{{
`${userDetail.user.displayName}${userDetail.user.cachedBungieGlobalDisplayNameCode ? `#${standardizeBungieGlobalCode(userDetail.user.cachedBungieGlobalDisplayNameCode)}` : ""}`}}
{{getUserClans()}}
Editado por Lost Sols:
11/29/2018 6:57:44 AM
42
QuickPlay Matchmaking By the Numbers (or how 10% dictate the experience of the other 90%)
So I've been posting a lot on the state of matchmaking in quickplay and how FUBAR it is since all skill factors have been removed. I've posted [url=https://www.bungie.net/en/Forums/Post/249728850?sort=0&page=0]my Iron Banner experience[/url] as well as experiences [url=https://www.bungie.net/en/Forums/Post/249787911?sort=0&page=0]from select QP games[/url].
To all of this I have heard the same false accusations against sbmm that have always been leveled as well as claims that the numbers were random and not indicative of the entire experience, so I've been playing more QP to give a deeper field of view on what is really going on and again, I think people will find the numbers very interesting.
[quote]Games Played: 40
Record: 20-19-1 (the stats from the tie not included as both teams were placed together in the stats)[/quote]
[quote][b]Control[/b]: 24 games (62%)
Win/Loss: 12-12
Avg Score: 141-84
Avg Victory Margin: 57 points
[b]Avg Combined Elo[/b]
Winning team: 7831
Losing Team: 7119
[b]Avg individual Elo[/b]
Winning Team: 1305
Losing Team: 1187
[b]Mercies[/b]: 3
Closest Game: 127-124
Worst Blowout: 155-50
[b]Highest Rated Player[/b]
Winning Team: 19 games
Losing Team: 5 games
Largest Player Elo difference in a match best to worst player: 1140
Closest Player Elo difference in a match best to worst player: 211
Avg Top Individual Elo
Winning Team: 1612
Losing Team: 1330
[b]# of players by rank:[/b]
Winning teams-Losing teams
Bronze: 9-32
Silver: 91-89
Gold: 18-21
Platinum: 16-2
Diamond: 9-0
Master: 1-0
In 14 of 24 Games, the winning team had at least one Platinum or better player vs 2 of 24 for the losing teams.
Bronze, Silver and Gold ranks make up 260 players vs 28 Platinum, Diamond and Master ranked players.
That is 90% versus 10%[/quote]
[quote][b]Clash[/b]: 15 Games (38%)
Win/Loss: 8-7
Avg Score: 88-60
Avg Victory Margin: 28 points
Closest Game: 98-92
Biggest Blowout: 64-13
[b]Avg Team Combined Elo[/b]
Winning Team: 7981
Losing Team: 7447
[b]Avg Player Elo[/b]
Winning Team: 1330
Losing Team: 1241
[b]Team w/ highest ranked player[/b]
Winning Team: 11 games
Losing Team: 4 Games
Mercies: 2
[b]# of players by rank:[/b]
Winning teams-Losing teams
Bronze: 7-11
Silver: 41-54
Gold: 31-19
Platinum: 5-3
Diamond: 4-3
Master: 2-0
In 6 of 15 Games, the winning team had at least one Platinum or better player vs 3 of 15 for the losing teams.
Bronze, Silver and Gold ranks make up 163 players vs 17 Platinum, Diamond and Master ranked players.
That's 90.5% versus 9.5%[/quote]
A lot of interesting numbers that aren't surprising, but there was a number that was. The team with the highest combined Elo won 23 of 24 Control games.
MM is literally determining who wins.
as expected, lack of SBMM hurts Clash less, but still heavily favors the team with the highest combined Elo as they won 73% vs 96% in Control.
It's also odd that the average team skill difference is higher in Control than Clash and that the game puts us in Control more often, even though it's much more punishing with no sbmm. Also once more, Control is broken as well. 57 points is an insane average victory margin.
So essentially 90% of the population are having their experience dictated by 10% of the population.
Sounds all too familiar. The game is for everyone and that includes Crucible. If players are going to be tasked to play Crucible to level, do clan bounties, grind things like Ace of Spades, etc, then it needs to be a place they can go and not just get rolled by players they have no business being matched against.
-
55 RespuestasEditado por biggestLOUser: 11/29/2018 3:29:10 PMI love the arguments people have against SBMM 1) I’m really good so why should I have to sweat all the time? (But Not realizing that it makes the other 90% of the player population have to sweat every match just so they can “relax”) 2) No SBMM is better than lag every match (guess what, with an atrociously low player population lag may be an issue from time to time depending on region and time of day, but notice how when the player population is super high, a vast majority of the “lag” complaints go away) 3) Git Gud (My favorite, well guess what, not everyone will be able to get as good as an MLG 3.0 KD 7500 ELO 987x flawless crucible god, every human being is different, and everyone has a skill ceiling, that’s just fact) 4) People want everything handed to them, to be protected (No, people don’t want to be protected, they want to have FUN, in a GAME, and who has any fun getting pub stomped into the dirt every match?) So let’s just call it like it is: The Pro’s want to be able to drop 50’s on the other 90% of the player base, and those players experience be damned. Who cares if people avoid the crucible like a leper colony, as long as I can cut my YouTube clip I’m good. Well done on the #’s OP, a very interesting read.
-
Bungie better come up with a good response to this because its about damn time we get a fix to matchmaking. These numbers reinforce my own gameplay experiences quite a bit. Hell back in Season 1 I had a game where me and 3 golds were put up against a full team of randoms who were all diamond. Not even cool.
-
2 RespuestasI totally agree, the PvP side of this game has been ruined by a small number of sweaty nostalgic D1 players who pretend to dictate how the game should work because they believe their opinion counts more while the overwhelming majority of players has awful PvP experience and struggle to even have fun
-
I did a similar breakdown on one of my matches that was a lopsided, 150-52 loss. The winning team had a player with an ELO of 2131 (Grand Master?)----whose profile showed that he literally does nothing in the game except play Crucible and Gambit----in a lobby of one Gold player and the rest Silver and Bronze. He "relaxed" his way to an Efficiency of 7.25 for the match (the next highest was 2.35) and 29 kills. Nine more than the next highest on his own team. DOUBLE that of the highest scoring player on my team, where the highest efficiency was 1.15. [quote]The game is for everyone and that includes Crucible. If players are going to be tasked to play Crucible to level, do clan bounties, grind things like Ace of Spades, etc, then it needs to be a place they can go and not just get rolled by players they have no business being matched against[/quote] THIS.....is especially why this game needs skill matching. How many people's efforts to gain access to the Malfeasance quest is my Grand Master ruining by wracking up team wipes in Gambit? People who are so relatively unskilled in comparison to this guy that they're dying like grass in prairie wildfire??
-
Editado por Orpheus49: 12/2/2018 3:05:51 AMI'd rather have a better connection. At least it makes the game playable
-
21 RespuestasEditado por Optimus_immortel: 11/30/2018 10:25:18 AMRegardless of the numbers quickplay should never have skill tied to match making. If you want to be facing skill v skill to improve that's when you play comp
-
Also, if you can't find a way in this environment you never will. This pvp has seen it all over the years and it's never had so much at your disposal. Time to acknowledge and accept where you are.
-
11 RespuestasALERT - SBMM IS NOT WHAT YOU THINK IT IS!!!!!!! The creator of SBMM works for 343i now. His name is Josh Menke. He has multiple tweets describing what SBMM actually does and here are some of them - https://twitter.com/joshua_menke "Trueskill (SBMM) will simulate your average kill vs certain players, then pit you against them with players to either help or lower your overall kills as a team. The system is rewarding players at a 50% win ratio" "We can change the system but the players in social will complain" "My system does not put you against players of equal skill, it puts you with players you can match with based on connection and their past few games. As you play more the system finds you skill, and adds players to your team to help have a 50% win ratio vs the other team" In other words that system does NOT put you WITH people in a skill bracket. BUT put you with WORST players vs BETTER players than them. So here is an example - 4 kd .5 .5 .5 vs 2 kd 2 1.5 1.5 Now the system thinks - Hey that 4 kd guy will average XX amount of kills vs a 2 (Should be a 2KD atleast)...Problem is it gave me .5's on my team to compensate and make my chances of winning literally none. .5's will not average .5's vs 2 KDs. They will get smashed and I will have to literally 1v4 this game and sweat so hard just to win. It wont happen. The .5's will quit and come to the forum to blame matchmaking, while I will come to forums yelling about SBMM. So what you want to be in part of the .5 players TRYING to be carried by the 4.0's? Sounds no different that what you have now kiddo
-
Editado por Willakkuma: 12/1/2018 2:51:19 PMElo is an outside 3rd party metric that IS ONLY, AND NOTHING MORE than a wildly fluctuating reflection of game mechanics. End of file. For YEARS now I've seen elo rage between Diamond across the board to 1300, depending on how many losses I'm supposed to take while pulling toward the 50/50 w/l theoretical co-op equilibrium. All the while.....my game changes very little in fact personal stats steadily climb. Elo is a moot point in your argument. Understanding this is Destiny PVP 101. Get with it.
-
21 RespuestasQuickplay is meant to be like this. Its how its designed. Adding SBMM to it ruins the connections. Remember D1? The lag was horrendous and unacceptable because of SBMM. Comp matchmaking doesn't exist. Its basically quickplay with even more 4 stacks.
-
-
Good job.
-
Thanks for the data, Didn't know it was this bad.
-
1 RespuestaAll they need to do is create a solo que only and chances are you wint get curb stomped near as much
-
13 RespuestasEditado por Kyreek: 11/29/2018 8:46:40 PMActually 90% of the population are having their PvP experience dictated by their uselessness in Crucible. Fact is 90% of players in crucible are bad, worst than bad. And i’m not talkin bout havin a pro gamer MLG cyborgs’ aim. Enter the crucible and take a look around you. No game sense, no map awareness, unadapted spec, shit loadouts, crap gear, wrong mods, peoples at level 12, never holding heavies, never sliding, never jumping... Hell, 4 years of Distant Shores and there are still peoples running to cap A with 4 mates (-blam!-in useless move) and leaving B, C and the heavies to the opposite team. How many times i had to fight one, two, three opponents just to get kill (by a margin) by the fourth enemy while having absolutely no support? Because my team is filled with headless chickens running in circles, wondering why the circle in the top left of the screen is glowing red? Funny thing : those peoples want Redrix, they want Luna, they want Not Forgotten but they don’t want to improve. Most of players will have a better PvP experience when they’ll stop being bad.
-
1 RespuestaHere's an anecdote for your thread, very fresh still steaming: Two matches ago as of now I actually got teamed up with a three man team of [Last Rites], Qaa51mB, MightGuyy and LokiSlays. I am an absolutely average player, maybe even worse than that, why are these people even anywhere close to being considered to be put near the shadow of a thought of a team match up for me? I am against skill based matchmaking but how about we introduce commonsense based matchmaking? It was a free win but it was boring beyond belief. I might as well have stayed in spawn they did just fine without anyone else on the team.
-
6 RespuestasEditado por Megrallo: 11/30/2018 1:40:46 PMHey @Lost sols. After our debate at Twitter (was very toxic from my side, sorry for that) i thought a lot about the issue. I now can understand what your problem is and that you want MM to be changed. But please think about the "10%" players too. For me personally SBMM makes me dont wanna play PVP anymore. Imagine having a 3.0 KD in QP and your opponents ALWAYS are the best of the best in casual QP. This isnt fun aswell and QP isnt meant to be this way(in my opinion). You always have to play the meta to compete, you always have to try your hardest to compete and the game forces you to lose 50 % of your matches because of this 50/50 winratio crap(bad connections included). I like mixed lobbys ( idont want to stomp all the time, believe me or not) I guess MM will never be perfect,it isnt possible to make everybody happy and i really dont know what the best solution would be. Sorry for my poor english. Peace
-
27 RespuestasSounds like it works well. If IB uses sbmm then it works for casual pvpers (who only really play when IB is on anyway), while making it feel like an endgame event for the hard-core. The matchmaking for quickplay seems to work well too. When sbmm was in quickplay you had like 200/300k players (destiny tracker), whereas since the change it has stayed at 800k/1m. Not only that it looks like a higher share of the overall playerbase now plays PvP. If the 90% didn't like it you'd think the amount of crucible players would decrease and not increase. Bungie has always wanted sbmm, so the fact they haven't reverted back speaks volumes.
-
1 RespuestaPvP has a feel that Bungie was trying to force me into a 50/50 win/loss ratio. I'd say your data supports that. I think it's crap that they have win streaks for a situation that is basically predetermined, in other words I should just not even care and just go along for the ride. (probably some good advice for myself though really)
-
SBMM DOES NOT MATCH YOU WITH EQUAL SKILLED PLAYERS!! I have posted numerous times about SBMM and how it is designed to win 50% of your games, designed to DETERMINE your kills based on feedback from previous games. This makes it so you either have to carry terrible players (Regardless of your skill) or you will win by being on the right side of the 50%. CBMM makes for better connection + same odds of playing players as everyone else. Ill play a Top 1%... at 1% of the time. SBMM is a TERRIBLE system that tells players to not play the game if you are good unless you run in 6's. If they put SBMM in the game, players who are playing solo will now run in 6's... and you will be back here complaining again.
-
This game has the worst pvp ever. And trust me, I played a lot of pvp in many different games.
-
https://www.youtube.com/watch?v=JLYeQrmwcmM
-
2 RespuestasWhy do I constantly get stuck on a shitty team. Pisses me off
-
7 RespuestasI would much rather have bad match ups than bad connection any day. Sbmm makes Destiny so bad the skill level you play against doesn't even matter due to how often lag could kick in. For all I care, I'm smack right in the middle of all of this. Sometimes I stomp, sometimes I get stomped, sometimes I get close games, but I will ALWAYS take that randomness over the randomness lag brings. They need to create a better system for matchmaking as having one or the other is COMPLETELY idiotic. Unless they do so, I'm completely against sbmm for quickplay, especially since it's 6v6.
-
I don't ever play crucible anymore. Get rolled by a team of 6 or match up against the thumbless. Either way its not fun.
-
4 RespuestasSols, Could you run this same data gathering in competitive as well? It will give us a better feel as to how well the SBMM works there and could serve as a form of control group for this discussion.
