When things don’t go so well in Overwatch’s rated mode, people like to blame it on matchmaking. That would create unfair games and “always play as a solo player against full groups” and similar reasons shoot through your head. Scott Mercer from the Overwatch team took the time to explain many details about the matchmaker.

Group size has no influence on SR. One of the biggest misconceptions in the community is that group size in rated mode would have an impact on skill rating gain or loss. This is simply not the case. Whether you sign up alone or with a group of six, the potential gain (and loss) in SR is identical.

What affects skill rating gain/loss?

• The chance of victory calculated by the system. Above 50% you earn less skill rating when you win, below 50% you earn more skill rating.
• Consistency. New accounts or those who haven’t played in a long time see larger fluctuations. This will subside with more games.
• Was your skill rating over 3000 and then “expired” due to inactivity? Then you will get more skill rating when you return, until you are close to the old level again.
• (Platinum and lower only): If you have played particularly well or poorly and thus deviate from the average of players of your rank, the skill rating increases or decreases more.
• If the skill rating is already very high, you will receive less when you win, as you approach the end of 5000 SR.

Group sizes and their matches

Who plays against whom? A few internal statistics from Blizzard should also be interesting. Mercer revealed how often group constellations played against other constellations. Here are a few exciting key data:

• Only 16% of all games consist of complete solo players.
• 28% of all matches are made up of player groups of 2-1-1-1 versus 2-1-1-1.
• If you are on a team of solo players, 73% of the time you are also fighting complete solo players. 24% of the games contain a 2 player team and only 3% of the players contain larger teams.
• Teams of six players gamble against other teams of six players in 92% of matches.

Victory rates of different group sizes

The victory rates – that is, how often certain group constellations win – are also exciting.

• Combining the win rates of all possible matchups of complete solo teams (i.e. 1-1-1-1-1), these teams have a 49.5% chance of winning.
• 2-1-1-1 teams win in 50.03% of all matches.
• 2-2-1-1 teams win slightly less, 49.46% of all games.

All numbers and matchups combined give an average win rate of 49.94% for a single solo player.

Basically, all these numbers are very close to the targeted 50%, which speaks for good matchmaking.

“Search by group” feature should remedy the situation

Despite all these quite positive numbers, the gaming experience for solo players is often lacking because the negative moments are remembered more than the positive ones. The “Search by Group” tool is supposed to help with this, so that players can find like-minded people beforehand. After all, a pre-determined group of players has many advantages:

• More positive communication among each other
• Better conflict resolution (who plays what role?)
• Stronger cohesion. Devastating defeats don’t matter so much because you have fun together