💻 computer science

"Can You Play Anything Else?" Understanding Play Style Flexibility in League of Legends

This study analyzes the relationship between play style flexibility and team success in League of Legends, revealing that adaptability varies by skill level and preferred style while significantly influencing match outcomes and highlighting the strategic balance between specialization and flexibility.

Original authors: Emily Chen, Alexander Bisberg, Emilio Ferrara

Published 2026-03-18

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Emily Chen, Alexander Bisberg, Emilio Ferrara

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine League of Legends (LoL) as a massive, high-stakes orchestra. In this orchestra, there are 168 different instruments (called "champions"), and every musician (player) has to pick one to play. Some musicians are virtuosos who only play the violin, while others are jazz improvisers who can switch between the sax, the trumpet, and the drums depending on what the song needs.

This paper asks a simple but profound question: "Can you play anything else?"

The researchers wanted to know if being able to switch instruments (being flexible) helps a team win, or if sticking to one instrument (being specialized) is the secret to success. They also wanted to see how players change their minds over time.

Here is the breakdown of their findings, translated into everyday language:

1. The Two Types of Musicians: Specialists vs. Generalists

The researchers invented a "Flexibility Score" to measure how many different instruments a player uses.

The Specialists (Low Flexibility): These are the "One-Trick Ponies." They pick one champion and play only that one. They are like a chef who only knows how to make pizza. They might be incredibly good at it, but they can't make a salad.
The Generalists (High Flexibility): These players try out many different champions. They are like a food critic who eats everything on the menu.

The Big Discovery:
In the lower ranks (the "casual" crowd), players tend to be more flexible. They try everything. But as you get better and reach the Elite ranks (the top 1%), players become much more specialized. The best players tend to stick to their "pizza" because they have mastered it.

2. The Paradox: Specialization Wins, But Flexibility Grows

Here is where it gets interesting. The study found a strange contradiction:

To Win Right Now: Being a Specialist helps you win more games. If you want to climb the ladder quickly, pick one champion and master it. The data shows that players with lower flexibility scores (specialists) actually have higher win rates.
To Grow Over Time: Being a Generalist helps you evolve. The researchers tracked players over time and found that Elite players actually started trying more new champions as they got better. They became more flexible over time, even though it might cost them a few wins in the short term.

The Analogy:
Think of it like a video game character.

Non-Elite players are like beginners who keep trying new weapons but never get good at any of them. Over time, they get bored and stop trying new things (their flexibility goes down).
Elite players are like masters. They start by mastering one weapon to win tournaments. But once they are the best, they get bored and start experimenting with new, weird weapons just for fun or to stay sharp. They become more flexible because they are already so good.

3. The "Role" Matters

Not all positions are created equal.

Players who like Assassins (high-speed attackers) or Marksmen (snipers) tend to be more flexible. They are willing to try different variations.
Players who like Supports (healers/helpers) or Tanks (defenders) often fall into two camps: either they are super-specialized "One-Trick Ponies," or they are super-flexible. There is very little middle ground for these roles.

4. What This Means for the Real World

The authors suggest that this isn't just about video games; it's about how we work in teams.

In a Crisis: You want a Specialist. If the building is on fire, you want the firefighter who has done only that job for 20 years, not the guy who tries to be a doctor, a plumber, and a firefighter all at once.
In a Changing World: You want a Flexible person. If the market changes or the rules shift, the person who can adapt and learn new skills is the one who survives long-term.

The Bottom Line

The paper concludes that there is a delicate balance.

If you want to win today, specialize. Stick to your strengths.
If you want to grow and adapt for tomorrow, be flexible. Try new things, even if you aren't perfect at them yet.

The most successful players (the Elites) seem to understand this duality: they specialize to win the game, but they stay flexible enough to keep evolving as the game changes.

1. Problem Statement

The study addresses a gap in existing gaming research, which has historically focused on matchmaking, user archetypes, and team communication. The authors investigate player flexibility within League of Legends (LoL), a competitive Multiplayer Online Battle Arena (MOBA). Specifically, the research seeks to understand:

How to operationalize and measure a player's willingness to adapt to different roles (play styles).
The relationship between a player's flexibility, their skill level (rank), and their preferred play style.
Whether flexibility impacts match outcomes (win rates) and how adaptability evolves over time (temporal flexibility).
The trade-off between specialization (mastering a few roles) and flexibility (playing many roles) in competitive environments.

2. Methodology

Dataset Construction

Source: Data was collected via the Riot Games public API from the North American server between June and September 2023.
Sampling: Users were sourced from third-party tracker OP.GG. To balance the dataset, the authors stratified users into Elite (Challenger, Grandmaster, Master) and Non-Elite (Iron through Diamond).
- Sampling Strategy: 100% of Elite Challenger/Grandmaster, 50% of Master, and 0.5% of Non-Elite users were retained as seeds to ensure a representative sample despite the top tier comprising <1% of the population.
Scope: The final dataset includes 4,018 elite and 5,621 non-elite users. Analysis focuses on users with a minimum of 100 games in specific queues (Solo Queue, Flex Queue, or All).
Game Modes: Primary focus is on "Solo Queue" (highly competitive, restricted pairing) and "Flex Queue" (broader skill levels).

Metrics and Definitions

Functional Play Styles: Based on Riot's 6 official categories (Assassin, Fighter, Mage, Marksman, Support, Tank) and 168+ champions. The study defines 21 "functional play style" pairs (primary + secondary) for categorization.
Flexibility Score ( $F$ ): A normalized metric calculated using the distribution of a player's games across the 21 functional play styles.
- $P_i$ : Percentage of games played in style $i$ .
- $P_{max}$ : Percentage of the most played style.
- $P_{norm}$ : The average absolute difference between $P_{max}$ and all other $P_i$ .
- Formula: $F = 1 - P_{norm}$ .
- Interpretation: A higher score indicates greater flexibility (even distribution); a lower score indicates specialization (narrow focus).
Temporal Flexibility: Measured by applying a rolling window of 10 games across a player's chronological history. A linear regression is fitted to these sequential scores to generate a Temporal Flexibility Coefficient (Temp Flex Coef), indicating the rate and direction of change in adaptability over time.

Statistical Analysis

Multivariate Linear Regression (MLR): Used to predict:
1. Solo Queue Win Rates (based on flexibility and skill).
2. Flexibility Scores (based on win rates and engagement).
3. Temporal Flexibility Coefficients (based on skill and game history).

3. Key Results

Flexibility Scores and Skill Levels

Inverse Correlation with Rank: Elite players consistently exhibit lower flexibility scores than non-elite players. As queue competitiveness increases, flexibility scores drop for all users.
Role-Based Variance:
- Players favoring Assassin-Fighter, Marksman, and Mage roles tend to show higher flexibility.
- Players favoring Mage-Support and Fighter-Tank roles show a bimodal distribution. In elite ranks, this bimodality is driven by a distinct cluster of "One Trick Ponies" (highly specialized, low flexibility) alongside a cluster of highly flexible players.
Elite Specialization: Elite players generally specialize more than non-elite players, suggesting that mastery of a limited set of roles correlates with higher competitive success.

Temporal Flexibility Trends

Divergent Trajectories:
- Elite Players: Show a significant increase in flexibility over time. They are more likely to experiment with new play styles, even if it temporarily impacts competitive success.
- Non-Elite Players: Show a decline in flexibility over time, tending to settle into specific roles.
Rate of Change: The rate of flexibility increase among high-ranked players is over six times faster than the rate of decrease among low-ranked players.

Predictive Modeling (Win Rates vs. Flexibility)

Win Rate Prediction: Both skill classification and general flexibility score are inversely correlated with Solo Queue win rates.
- Finding: Lower flexibility (specialization) correlates with higher win rates. This suggests that in high-stakes competitive environments, deep mastery of specific roles is more valuable than broad adaptability.
Flexibility Prediction: Win rates and the proportion of ranked games do not significantly predict a user's flexibility score. However, skill classification is the only statistically significant predictor for the Temporal Flexibility Coefficient, confirming that non-elite players tend to decrease flexibility over time while elites increase it.

4. Key Contributions

Operationalization of Flexibility: The paper introduces a novel mathematical framework ( $F$ score and Temp Flex Coef) to quantify player adaptability in MOBA games, moving beyond simple role counts.
The Specialization-Flexibility Duality: It empirically demonstrates a trade-off: Specialization drives immediate competitive success (higher win rates), while Flexibility is a trait of elite players who evolve over time.
Temporal Dynamics: The study reveals that player adaptability is not static; it evolves differently based on skill tier. High-level players become more versatile over time, whereas lower-level players become more rigid.
Role-Specific Insights: It highlights that the "One Trick Pony" phenomenon is most prevalent in specific role categories (Mage-Support, Fighter-Tank) within elite ranks.

5. Significance and Implications

For Game Design & Esports: The findings suggest that game balance and matchmaking systems should account for the fact that high-level play rewards specialization, while lower levels may benefit from encouraging exploration.
Strategic Planning: In competitive team environments, understanding a player's flexibility score can inform resource allocation and team composition strategies.
Broader Application: The concepts of flexibility vs. specialization apply beyond gaming to any dynamic, collaborative environment (e.g., corporate teams, emergency response), offering a model for how stress and competition influence adaptability.
Future Research: The authors propose extending this work to longitudinal datasets and global regions to understand cultural and systemic differences in player adaptability.

Limitations

Data Scope: The dataset covers only a partial history (up to 100 recent games) and a specific 4-month window, potentially missing long-term behavioral shifts.
Regional Bias: Data is limited to North American servers.
API Constraints: Rate limiting and the nature of API data (aggregated, not raw telemetry) restrict the granularity of behavioral analysis.