Breaking Down Matchmaking Test Results
September 30, 2026
Table of Contents
INTRO
During our last season, Overclocked, we tested two temporary changes to Ranked matchmaking. Both went after the same thing, something we hear constantly and take seriously: it does not feel good to run into a stacked squad of the game's best players, especially when their Rank feels distant from your own.
Ranked Matchmaking only works if you trust it. That means the system has to represent your skill accurately, show you honestly who you are up against, and put you in matches where climbing is possible if you play your best. Apex is a competitive game at its core, and Ranked is where players choose to test that.
Below is what we found, and what happens next.
THE PROBLEM WE WENT AFTER
Premade stacks of high-skill players can create bad matches for everyone else, especially in relatively lower-skill lobbies. We see it in the data. Across Ranked, a squad with a premade duo plus a solo fill wins 19% more often than a squad of three solos. A full premade trio wins 17% more often. The behavior concentrates exactly where you would expect: 85% of Masters squads are premade three-stacks.
&q=40)
Both tests were aimed at tackling the problem of premade squad advantages.
THE TWO TESTS
Test 1, premade tier restriction. You could still squad up, but only with players within one tier of you instead of two. The goal was to stop wide-band squads from dragging lower-tier players into high-tier lobbies.
Test 2, solo queue only. Diamond, Master, and Predator players were locked to solo queue. No premades at all at the top end of play.
For both tests, we wanted to measure two things: whether match quality actually improved, and how the change felt to play. The second matters as much as the first, and it is the one we can only answer by hearing what you tell us.
TEST 1: THE TIER RESTRICTION CHANGED LITTLE FOR A SMALL GROUP OF PLAYERS
Diamond players faced about 8% fewer Diamond opponents and roughly 8% more Platinum opponents. Platinum players saw more Diamonds in their lobbies. Outside those two tiers, nothing meaningful changed.
A small improvement for a narrow slice of players was not worth the constraint on everyone else who was used to playing with friends within two rank tiers from them. As a result, at this time, we are not making the one-tier restriction permanent.
TEST 2: SOLO QUEUE WORKED MECHANICALLY, BUT IT DID NOT DELIVER AN IMPROVED EXPERIENCE
Premade three-stacks at the top of the ladder went away but the rest did not go the way we hoped.
Who wins did not change
If removing premades squads results in more even matches, we would expect the distance between the strongest squads in a lobby and everyone else to shrink. Coordination is an advantage that RP does not measure, so taking it away should have pulled the top of the lobby back to a more even playing field.
Unfortunately, this number barely moved. In Diamond, match result distributions stayed roughly where they were. In Masters we saw a small dip in the top squads' win rate, but the win-rate gap between the top 10 and the bottom 10 teams held at 9%. The best squad in the lobby still wins, whether or not they queued together.
Exposure to top players got worse, not better
This is the result we did not expect, and it is the most useful thing we learned.
First, a brief overview of how a Ranked squad gets built. The matchmaker pulls 60 available players of similar RP into a lobby, sorts everyone by RP, and splits that list into thirds. Each squad gets one player from the top third, one from the middle, and one from the bottom. That is how it tries to keep teams even.
When we took away the ability for Masters and above to squad up, most of them did not leave the lobby. They were distributed one per squad instead of clustering into a few, which means more squads in more matches contained a top-tier player. We broke up the stacks and spread them across the board.
For a Diamond player, that reads as more matches with a Master or Predator in them than before.
It also looked worse than it was on our pre-match skill display. A squad with one Masters player and a squad with three Predators currently show up identically. So when more players ran into solo high-tier players more often, the result was more red on screen before the match even started.
This is the typical shift that was visible to players. These two games have exactly the same distribution of opponents (6 Predators, 36 Masters, and 18 Diamonds).
&q=40)
We are going to continue to iterate on this display.
Ranked play dropped among the players who play Ranked socially
&q=40)
Return rate here measures how many players came back to Apex during the test after playing in the days right before it. The weakest numbers were in Diamond and above cohorts who had been playing in duos and trios. These were not casual players drifting off. They were some of the most engaged players we have, and their Ranked play fell off during the window.
That is ultimately the part of the test we cannot accept. Ranked matchmaking gets better when there are more players at every skill level in the queue. More players means tighter bands, shorter waits, and closer matches for everybody. Any change that pushes people out of Ranked worsens the thing we were actively trying to fix. This test was meant to improve the Ranked experience. Instead it thinned out Ranked. That is a bad trade.
Where we landed
We are not rolling out the solo queue experiment as a permanent requirement for high-skill lobbies. That does not mean solo queue experiments are dead. It means this version was too broad.
What we are carrying forward is the goal without the blunt instrument: reduce how often you run into fully coordinated top-tier squads, without taking away your ability to play with your friends.
WHAT IS NEXT
Our teams continue to prepare various tests to continue to address matchmaking pain points, the details of which we’ll share as we get closer to implementation. The biggest upcoming changes that are coming as a result of these recent tests will be landing in Ranked in Seasons 31 and 32; we look forward to sharing those changes in October.
IN CLOSING
Apex is a competitive game at its core, and the best possible experience for our whole player base comes when we have players at every skill level playing Ranked. It is the mechanism by which matchmaking gets better for everyone.
Thank you for playing through these tests with us, and for sharing your feedback throughout. Keep an eye out for Ranked changes to come in the next few seasons!
Play Apex Legends for free* now on PlayStation®4, PlayStation®5, Xbox® One, Xbox® Series X|S, Nintendo Switch™ 2, and PC via the EA app, Epic, and Steam.
Follow Apex Legends™ on Twitter, Instagram, and TikTok, subscribe to our YouTube channel, and check out our forums.
Sign up for our newsletter today to receive the latest Apex Legends news, updates, behind-the-scenes content, exclusive offers, and more (including other EA news, products, events, and promotions) by email.
This announcement may change as we listen to community feedback and continue developing and evolving our Live Service & Content. We will always strive to keep our community as informed as possible. For more information, please refer to EA’s Online Service Updates at https://www.ea.com/service-updates.
*Applicable platform account and platform subscription (sold separately) may be required. A persistent internet connection and EA account required. Age restrictions apply. Includes in-game purchases.