Visit the Kill Switch Master List for more information on this and other current known issues: https://forums.bhvr.com/dead-by-daylight/kb/articles/299-kill-switch-master-list
Sports Analogy
Gooooooood comparison with Hockey.
Both teams can put 10 goals on the board and get it into a shootout, but the ONE goal difference at the end means all 10 scored by the losing team were pointless.
Great comparison. Maybe pick a sport that can't go into an OT shootout to determine a win after an INSANELY well played match by both sides.
Comments
-
Also the way sbmm works its a 1v1 for each survivor and if the team escapes that doesn't effect the 1 person who was killed. Also players skill in sports IS RATED DIFFERENT, you can have a more skilled player on a worse team. Makes zero sense whatsoever.
4 -
It was a bad analogy.
Look at the perhaps the two most skilled players in the entire NHL. Connor McDavid and Leon Draisaitl.
Connor McDavid's stats:
Leon Draisaitl's stats:
Now, the team they play for.
As we can clearly see, goals does not equal wins. Edmonton has lost 7 straight and lost 6-0 tonight.
14 -
Not to mention that the player who stole the puck at the other end, the player that ran the puck down the rails, the player that delivered the cross right into the goal-mouth get nothing, while the statue that tapped it in gets a win.
3 -
The hockey analogy was terrible for SO many reasons.
Firstly, the NHL has changed rules over the years to encourage MORE goals like in 2018 the NHL changed the requirement of goalie equipment to be smaller. The NHL has seen that fans enjoy games with alot of goals because it's not just about who wins/loses. NHL knows the game itself has to be fun to watch to get fans in seats and watching from home. As we've told the devs we want MORE hooks/interactions to be encouraged in the game and currently SBMM does not do that - it does the opposite which is encouraging facecamping/hiding for kills/escapes.
It's a team sport 5v5 not assymetrical 4v1. The whole team wins, even if only one player scores a goal - it's not a "that one player won." Even the "win" scenarios are a 1v1 basis not a team so comparing a team sport does not work.
Winning the Stanley Cup was also a bad example since to win the cup once you're in the finals you have to win the best out of 7 games. You don't win one game with 1 goal and go home with the cup - you need to get the winning goal in 4 games.
For the Stanley Cup analogy to work evenly slightly the shots on goal would be closer to hits, not hooks. You can have 10 shots on goal without scoring like how you could hit a survivor 10 times without getting a hook. Goals would be a better analogy to hooks since you could win a single game with 1 goal or 1 hook - although if there's 1 goal in a hockey game that usually means both teams had goalie/defense playing hard where as in DBD of you got 1 hook/kill it generally means the survivors dominated the killer or the killer just facecamped one survivor and called it there. The games themselves though would be sacrifices since it takes 4 games to win a cup and you need 4k for ultimate win in DBD.
There's more reasons why it's a bad analogy but I'm going to stop there.
4 -
At the very least, we have a clear picture that kills/escapes mean a win in the devs playbook. After years of skirting around the issue, we have an answer to that.
3 -
The more questions the lead game designer answers, the more I understand why this game is so broken and will be for the foreseeable future.
3 -
Survivor 1v1 should be nerfed across the board, full stop. It's just like how they changed the NHL to give the goalies smaller pads for higher scoring games. Even if you don't kill anybody, if you downed and hooked survivors 6+ times every game, you're going to feel like you were having a back-and-forth.
The game is wildly more fun for killers if they can hit, down, and hook survivors without losing 2-3 generators per hook.
Being able to stand a chance at downing survivors discourages camping because you're not going to stay to secure one hook when you can just go off and get another one.
Having more hits means survivors have to get off of gens to save their friends, slowing down the gen progress dynamically instead of artificially putting in elaborate perks or nerfing repair speeds.
Having survivors take more hits means more health states to be healed, which gives survivors more points each game.
It's the right thing to do and the best way to make the game more fun for everyone (survivors included).
1 -
Yeah, it was a terrible analogy, with any sport. Individual players can be awesome, but if their team doesn't play well, they lose. All the players have their own stats that show how good they are: Goals, assists, shots on goal, touchdowns, yards rushing, receptions, etc... The skill of the players is still seen even if their team sucks and can't win games.
If anything, it shows exactly why the SBMM doesn't work. It is not based on the players skill. Making the analogy even more invalid is that survivors don't lose or win as a team. The best player can carry the team and die on a single hook, while the worst player can get a win by simply hiding and doing as little as possible. While the survivors should be playing as a team, the SBMM system makes it so they are playing as 4 individuals out for themselves.
2 -
Survivors throwing the game doesn't make DBD similar to Hockey. If professional Hockey players could just leave to win, they would. Only survivor mains would choose the opposite, dear friend. And I love them for it. Such creatures are beyond majestic and deserve all the protection in the world. Unfortunately it is my job to slaughter them. Quite the moral dilemma.
2 -
The problem is that if you are a very skilled player in any team sport and your team does not perform well then another team will offer you to change to them. Usually skilled players get more money and teams from higher level will lure them, bringing together players with similar skill. This way an enviroment is formed where all individuals are at an equal skill level and then winning is all that matters.
DbD does not offer this because there is no "pre-selection". An equivalent would be if the system would look at your emblems and decide to offer you an higher MMR. If you take it you will get to the "better team" and here only win/loss matters.
However in professional sports the individual performance is also closely monitored. DbD would not do this as the equivalent would be that at high MMR the system would check your emblems and then tell you: "Sorry sweety, your emblems are so bad and while the team is winning you are playing like crap and we could play much better without you. Therefor we send you back to a lower MMR."
0 -
There is no genuine comparison to make. It was such a silly analogy. You cannot compare a symmetrical team sport to an asymmetrical video game. In hockey, each teams are equal in terms of what they have available to play with and do, and each team is working toward the same goal to win. In DBD none of that is true. So comparing a team sport where each side tries to win by scoring goals, to a game where one side wins by escaping and the other wins by killing, is just silly. You might as well compare DBD and Mouse trap.
2 -
Who call bout rats? 👀
0 -
Sports are a terrible analogy to use in refards to SBMM. There are numerous metrics used in professional sports to calculate individual skill and team skill because....wait for it...wins and losses don't tell the whole story. The NFL uses DVOA. The NBA uses win shares, PER, offensive/defensive rating, etc. Baseball has WARP. The list goes on. Sports calculate the value of individual performance outside of wins and losses all the time. A great player can play on a losing team.
I can totally accept that the devs can't make the math work or figure out all the nuance. But to act like wins and losses are the end all be all of performance is laughable.
1 -
Calvin Johnson is no doubt one of the best wide receivers of all time. The man was a beast at what he did, but according to this terrible sports analogy by Patrick, Calvin Johnson wasn’t good because his team was never good.
The Lions have sucked for so long, pretty much wasting Megatron’s (Calvin Johnson’s nickname) career. That never made him bad.
4 -
Thanks for those stats, literally had no clue about this analogy when it was mentioned in the Q&A. lol
0 -
It was a terrible analogy to defend a terrible system. I was actually pretty satisfied overall with the Q&A (definitely better than past ones), but this was the low point of the stream for sure
1 -
It’s good to know that I’m officially one of the best survivors in the entire game, and with under 1k hours. Impressive.
1 -
What you are all failing to see is Kills + Skills + Hockey = JASON from Friday the 13th CONFIRMED
2 -
There is absolutely no sports that would make up a decent analogy at all. And I cant understand anyone that seriously thinks a loosy Escape/Kills would be a decent proxy for skill in DbD, while you have FOR YEARS emblems working in the background. This statement was the peak of ignorance.
1 -
Golden Knights ftw, haters mad, haters hate.
0 -
Okay, so I just reported this to Papa Patrick and everyone here just lost their cosmetics.
1 -
Personally i found that the analogy made total sense.
What he said was "shots taken lead to goals so you can cut the middle man out and only look at goals to determine the winner"
That statement is not incorrect.
But for some reason everybody heard "Hockey and DbD are practically the same game"
All analogies fall apart if you start looking too deep into it.
1 -
Well we can confirm one thing clearly Patrick never watched or played a sport in his life.
1 -
It depends from a perspective.
SBMM wants to put you in matches, where there is supposed to be “50:50” to win or loose.
So from where the thinking drops:
All the hard effort and hooks (where goals should actually be hooks) leads at the end to mmr loss.
At this point:
facecamp bubba vs altrustic teams
or
rancor + noed vs any team
Can get you easier 2K then somebody who whole match does pressuring and ends up with 8H - no camping, no tunneling
Bench warmer aka claudette in locker who does nothing and escapes gets boosted. Same as killers from above scenarios.
This promotes camping and tunneling. They also gave a hint with making BT a base kit. Which is contradictory with killer strat they promote for mmr plays.
top killer vs top swfs will be always survivor sided. And playing “fair” will be rarely achieved with 4K.
Look at optimal play of killer, 1-3 hooks = 1-2 gens done.
Mostly because of on top of gen spawns.
Matches will also never balanced by SBMM since builds and TB w/ BNPs for instance are not taken into account, and with mentality of gen rush and 0 to nothing altruistic plays, as killer you will never win.
Worst thing is, hockey is not dbd, it works differently.
Its asymmetric game, which will never be balanced nor player ratios with mmr included to make matches balanced.
AFTERALL, balanced matches and the gameflow right now, its much more intensive for killer then survivor.
This is why killers leave or swap sides. It takes too much focus and energy to play it, its not fun.
If you decide so to play casually. You drop mmr, you get matched with much worse opponents, this time you stomp them.
Did you enjoy it? Most likely not.
Have they enjoyed being stomped? Most likely not
Limbo of unhappiness
1 -
This whole thread is based on a strawman Patrick didn’t say. He didn’t say DbD was the same as hockey. Here’s what he actually said
"There's a common misconception that by just using kills and escapes to adjust MMR, it is "ignoring" your skill in the game. This isn't the truth though, this is false- kills and escapes are proxys for skill, in that skilled play inherently leads to kills and escapes; think about sports, and I'll use hockey as an example, because I'm Canadian- you wouldn't say "They only count wins when playing teams in the standing, but they should also count shots on goal too! A more skilled player takes more shots!"
"And while that's certainly true, it's important to remember that shots lead to goals, which lead to wins, so you can cut the middleman out there. And that's the key- skilled plays lead to wins. If they don't, are they really skilled plays? If your skilled play doesn't get you a win, was it really a skilled play? And that's the same in DBD- killing survivors or survivors escaping alive. That's what MMR predicts- the chance of the killer killing survivors or the survivors escaping alive, and the system is very good at predicting that.
In other words he’s saying that MMR is intended to rate your chances of winning or losing the game on average, and in DbD a win or a loss is a kill or escape. How you get the win or loss doesn’t really matter for purposes of your probability of doing it, all that matters is the end result. So the MMR doesn’t need to try to artificially track exactly how people are getting kills or escaping, it can on average look at your win loss records against other players and if you win versus higher rated ones move you up more than if you win versus lower rated ones. The only thing he said above about hockey was you wouldn’t say a team “didn’t actually win” if they played a certain “cheesy” way but still got more goals, or if they took less shots than the other team, or that a team didn’t win if they only got 1 goal but the opponent got none, etc.
Really the only sort of new thing Patrick implied was the devs do indeed consider kills and escapes to be win conditions. Everything else are just tools to get those. Previously they’ve been intentionally vague about what a win or loss is in their mind, “leaving it up to the player to decide if they won or lost”. That’s why they have all these other weird systems in place like bloodpoint scores and emblems and pips that measure all the other smaller things that happen in games. So a player can look at their score and say “yeah I died, but I still did really well because I got almost 30k points and one pip!” or whatever.
The whole goal of MMR though is to make it so that survivors have about a 50/50 shot of escaping by rating survivors and killers on the chances of escapes and kills happening. That’s it, that’s all it’s meant to do. So all these other things that happen in a game that make it “more exciting” or whatever don’t matter for that purpose, it just needs to make adjustments over time based on kills and escapes and the tools that get those will on average already be influencing those results.
And from what I can tell the MMRs are actually accurate in that if everybody in a match is roughly the same MMR then the match will be close, with the exception of brand new players that might start with an initial rating that is too high. The matchmaking system itself though could be inaccurate beyond that if maybe it is using too simple a formula estimating the collective MMR of groups of survivors which has a wider gap in individual ratings. If it’s just doing the average of the individual ratings it could well be that that formula isn’t accurate because for example maybe small gaps make a big difference but once a gap is large there are diminishing returns for one player being awesome and everybody else being terrible or vice versa. So I am curious how the system calculates the optimal killer rating to go against a given group of survivors based in their individual ratings.
0 -
Ummm...no
That would be like saying riding a bicycle is like riding a motorcycle because you just need the balance. So, since you can ride a bike, you can just hope on any ol' motorcycle and be on your way.
Except, oh wait, you're going 80 mph and if you fall you can't just get back up...and, oh wait, there a motor on this device...and, oh wait, you have to obey traffic rules...
So, we realize they're nothing alike. Therefore, the analogy makes no sense. It's not thinking too deep. It's being realistic about the problem.
"Just using kills and escapes to adjust your MMR - it's ignoring skill in the game...this is false"
If you're trying to prove that MMR being based on kills/escapes ignores skill, then you should use an analogy that makes sense. There were so many factors that were left out and we're left with still wondering what the answer to the initial question even was all the while still getting horrible matches.
1 -
And this is another misconception that people make about mmr. They hyperfocus on the skill part
Mmr is first and formost a matchmaking tool. Not a skill calculator.
Sure a noed+rancor is going to get easier kills then a "fair" killer but if that noed+rancor gamer gets constant 4k's with that build. Do they need to go up in ranking and face stronger opponents who will punish the lack of perks early and who as obsession will make sure they aren't targeted when the last gen pops or weaker opponents that they can crush easilly again?
Same with hide and do nothing claudette's. Do they have to go down in rank and face weaker killers? Or go up and start facing killers who don't lose a 3v1?
Your skill isn't the only thing that matters in matchmaking, your opponents skill matters as well. If you win with "cheese strats" you need to start facing people who can deal with that. Regardless of how little skill those strats take
0 -
MMR balance will never be achieved via 4:1 ratio.
Lets keep it in mind. The higher you get, the less you enjoy the game. To be every game the same, from 5 meta perks to speed of gens, same strats on both sides, just wanting to win. It brings the burnout syndrome, you dont ENJOY the game. Eventually you swap side - less stress or stop playing the game.
Why is there much less killers? People dont enjoy the role in current meta and matches.
How will SBMM solve this problem? It cannot. It will just get a harder job with creating aprox balanced matches with reasonable queue times
1 -
I think it was a bad analogy for many reasons, but sometimes we make bad analogies when trying to express something on the fly. I hope that whatever new mode they were hinting at will be aimed at casual players they have alienated with SBMM.
0 -
I mean that the higher you get the less fun it is isn't mmr's fault.
It just highlighted it.
You can change the mmr system all you like those matches will not suddenly be more fun. They'll just be less frequent and replaced with stomp matches which also aren't fun.
Mmr's job is putting people into brackets and for the most part it's doing a good job with it.
That the highest bracket is miserable has nothing to do with mmr and is up to the devs to fix those problems.
People are just yelling at the wrong thing
0 -
the only thing that really bothered me about the hockey analogy is it was unnecessary, Patrick said the right thing during the last QnA, its impossible for that system to capture that nuance (which is true). I also disagree that hook cant be factor (from ideal of concept. programing it in is another story). Basically a survivor who escaped without being hooked once should go up in MMR higher than a survivor who did, just like the killer should lose less MMR if survivor who got hooked twice escaped.
0 -
Actually, that's a wrong mindset and a wrong analogy. A player or players are not determined by their team, so their team might be bad yet that doesn't mean they are... also, even if you lose. Is that a bad thing? Many people who are not athletes believes they can judge it and say "oh, but he/she is second, she lost" when you do not see how much time they've been training. As an example, I do like boxing a lot and I practice boxing myself as well... a well connected punch can make a great fighter fall. Does that mean that everything else didn't count? Many will say "he lost by KO in the first round. What a loser" but actually, it's getting there and fighting what matters, training everyday, following a routine... if you're there is because you trained hard for it and you're not a loser, you lose a match which is different.
1
