• Topic Archived
You're browsing the GameFAQs Message Boards as a guest. Sign Up for free (or Log In if you already have an account) to be able to post messages, change how messages are displayed, and view media in posts.
  1. Boards
  2. Call of Duty: Modern Warfare
  3. New study shows that SBMM is in the game.

User Info: Nordini

Nordini
2 weeks ago#1
https://www.youtube.com/watch?v=BcUzLHhdaKg

https://www.youtube.com/watch?v=qUcb58WDtVA

Drift0r and XclusiveAce just did a study to test for SBMM and both concluded that it does exist in the game. They borrowed accounts from people of varying stat levels and tested them across many games.

"It does appear that there is relatively strict skill based matchmaking in modern warfare"

Main points of the video:

  • The game must use some sort of hidden MMR system independent of your listed stats (so not K/D, SPM, etc)
  • Lifetime stats in this game do not determine skill or the matchmaking. This is due to the fact that lifetime stats in this game are obviously impacted by the very fact that sbmm exists, so two highly skilled players are constantly playing each other resulting in, on average, them having similar stats to players who are much worse than them.
  • The game does match you to a certain extent based on the average stats from your last 5 matches.
  • The game does not prioritize SBMM over connection. Connection is prioritized first. I personally did not care about this one, but I know a lot of people did.
  • Forgot to add this one: Matchmaking does also take level into consideration


Don't even want to say "I told you so". I just want it fixed.

User Info: Machine1136

Machine1136
2 weeks ago#2
Nordini posted...
This is due to the fact that lifetime stats in this game are obviously impacted by the very fact that sbmm exists, so two highly skilled players are constantly playing each other resulting in, on average, them having similar stats to players who are much worse than them.


Pure speculation.
NNID, PSN: Machine1136 (I also have a secondary account on PSN, but it's top secret)
ICE - $2B per year toward abusing human rights

User Info: seven13

seven13
2 weeks ago#3
I knew sbmm had nothing to do with peoples connections. Hopefully this will shut them up about that. I think these findings are probably the best people could hope for with having sbmm in the game. Its obviously a lot more complex than people original thought and should be the fairest for all players.

User Info: Machine1136

Machine1136
2 weeks ago#4
Have you looked at the actual data? Do you know whether they were looking at the K/D of the last 5 matches for the enemy but not themselves, or were they comparing apples to apples (last 5 matches to last 5 matches)? What was the confidence interval on their correlation coefficients? All of this type of consideration should be done before jumping on a speculation bandwagon.
NNID, PSN: Machine1136 (I also have a secondary account on PSN, but it's top secret)
ICE - $2B per year toward abusing human rights

User Info: Maestro_AN

Maestro_AN
2 weeks ago#5
as a player who almost exclusively play 2v2. i like sbmm. i was wrecking people early on. they where constantly leaving matches.

now i am lvl 40 i play with 70-120 lvl opponents.

but in regular modes, it’s bad. i almost never get killed by killstreaks above 5.

i was never killed by wheelson or century gun, 1 by a tank killestreak. a few times with vitol. saw gunner chopper once. and nothing above that. it’s stupid.

User Info: MaxxisKenda

MaxxisKenda
2 weeks ago#6
So...while being totally ambivalent because I don't know enough about it, conjecture is proof? Test all you want, you think they're going to include the instances where things didn't conform to the way they wanted them to? The guys mentioned will only post what they want to see, not what they want everyone else to see.
Proof is when IW outright state that they have it programmed in. The haven't and refuse to.

I do not know enough about SBMM to offer an opinion other than: YT is not proof.
I saw a fantastic video about the PKM and how it's the best gun in the game a few days ago. I did not see the deaths for every kill montaged, and there were hundreds of them in a very short vid.

When the guys mentioned can tell me how I get a rank 3/4/5 and a rank 155 in my lobby, maybe then they may convince me. If they have a compelling argument.

I'm not saying it's not in there, I'm not saying it is. I do know my K/D is lower than any previous CoD for no reason that I can tell other than age and reflexes.
All I'm saying is that YT is hardly the definitive answer. That lies with IW.

Now back to those Flat Earth vids I was watching..
My callsign shall be Skymonkey. Wait, Air Dragon. Wait, Air Jordan. Wait, Applesauce.

User Info: Lauz84

Lauz84
2 weeks ago#7
Since when are ranks that important.
I used to be a hardcore gamer years ago, now I'm just an average Joe, a gaming casual.

I can't say I'm that good, however I'm level 111. But I can tell you that I can get wrecked by a lvl 30 or so.. so how big of a deal are ranks? Or is the sbmm based on k/d ratios?
"Those who criticize our generation forget who raised it"
PSN: Viol8-R

User Info: josifrees

josifrees
2 weeks ago#8
Machine1136 posted...
Have you looked at the actual data? Do you know whether they were looking at the K/D of the last 5 matches for the enemy but not themselves, or were they comparing apples to apples (last 5 matches to last 5 matches)? What was the confidence interval on their correlation coefficients? All of this type of consideration should be done before jumping on a speculation bandwagon.


You should watch the video. They tested a few things. Lifetime K/D and lifetime SPM had confidences in the .20s and recent K/D of themselves and the players they faced was .83

they were able to look at the statistics of the other players in their matches 5 games back. I only got to watch the first 5 minutes or so and that is what they said so far
Quit Crying

User Info: Nordini

Nordini
2 weeks ago#9
Machine1136 posted...
Pure speculation.


More like a conclusion that necessarily follows the premises. The premises being that there is a hidden MMR, and that sbmm exists. If that is the case, then it necessarily follows that peoples lifetime stats will not be an accurate indicator of skill because they are playing people of their skill level.

If team Optic plays team Faze only, their stats are going to be all closer to 1, whereas they would all have 4+ K/Ds if they were playign random players.

Also if you want to look at the raw data:

https://docs.google.com/spreadsheets/d/18aXEK_QgM12z8EuwDZMAXcf6jIJ7JYipmqIEhkSMCGY/edit#gid=0

User Info: Nordini

Nordini
2 weeks ago#10
Maestro_AN posted...
as a player who almost exclusively play 2v2. i like sbmm. i was wrecking people early on. they where constantly leaving matches.

now i am lvl 40 i play with 70-120 lvl opponents.

but in regular modes, it’s bad. i almost never get killed by killstreaks above 5.

i was never killed by wheelson or century gun, 1 by a tank killestreak. a few times with vitol. saw gunner chopper once. and nothing above that. it’s stupid.


I actually agree that SBMM isn't nearly as bad in 2v2. In fact, SBMM is usually desirable in 2v2.

That's because 2v2 plays completely differently though. It's mean to always be "tactical" to the extreme, and you shouldn't be getting killstreaks in 2v2 every game.
  1. Boards
  2. Call of Duty: Modern Warfare
  3. New study shows that SBMM is in the game.
  • Topic Archived

GameFAQs Answers