we'd need to group up rank 1 players as well, but don't have enough data to do a weighted average). It's way too high.
@Peanits Can we get this discretized data in full?
2 -
Okay, here's the link to Steam Charts:
https://steamcharts.com/app/381210
You would be looking at March 2017. It wasn't a 50% reduction in the player base.
Here's the Steam forum thread, showing that the discussion started around March 3, and the protest was planned to happen on March 13. So, you cannot take January or February into account, as that is not when the strike happened.
January was the release of RE7. March was the release of the final DLC for DS3. So, there may have been at least a few other factors.
2 -
@245_Trioxin The organized strike started in March, yes, but the overall drop in players was due to killers quitting en masse, for one reason or another. And yes, there were other factors, but you have no evidence to suggest those releases were responsible for anything by themselves. I do remember players saying they were quitting to play other things due to being frustrated with DbD, however.
3 -
Regarding the survival rate at rank 1? Funnily enough I actually spoke to the data team this morning. There's a few things that affect that data:
- The data was collected between April 1st and April 7th. This was after the 2.6.0 update that made it harder to pip and before the update that would lower the survivor pipping thresholds.
- The data uses your rank AFTER the match, not before.
With that in mind, generally speaking, the survivors at rank 1 at the time were not your average survivors, they were the best of the best, the ones that could maintain that rank while everyone else struggled. In addition to that, were they to die and derank, their death would have been counted as a rank 2 death.
Generally speaking, at the time it was incredibly hard to die at rank 1 without depipping, so naturally, the survival rate at rank 1 was very high since those who did die would likely derank and be counted as a rank 2 player instead. Being able to die and not derank at that time was incredibly rare, basically.
We ran a quick test and checked the survival rates after the pipping thresholds were lowered and the survival rate was much closer to 43%.
(I've also updated the blogpost with this info. This is why we suggest not drawing any conclusions from the data, there's a whole heap of factors that could affect them.)
6 -
Most of the games including swf have duos, not 3 or 4-man teams, which DRAMATICALLY contrasts with the forum's mentality of "every game is a 3-4man swf that bullies me". My point still stands.
5 -
@Orion Your argument was that it is objective fact that the strike was the cause of a massive drop in the player base and caused survivor wait times to sky rocket.
This is not true, as the strike had less impact than that if we look at the stats.
The only reason I pointed out that there were at least a couple major releases was to show that there is possibly some other influence on top of the strike itself. January could be written off due to it being directly after the holidays. People get games for the holidays, and stop playing games they usually play to try out the new games.
Correlation is not causation. There are a number of variables that could have come into play there.
I am not saying it didn't have an impact, but I am saying that it had less of an impact than you initially implied.
3 -
Ah, my mistake. There's a couple different conversations going on here. For that data, we've only requested the most used killer for each rank per platform, so I unfortunately don't have the data for every killer. If it helps, I can share the usage rates for each in a bit, but as mentioned before, there's nothing particularly shocking.
5
