Call of Duty SBMM Explained: How Matchmaking Really Works (MW4)
"Call of Duty SBMM" is what players call it. Activision calls it matchmaking and says skill is only one part of it. This guide sticks to what Activision has published: its matchmaking white papers and Call of Duty blog posts. MetaPatchAI did not test anything. Claims from players and creators are marked [Community claim].
Steps: what you can actually control (do these first)
- Get your connection right. Activision says connection ("Ping is King") is the most critical and most heavily weighted factor. Play on a stable, low-latency connection. Source: White Paper #2, https://www.activision.com/cdn/research/CallofDuty_Matchmaking_Series_2.pdf
- Play more than one playlist. Activision lists time to match as the second most important factor and also names playlist diversity. Fewer players in a queue gives the system fewer options. Source: same white paper.
- Pick Ranked Play if you want skill to count openly. In Ranked Play, every player has an SR (Skill Rating). You gain SR for wins and lose it for losses. Your personal performance and the margin of victory change how much. Source: Matchmaking Intel White Paper #3, https://a.atvi.com/RPMatchmaking
- Know what "skill" means here. It's a prediction of how you will perform based on recent and past matches, not a fixed label (see below).
What Activision says matchmaking uses
From the January 2024 blog post "An Inside Look at Matchmaking" (https://www.callofduty.com/blog/2024/01/call-of-duty-update-an-Inside-look-at-matchmaking): "Call of Duty does consider skill (or more specifically player performance) as a component... but skill is not the dominant variable."
From White Paper #2, "Matchmaking Series: The Role of Skill in Matchmaking" (July 26, 2024, per Activision Research: https://research.activision.com/publications): - Connection is the most critical and most heavily weighted factor. - Time to match is second. - Other critical factors include playlist diversity. - Activision defines skill as "how well a player can be expected to perform against the rest of the population in a given game mode, based on their previously observed performance." - The skill value is meant to be bounded, predictive, summable (so a party's average predicts the team), quick to adapt, and resilient when the player population shifts. - Skill is also used for team balance, because blowouts make players quit.
What Activision says happened when skill mattered less (White Paper #2)
- MW2019 looser-matching test: Activision reports "a significant decrease in the number of players playing Call of Duty: Modern Warfare (2019) and an increase in the overall match quit rate, when treated with a looser skill matching."
- Protecting only the bottom 25%: a follow-up test kept skill matching for the bottom 25% of players and loosened it for the other 75%. Activision says this "also had clear negative effects on player counts in two weeks, with increased quit rates, and reductions in total hours played."
- Deprioritized Skill Test: the returning-player rate fell significantly for 90% of players, and only the top 10% came back more. The quit rate rose significantly for 80% of players. Kills per minute fell significantly for the bottom 20–30% of players, while the top 10% saw higher KPM.
- MWIII tightening test: with skill matching made tighter, the quit rate fell for 90% of players. It hurt high-skill players, though, so Activision says the change "was not rolled out as a standard approach."
- Activision's overall claim: "When skill is utilized in matchmaking, 80-90% of players experience better end-of-match placement, stick with the game longer and quit matches less frequently."
Is MW4 different?
Not confirmed. We found no MW4-specific matchmaking document from Activision. Everything above comes from the 2024 white papers and blog posts. Activision's third paper, "Call of Duty Matchmaking Intel – White Paper #3" (https://a.atvi.com/RPMatchmaking), covers the Ranked Play SR system (the source for step 3), not MW4 public matchmaking.
Community claims (not from Activision)
- [Community claim] "Reverse boosting" and stat-padding lobbies prove matchmaking uses only K/D. Activision's papers say skill comes from observed performance and is weighed with connection and time to match. They do not publish the exact inputs or weights.
- [Community claim] Creator analysis videos (Drift0r, below) give opinions on the white papers. Treat them as commentary, not official statements.
Confirmed / Not confirmed
Confirmed (Activision): - Skill is one factor, and Activision says it is not the dominant one (blog, Jan 2024). - Connection is weighted highest, then time to match (White Paper #2). - Skill is used to balance teams (White Paper #2). - Ranked Play uses SR, with gains and losses affected by performance and margin of victory (White Paper #3).
Not confirmed: - MW4-specific matchmaking changes. - The exact formula or weights for skill. - Any "SBMM off" playlist in MW4.
Sources
- Activision, White Paper #2 (PDF): https://www.activision.com/cdn/research/CallofDuty_Matchmaking_Series_2.pdf
- Call of Duty blog, An Inside Look at Matchmaking: https://www.callofduty.com/blog/2024/01/call-of-duty-update-an-Inside-look-at-matchmaking
- Activision, Matchmaking Intel – White Paper #3 (Ranked Play): https://a.atvi.com/RPMatchmaking
- Activision Research publications: https://research.activision.com/publications
Related
Up: Gaming
