Specialization and Familiarity: Ingredients for a Competitive Edge

Hello everybody and welcome to a freshly squeezed episode of Gaming Science. 🙂

How specialized and familiar with each other should you and your teammates be to increase your chances of winning? In this episode, we discuss how specialization and familiarity impact performance.

🫂 Why You want to be Familiar

In team games, such as Counter-Strike and League of Legends, some players tend to specialize by playing the same role and/or champ repeatedly. On the other end of the spectrum, non-specialists are your average casual players who enjoy playing whatever role or champ they feel like.

Another characteristic within gaming is that gamers and teams vary in their degree of social and functional familiarity.

1. Social Familiarity: When you queue up with your buddies, you, as a team, have a high degree of social familiarity because you’ve interacted with them before on various occasions.
2. Functional Familiarity: If you play your favorite champ in League (which, of course, is Nami!) or your role (AWPer, because they're just the coolest 😎), you will have a better understanding of the "subtleties, difficulties, and opportunities associated with performing a particular role [and champ] [1]."

Teams of specialists may have a great understanding of what they’re doing individually but may lack knowledge of what their teammates are doing. For example, your very cool Nami player knows how to CC, heal, and place wards but doesn’t know how to last-hit minions. Teams of non-specialists have a great understanding of what everyone else is doing but know far less about how to efficiently play their champ or fulfill the specific role.

This is where familiarity enters the stage. Specialized teams may need to be more familiar with their teammates (socially) and roles/champs (functionally) to perform well—compared to non-specialist teams.

Therefore, this study investigates how team composition (specialists vs. nonspecialists) and team familiarity (social and functional) influence performance.

👥 Analyzing 9.2 Million Matches

The authors of the study downloaded data from Dotabuff.com, including 4,272 unique professional players. The entire game history of these players created a sample of 9.2 million matches, which included 306,949 individual players over a five-year period.

For teams to be considered socially familiar, at least three non-anonymous players had to be on the team (6,444,502 matches in total). Specialists were classified as players that usually play the same role (e.g., support or carry). Specialist teams had to have at least three individual specialists. The same criteria were applied to non-specialist teams. To evaluate team performance, the match outcome (win/loss) was used.

🧗‍♂️ If you want to Climb the Ladder You have to...

"... specialist teams are 1.75 percentage points more likely to win relative to nonspecialist... teams [1]."

In addition, an increase in functional familiarity increases the probability of a specialist team winning by 1.4% compared to non-specialist teams with the same level of functional familiarity.

"We find that team specialization and familiarity are complements—the returns to specialization increase with familiarity, relative to nonspecialist teams with the same level of familiarity [1]."

The more specialists on your team, the stronger the interaction with familiarity, and the higher the likelihood of winning. If you face a team of specialists with high familiarity, you’ve got a tough challenge ahead, as they reduce your team’s chance of winning by 1.85%.

The second, more interesting finding is that teams consisting of non-specialists (the guy on your team who plays whatever role/champ) are better at coordinating before the game but fall short to highly familiar specialized teams during the game. This makes sense because non-specific players are more "flexible" and can adapt more easily before the game. However, their lack of deep knowledge about their champ/role makes them vulnerable when it comes to adapting to the chaos and dynamics of a match.

🤔 What can We Learn from It?

As we’ve seen, teams consisting of players who know their teammates and have more experience with the champ/role they’re playing have a higher chance of winning a game. What does that mean for you?

If you’re someone who enjoys climbing the ladder and playing a specific role/champ over and over again, you want to message your buddies who know their sh*t as well and play with them. If playing multiple champs/roles brings you joy, you may have to accept that you’re more likely to get stuck in ELO hell (I’m exaggerating... or am I? 🤔).

Because playing with people you know has such a big impact on performance, it’s easy to see why team games don’t match pre-made teams against random groups. However, queuing up as four and getting matched against a pre-made team of three may give you an advantage (depending on whether you’re specialists or not).

Thanks for reading, and I hope you all have a great week.

Christian 🙂

References

[1] Ching et al., 2021