What's better
So after tonight's Q&A and hearing Patrick talk about SBMM, i wanted to see what the community things of this.
What's better 38 votes
5 votes
33 votes
Comments
-
He's right, escape/kills is the only way to determine skill
I find this discussion is often far too binary, but when he says that kills and escapes are the metric to focus on, he's right.
Personally, I think how well you scored in either the emblem system or the bloodpoints system should be used to determine how far you go up or down after that, but basing the system solely on those clearly doesn't work because that's what the emblem system of matchmaking already did.
1 -
Chasing, downing, hooks, heals, hit and everything else in between is the way
Considering it's possible to escape by doing very little and stumbling across a hatch, I'd strong disagree with that statement. This also is illustrated when looking at endgame scores, and there have been many times where a sacrificed survivor has done far better with BP.
You can strongly argue that a killer kill rate is more based on skill; moreso than from a survivor escape scenario, although even then there are probably some tweaks.
I don't know if they said exactly what the option suggested as I haven't seen it yet, but if they did then I feel that answer isn't fully supported, certainly from the survivor perspective.
1 -
He's right, escape/kills is the only way to determine skill
It's worth mentioning that the hatch does not count as a win for the survivor, it is a net zero impact on either the survivor who escaped or the killer who was hunting them at that time.
1 -
Chasing, downing, hooks, heals, hit and everything else in between is the way
That's a good point, thank you for setting that part straight for me. I'm still not sold on escapes being a true proof of skill, but at least that part takes away some of the issue. Take my upvote.
2 -
Chasing, downing, hooks, heals, hit and everything else in between is the way
That is a very interesting poll
1 -
He's right, escape/kills is the only way to determine skill
If it is based on gen time, hooks, chase time, hits, etc then it is a farmable mmr, which is just absolutely awful for any game.
1 -
He's right, escape/kills is the only way to determine skill
Should other things be taken into account? Absolutely. But the primary focus should be kills/escapes with other things effecting it to a lesser extent.
1 -
Chasing, downing, hooks, heals, hit and everything else in between is the way
Quite clear IMO.
The thing is, both dont indicate Skill. There are so many factors which make it impossible to see if someone is better just based on results.
IMO the Emblem System was not perfect for a Matchmaking, but with a few Tweaks, it could have been solid. It should have been harder to rank up for both sides, you had Killers and Survivors at Rank 1 who did not belong there. Also, Rank Reset should have been bigger.
But, the Devs are stubborn and think an MMR is needed and that their vision of an MMR is the correct one. Those two things will prevent any improvements anytime soon.
3 -
Chasing, downing, hooks, heals, hit and everything else in between is the way
Well said.
1 -
Chasing, downing, hooks, heals, hit and everything else in between is the way
I have a couple of points to make about this. If there's any statistic experts out there forgive me as I'm probably thinking of this the wrong way. I just wanted to explain where my brain goes with this.
- There's a decent amount of games I play where all survivors will escape all on last hook. That means I got 8/12 hooks, or 67% of hooks needed to "completely win" the game, or get max points. But I end up getting a complete loss.
- Lets compare that to a match where a killer tunneled and/or camped one survivor out of the game, and it's the only kill + hooks they got. They got 3/12 hooks, or 25% of hooks needed to "completely win" the game or get max points. But they end up getting more MMR points than I do? When ultimately I was 42% above them in hook rate and completion towards hook objective? By standing there and/or getting an easy down?
This doesn't really sit right with me. If I were to look at this gameplay side by side, even if I didn't know how DBD worked. I would say "one guy seems to be doing a lot, and the other is just standing there, or just focusing the easy kill. The guy that's doing a lot seems to know what they're actually doing." While I'm not really against anyone that camps/tunnels as a strategy, I think we can all agree that it's a lot easier to do that (especially with a killer like Bubba or Trickster) rather than actually playing the game without the camping/tunneling.
Secondly, I wish there was a way to change MMR similar to League of Legend's MMR system. I don't know all the ins and outs of how it works as I don't play it as competitively as DBD. From what I understand/think you gain MMR points or lose MMR points completely on YOUR performance.
- Meaning if you perform well, and your end game stats are high, you'll gain more if you win the game, or lose less if you lose the game. Sometimes the game will even completely save you the loss of your MMR points even if you lose if there was a DC, someone purposely losing the game on your side, or you simply did that good and outshined the rest of your team.
- If you perform badly, you'll lose more if you lose the game, and win less if you win the game.
I know most games with a competitive mode aren't like this. Most of them are based on an overall win/lose system. But DBD has a lot more complexities than most games. League of Legends also has really complex gameplay wise as well, which is probably why they implemented their personalized MMR system.
P.S. My first point goes along with survivor sides as well. I disagree that someone that looped a killer for many gens and ultimately dies in the end gets less points than someone who hides in lockers the whole game and ultimately ends up escaping.
1 -
Chasing, downing, hooks, heals, hit and everything else in between is the way
Not always no, that's fair. Sometimes it is. But on the whole since BP are awarded for actions of various types, it's safe to say a higher BP result means that person has done more than those below. That's the frustration with the system being refered to as skilled.
For the record, I prefer the new system more as my experience has been far better. It's just I wish they don't refer to the system as SBMM when it's really a system that works on balance in terms of results. If they drop the hang-up of skill, at least it would make much more sense. It's just causing confusion and frustration with some at a time when the game is a bit tender.
2 -
Chasing, downing, hooks, heals, hit and everything else in between is the way
Play-Style is extremely important. Judging a book by its pages is better than judging it by its cover - no matter how important the cover is.
An altruistic or unlucky survivor can be deemed 'garbage' by the system and get stuck in ELO hell until they prioritize selfish survival over winning matches and helping allies get to the exit gates.
A Killer that did nothing and managed 4 kills because the opposing team was trolling with an AFK and got cocky to NOED doesn't mean jack about their skill, but now they're going to be elevated to hardcore survivors for no good reason.
Not sure why everyone is just using W's and L's to tally in this tacked-on system when the BP scoreboard at the end is already a pretty good indicator for the score.
While they could easily implement Score Per Minute, SPM to generate an accurate number to display player skill - you could also distribute points in a ranking order.
5 people in a match - tally up their scores and put them in placement.
Place | ELO
1st | +2
2nd | +1
3rd | 0
4th | -1
5th | -2
Not as effective, but hell - at least it binds rank to in-game scoring events that can be tweaked to maintain healthy scoring. All the while it's using pre-generated in game information without relying on tracking kills and escapes to generate an abstract number.
0 -
He's right, escape/kills is the only way to determine skill
You cannot feasibly implement an MMR system with the number of variables the community wants in play. Wins are an indicator of skill, and it simplifies the process. You can't say it doesn't work because the devs said in that same QnA that the MMR system can pretty accurately predict the outcome of matches. The matchmaking is what needs work, partially because of the lack of killers that lead to increased matching parameters, creating a mismatch in MMR between survivors and killers in lobbies. People want MMR to work for them on a game to game basis, but again that's not what it's designed to do. It's designed so that, GENERALLY SPEAKING, if it works perfectly and ideally, you should be winning 50% of your games(at least according to what the devs think is a win)
0
