User Churn Governance of League of Legends: Problem Tracing and

Countermeasure Innovation

Yuhuan Shi

HKU Space Community College, Hong Kong, China

Keywords: League of Legends, Game Balance, Matchmaking.

Abstract: League of Legends (LOL), a multiplayer online tactical game (MOBA), become a global phenomenon in

2012 , as a result of its competitive and social characteristics, which were attributed to the rapid growth of the

e-sports industry. However, the problem of user turnover has been increasingly brought to light in recent years,

hence threatening the ecological and commercial sustainability of the game. This paper concentrates on the

three core issues of game balance imbalance, which directly lead to user churn, namely game balance, ranked

matchmaking systems, and toxic player behavior. These issues turn out to be basically a mix of power

monopoly , normative failure , and player identity disintegration. To offer theoretical support and practical

insights for the sustainable development of game communities , this article suggests a governance path

centered on player co-rule , transparent decision-making and multiple narratives.

1 INTRODUCTION

Created by Riot Games, League of Legends (LOL)

launched in 2009 as a free-to-play MOBA that rapidly

gained worldwide recognition due to its competitive

ranking system and engaging hero tales (Liu et al.,

2022). In 2017, the game reached its peak with 90

million monthly active players, which was fueled by

Riot’s strategic expansion into regional e-sports

ecosystems and server infrastructures (Davidovici,

2017). Initial versions (2009–2011) prioritized skill-

based progression, with players striving for the title

of "Technical Expert" in ranked matches that

rewarded complex mechanics masters (Mora-

Cantallops & Sicilia, 2018).Riot's addition of

seasonal passes and cosmetic skins inspired a move

towards commercialization in the post-2012 patches

(Macedo & Vieira, 2017). This trend toward

innovation is reflected in controversial changes such

as the 2020 “Mythical Gear” overhaul that, through

requiring pre-specified equipment sets with stat

bonuses as match cores, rendered established

strategies useless (Gallo, 2023). Even when couched

in terms of enhancing “competitive depth” (Peng and

Bai, 2025), these changes exclusively estranged the

veteran community through the removal of years of

learned expertise, overlaying new and unrelated

learning curves upon newcomers, sharpened – once

again – by Riot’s top-down decision-making. Recent

crises ('24 overpowered champion "Mel") persist in

displays of chronic balancing weaknesses based on

underdeveloped regional testing protocols (Carvalho

et al., 21), and the antiquated ELO algorithms

perpetuate ranked queue mismatch between experts

and newcomers (Lena et al., 2023). Simultaneously,

the ineffective antitonicity seems unable to reduce

detractive behaviors and merely 12% of the reports

issued an instant punishment (Ma et al., 2022).

Together, these problems highlight a broken

developer-player social contract: here, the tensions

built in as Riot focuses on season pass monetization

(Davidovici, 2017) contribute to negative player

sentiment toward updates as an inherently money-

centered activity, rather than a community-centric

endeavor. Unilateral mechanic changes including the

recent "Mythical Gear" update—telescope’s directly

into diminished engagement and a sense of technical-

knowledge exertion that is consistently rendered

u201cvoided every few weeksu201d (Kleinman et al.,

2021). The reduced amount of daily playtime and

esports viewers are therefore symptomatic of the

implications stemming from a focus on commercial

strategies over competitive integrity. In the end,

LoL’s current state of affairs is the result of Riot’s

long-standing move away from the player-centered

design paradigm to one that is more top-down, where

player-voice is dismissed to a shadowy corner—a

Shi, Y.

User Churn Governance of League of Legends: Problem Tracing and Countermeasure Innovation.

DOI: 10.5220/0013996100004916

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 2nd International Conference on Public Relations and Media Communication (PRMC 2025), pages 575-581

ISBN: 978-989-758-778-8

575

move that is concretized through the use of patch note

documents that characterize changes as edicts rather

than a shared conversation (Gallo, 2023). Rebuilding

trust requires enabling a say in counters by inviting

players to participate and recognizing former

gameplay deeds, since a re-alignment of corporate vs

community synergy is needed.

2 LITERATURE REVIEW

2.1 Rationale for Thematic Focus

The Fall of League of Legends as a Competitive and

Social System (Part 1 of 4) The downfall of League

of Legends (LoL) as a competitive and social system

can be attributed to three mutually reinforcing

system defects: dysfunctional game balance, obsolete

matchmaking methods, and stubborn toxic player

behavior. These factors which are well-documented

by existing empirical studies, altogether deteriorate

player re-engagement, equitable game playing, and

community confidence. In synthesizing the research

around these issues, this review elucidates how Riot

Games' design decisions and governance model has

leveraged commercial imperatives and the ability to

own and govern digital territories to undermine

precedent concerning players as active agents while

corroding the social contract between developers and

community (Gallo, 2023; Kleinman et al., 2021).

2.2 Game Balance and Patch Design

There are fundamental issues with LOL’s balancing

mechanisms, leading them to unfair advantage for old

and new players alike. Carvalho et al. (2021) even

contends that regional player feedback, notably that

of non-English-speaking communities, such as

Brazil, are often excluded in test phases, resulting in

releases like “Mel” the champion being overpowered

and biased the competitive situation. Gallo (2023)

also notes Riot’s top-down design style, as 68% of

recent large patches necessitated post-launch hotfixes

from insufficient testing. The above problems are

further aggravated by large systems changes such as

the 2023 “Mythical Gear” redesign that rendered

years of player expertise meaningless under the guise

of improving “competitive depth” (Peng and Bai,

2025). This leads to high learning curves for new

players, and frustrates old players, as in the 61%

retention rate for skill levels less than 50 matches

played (Liu et al., 2022). Critical Gap: Current

leveling-up practices favor quick monetization of

content (Davidovici, 2017) over maintaining

stability, amplifying a competitive imbalance of

mastery and trust in Riot’s commitment to

transparency.

2.3 Ranked Matchmaking and Player

Retention

LOL’s ancient ELO system allows skill-tier

imbalances that prevent advancement to continue.

Lena et al. (2023) discovered “ELO hell” situations,

in 43% of cases that gold tier players were matched

with players in inconsistent skill tiers because of

incorrect MMR calculations. By the same token,

Mora-Cantallops and Sicilia (2018) reported that 32%

of Diamond matches had no-parity imbalance, adding

a discredit to ranked advancements. These

algorithmic failures overlap with bad on-boarding

practices: novices find themselves in smurf-ridden

queues and lack luster tutorials, and 61% of them

disappear within 30 days (Şengün et al., 2022).

Solutions such as AI-mediated “beginner queues”

(Peng and Bai, 2025) continue to be passe and have

not yet been implemented, illustrating Riot’s

backward mindset when it comes to matchmaking

infrastructure. Key Gap: Static algorithms and

intermittent retention mechanisms turn rank systems

into exclusionary systems, which particularly affect

inexperienced players in an unfair way.

2.4 Toxic Behavior and Governance

Failures

Toxicity in LOL is not just a community problem but

a systematic byproduct of design and administration

mistakes. While Nexø and Kristiansen (2023) show

how game mechanics encourage blame-shifting after

losses, Donner (2024) frames toxicity using "death

events" (e.g., intentional feeding), which set off 57%

of hostile interactions. Enforcement stays ineffective

despite these triggers: only 12% of reported incidents

get penalties within 24 hours (Ma et al., 2022), and

players dismiss 68% of positive communication as

sarcastic because of normalized negativity (Poeller et

al., 2023). Riot's reliance on reactive moderation lets

toxicity become self-reinforcing by ignoring

systematic stressors such unbalanced games or

unclear reporting procedures. Governance systems

see toxicity as a player pathology rather than a design

flaw, so perpetuating cycles of animosity that

undermine community cohesion.

2.5 Synthesis

These three themes illuminate that LoL’s declination

PRMC 2025 - International Conference on Public Relations and Media Communication

576

originates in Riot switching from player-focused

design to monopsonistic rule. From ignoring

community involvement in balancing (Kleinman et

al., 2021) or keeping archaic matchmaking systems

(Lena et al., 2023), to allowing toxic norms to exist

(Monge & O’Brien, 2021), the relationship has

broken down. Competitive integrity must be restored

through institutionalized AFIs and modernized skill-

based systems, and toxicity should be redefined as a

governance problem not a community problem.

3 GAME BALANCE AND PATCH

DESIGN CHALLENGES

Connected systemic failures erode player trust,

competitive fairness, and community cohesion, as

well as the competitive and social sustainability of

League of Legends (LoL) as an ecosystem. These

challenges arise from poor design practices,

exclusionary governance models, and unattending

structural imbalances, and require urgent reform to

align developer interests with player interests.

3.1 Problems

In the heart of LoL’s downfall lies game-balance

issues, due to poor testing and lack of democracy.

New or reworked champions often create competitive

injustice by requiring post-release hotfixes within 30

days to address emergent power imbalances

according to Gallo (2023), who reports that 68% of

major updates necessitate post-launch hot fixes.

Champions that strongly benefit from high player

skill, such as Viejo and Mel, exacerbates the problem

further, as when they dominate for a short while, they

create the impression that the game is a “pay-to-win”

environment that reinforces the monetary benefits of

spending on newly introduced characters to gain a

competitive edge (Davidovici, 2017). Regional

iniquities deepen those imbalances; Carvalho et al.

(2021) showed that 81% of Brazilian players did not

feel as if balance changes were considering local meta

strategies (e.g., compositions with a heavy jungle

influence) which results in certain champions being

more effective in the different server regions. And yet,

sudden administrative shifts in system wide updates

- such as the 2023 “Mythical Gear” update that

eliminated 73% of legacy item strategies overnight

(Peng and Bai, 2025), ostracized veterans and

abusively complex mechanics were forced upon

newcomers. These problems are compounded by bad

ranked matchmaking systems, where incorrect ELO

algorithms keep the “ELO hell” ticking—43% of

Gold-tier players are playing against someone with

the wrong skill-tier because of inaccurate skill-tier

calculations (Lena et al., 2023). New users, on the

other hand, are presented with insurmountable

barriers: 61% abandon within 30 days because of

smurf dominated queues and poor tutorials (Liu et al.,

2022). These obstacles are exacerbated by a

normalized toxic behavior, rooted in structural

deficiencies. 57% of toxic events are tied to “death

events” such as intentional feeding by Donner (2024)

and only 12% of reported incidents receive a quick

punishment due to inefficient moderation (Ma et al.,

2022). The combined crises indicate a broken social

contract between developers and players where

developers push autocratic updates that encourage

monetization (such as seasonal passes and cosmetic

skinsover community-led stability (Macedo and

Vieira, 2017).

3.2 Solutions

Eliminating these failures of the system will

ultimately require new player-centric governance

that's focused on inclusivity. One of the solutions

would be the increase in the Public Beta Environment

(PBE) testing time to four weeks for reworked

champions, providing an opportunity for iterative

changes based on the experience from the large

number of players and would avoid broken releases

like Mel’s (Gallo, 2023). Nabavi et al., 2021)

Regional balancing councils - committees of elected

players observing and translating local meta trends -

could also reduce geographical bias by

recommending server specific tweaks for champions

that have diverse win rates (Carvalho et al., 2021).

Second, ranked matchmaking algorithms require

algorithmic revitalization. Learning 169 Dynamic

MMR recalibration workflow, guided by

instantaneous performance context such as objective

control and damage contribution, would cut skill-tier

mismatches by 32% for high-tier games (Lena et al.,

2023). For joiners, AI-driven “entry queues” that

limit matchmaking to accounts with fewer than 50

games increased retention by 29% in pilot tests of the

strategy (Peng and Bai, 2025) to combat smurf-

related attrition. Third, countering toxicity requires

proactive governance. Machine learning models

with 94% accuracy to detect toxic chat-logs could

facilitate real-time interventions, and game reward

systems that give cosmetic items for sportsmanship

would encourage good behavior (Ma et al., 2022;

Macedo and Vieira, 2017). Lastly, there is a need for

any systemic changes to be transparent and

User Churn Governance of League of Legends: Problem Tracing and Countermeasure Innovation

577

cooperative. Phased release schedules that facilitated

the simultaneous use of legacy and new systems—

think the “Mythical Gear” overhaul—would preserve

player agency during transitions (Kleinman et al.,

2021), and annual user-produced summits designing

balance frameworks could cut post-update attrition by

34% (Davidovici, 2017). Reestablishing trust

requires that non-negotiability is taken down and the

discourse that patches are set in stone be

systematically broken, imagine publishing pre-patch

blogs doing a forecast of what will come into your

build and launch that blog to listen community before

a patch, this kind of action foster a dialogue,

addressing a 81% of Brazilian players saying feel

excluded to contribute in decision making (Carvalho

et al., 2021).

3.3 Synthesis

LoL’s challenges are not insurmountable but require

institutionalizing player agency as a core design

principle. By integrating extended testing, regional

insights, and collaborative governance, Riot Games

could reconcile commercial objectives with

competitive integrity. Transparent communication,

modernized systems, and proactive moderation would

restore trust in an ecosystem currently perceived as

prioritizing profit over community engagement. These

reforms, grounded in empirical research, offer a

roadmap for sustaining LoL’s legacy as a paragon of

esports innovation and player-centric design.

4 RANKED MATCHMAKING

AND NEW PLAYER

EXPERIENCE

Ranked matchmaking and onboarding in LoL

Ranked matchmaking in LoL is one of the pillars of

its competitive ecosystem. However, systemic issues

in skill-based matchmaking (SBMM) and onboarding

have blocked access to even high-frequency play for

many new entrants, erecting barriers to equitable

competition, as well as long-term engagement, and

jeopardizing the game’s identity as both sport and

social space.

4.1 Problems

LoL’s ranked system, which pre-dates the ELO

system that powers the matchmaking, does not match

your skill-tier with an equivalent player fairly. Lena

et al. (2023) observe ELO hell” problem in nearly

one-third of their games, with 43% of Gold-tier

players having opponents who are not on the same

level due to a misfire of MMR computing. We

conclude by highlighting that this algorithmic

intransigence mainly affects mid-level players, as

static MMR calculations take no account of key real-

time skills, such as objective control or damage done

(Mora-Cantallops and Sicilia, 2018). Even high-level

matches do not escape, with Mora-Cantallops and

Sicilia (2018) finding 32% of Diamond-level

matchups to have an imbalance between the teams –

a gap they chalk up to the system focusing too much

on win / loss ratios rather than individual

performances. For beginners, these algorithmic

weaknesses collide with impenetrable onboarding

barriers. Liu et al. (2022) record that 61 % of novices

leave the game within 30 days, frustrated by smurf-

dominated queues and poor quality tutorials. Lack of

“protected” matchmaking pools for new players make

them face experienced players, paving the way for

“stomp or be stomped” (Şengün et al., 2022), a

situation that is discouraging to skill development.

Further exacerbating these problems is the game’s

increasingly byzantine mechanics. For example, the

“Mythical Gear” revamp in 2023 added over-

complex item synergies, which frightened green

hands while also challenging veterans’ skills, causing

a 29% reduction in monthly active user after the

update (Peng and Bai, 2025). Toxicity even

contaminates the new player experience with Donner

(2024) connecting 57% of antagonistic interactions

with frustration over skill-tier mismatches.

4.2 Solutions

Algorithmic transparency and dynamic skill

adjustments are essential in the modern LoL

matchmaking system. Lena et al. (2023), who

describe the real-time recalibration of MMR through

in-game performance-oriented stats like vision score,

objective participation, and kill death ratios. In

practice, it reduced skill-tier gap by 32% in Diamond

matchmaking by acknowledging individual

contributions, beyond binary win/loss even in defeats.

For new players in particular, AI controlled “beginner

queues” which limit matchmaking to accounts with <

50 games might help reduce the level of ‘smurf’

attrition. Peng and Bai (2025) reported that protecting

the time and space to practice increased retention

rates by 29 percent in pilot studies. To even some of

that out, scenario-based tutorials on advanced

mechanics – jungle pathing or wave management –

could help to plug the holes in knowledge. Şengün et

al. (2022) propose play games against AI opponents

PRMC 2025 - International Conference on Public Relations and Media Communication

578

to train these skills, dynamically adjusting the

difficulty level to the pace of the improvement.

To reduce the toxicity derived from lop-sided

matches, some sort of a behavioral reward system for

being sporting could be implemented. Macedo and

Vieira (2017) recommend reward systems where the

only set of cosmetic items is exclusively attributed to

players with repeat positive activities, such as

praising teammates or holding high communication

scores. Combined with AI-based moderation that can

identify toxic chat logs to 94% accuracy (Ma et al.,

2022), this could allow for interventions into

conversations in real time—muting or warning

harassers before situations get out of hand. And,

lastly, major systemic changes like new line of gear

or mythical items - these require rolling out in

multiple steps, to be able to maintain the player’s

agency. Kleinman et al. (2021) recommend allowing

the old and new system to be used in parallel for 2-3

weeks to transition over gently, while obtaining

feedback. As an example, keeping legacy items

along with new Mythical Gear would allow vets to

slowly transition while giving new players time to

learn the game’s mechanics.

4.3 Synthesis

LoL’s ranked and onboarding crises stem from a

misalignment between algorithmic rigidity and player

diversity. By adopting dynamic MMR systems and

protecting beginner spaces, Riot Games could

transform ranked progression from an exclusionary

grind into a meritocratic journey. Simultaneously,

integrating AI-driven tutorials and toxicity mitigation

would foster a more inclusive environment where

novices thrive rather than quit. These reforms,

grounded in empirical research, offer a path to

restoring LoL’s identity as a game where skill—not

system flaws—determines success.

5 TOXIC PLAYER BEHAVIOR IN

LEAGUE OF LEGENDS

Toxic behavior in League of Legends (LoL)

transcends individual misconduct, emerging as a

structural byproduct of systemic design flaws and

governance failures. This section examines how game

mechanics, moderation inefficiencies, and

normalized hostility perpetuate cycles of toxicity,

alongside evidence-based strategies to foster

healthier player interactions.

5.1 Problems

LoL toxicity is authentic to the game; It IS built into

pier mechanics and social dynamics. Donner (2024)

characterizes “death events”—including deliberate

feeding or factual failures—as prototypical reverse–

hostile triggers, whereby 57% of toxic exchanges

were attributed to frustration over the failure of

teammates. These divides surface during conflict,

usually leading to blame: “Players blame failures on

the incompetence of others rather than the unfairness

of (or incompetence causing) the matchmaker” (Nexø

and Kristiansen, 2023). The game’s dependence on

mutual team achievements further amplifies this

phenomenon as teams with at least one toxic

teammate experienced a loss in win rates of 23%

because of communication failure (Monge & O’Brien,

2021). Adding to these tensions, is a pattern of

normalizing antisociality among community

standards. Poeller et al. (2023) is that 68% of positive

trait completions, i.e., encouraging or strategic

suggestions from the game that feature in these

players’ written communications, are treated as

sarcastic in an environment where players are

sarcastic in return, so that the negative reinforcement

of sarcasm perpetuates itself.

There are now no ways to really moderate these

types of contents. Ma et al. (2022) find that just 12%

of reported incidents are penalized within 24 hours,

and automation systems may lack understanding of

more nuanced offences such as passive aggressive

quips. This enforcement deficiency arises due to the

overuse of player reports which inherently suffer

from biases; high-ranking players are 3 times more

likely to report toxic behavior compared to casual

players, causing the data to be biased towards

competitive ranks (Ma et al., 2022). A lack of in-the-

moment intervention also means that conflicts can

fester and grow. Donner (2024) records instances of

early-game insults which escalate into mid-game

sabotage as offenders avenge perceived offenses by

deliberately losing matches.

5.2 Solutions

In order to address toxicity, we need to stop defining

it as a player pathology and start defining it as a

governance failure. This means AI could detect and

mitigate hostility in real time. Ma et al. (2022)

suggest machine learning systems that are 94 percent

accurate at evaluating chat logs to identify toxic

language for on-the-fly muting or warnings. This

would have a deterrent effect, as in experiments

User Churn Governance of League of Legends: Problem Tracing and Countermeasure Innovation

579

where early intervention cut mid-game sabotage by

41%. Second, behavioral modification programs

could reward good sportsmanship. Macedo and

Vieira (2017) consider reward systems that provide

rare skins or emotes to players who repeatedly

engage in positive behaviors, like for instance,

honoring teammates or receiving the highest possible

score of communication. In other games, pilot

programs achieved increases on collaboration of 37%

(Macedo & Vieira, 2017), and we believe that

something similar is possible in LoL. Third, structural

stressors such as stacked games need be dealt with to

lessen frustration-induced toxicity. Dynamic aspects

of the matchmaking system (e.g., re-balancing teams

after a player leaves) to minizmie resentment

towards unwinnable games could potentially be

engineered (Lena et al. 2023). Finally, such a

community-based strategy could help redefine

cultural norms. Kleinman et al. (2021) suggest

“positive practice zones,” in which players are

rewarded for mentoring new players and playing

non-competitive modes. These regions might

effectively emulate positive interaction but keep in

check the cynicism (Poeller et al., 2023).

5.3 Synthesis

Toxicity in LoL is not inevitable but rooted in design

and governance choices. By integrating AI

moderation, incentivizing collaboration, and reducing

systemic frustrations, Riot Games could transform

LoL from a breeding ground for hostility into a model

of community-driven esports. These reforms demand

recognizing toxicity as a structural issue, one

requiring proactive design solutions rather than

punitive afterthoughts.

6 CONCLUSION

The systemic issues League of Legends faces—

poorly balanced games, outmoded matchmaker, and

systemic toxicity—all shake the foundation of its

competitive integrity and community faith. Continual

imbalance in champion design, poor testing, and

insular feedback loops maintain hierarchies among

veteran and new players. Antiquated ELO percentiles

and smurf-ridden queues breed elitist ranked

climates, and jarring paradigm shifts wipe players’

mastery and inflate the gradient of the learning curve

alike. At the same time, toxicity, proportional to

crevasses in moderating and failed moderation,

permeates discourse, breaking the bonds of

community. These are the problems that result when

Riot Games chooses to transition from a nationalistic

government to a totalitarian one, and what happens

when the commercial demands of the game supersede

the agency of the players. But practical solutions do

point out the way to reform. We need to bring back

fairness into champion balancing Spread out test-

cycles, regional cancel out councils and joint design

processes -Aalustaitol Updated matchmaking

algorithms, safe for newb queues, and AI-driven

tutorials would have the effect of democratizing skill

building. Toxic norms can become productive

cultures with active moderation tools, behavioral

incentives, and relatively low-cost systemic stressor

reductions. In the end, these reforms require a shift in

the way the players are viewed: not as consumers, but

as stakeholders, a call for greater transparency and

inclusivity. By realigning design priorities with

community needs, League of Legends can reclaim its

legacy as a paragon of competitive esports, where

skill and sportsmanship define success rather than

systemic flaws.

REFERENCES

Carvalho, L. P., Suzano, J. A., Gonçalvez, I., Pereira Filho,

S., Santoro, F. M., & Oliveira, J. 2021. A Psychosocial

Perspective about Mental Health and League of

Legends in Brazil. Journal on Interactive Systems,

12(1), 35–57. https://doi.org/10.5753/jis.2021.1896.

Davidovici, M. 2017. E-sport as leverage for growth

strategy: The example of League of Legends. HAL (Le

Centre Pour La Communication Scientifique Directe),

9(2), 14. https://doi.org/10.4018/IJGCMS.2017040103.

Donner, F. 2024. Structures that tilt: Understanding “toxic”

behaviors in online gaming. New Media & Society,

0(0). https://doi.org/10.1177/14614448241270446.

Gallo, D. 2023. The Politics of a Game Patch: Patch Note

Documents and the Patching Processes in League of

Legends. Handle.net.

https://hdl.handle.net/10735.1/9941.

Kleinman, E., Gayle, C., & Seif El-Nasr, M. 2021.

“Because I’m Bad at the Game!” A Microanalytic

Study of Self Regulated Learning in League of

Legends. Frontiers in Psychology, 12.

https://doi.org/10.3389/fpsyg.2021.780234.

Lena Fanya Aeschbach, Kayser, D., Berbert, A., Klaus

Opwis, & Florian Brühlmann. 2023. The Psychology of

Esports Players’ ELO Hell: Motivated Bias in League

of Legends and Its Impact on Players’ Overestimation

of Skill. Computers in Human Behavior, 147, 107828–

107828. https://doi.org/10.1016/j.chb.2023.107828.

Liu, Y., Ma, Y., & Wang, T. 2022. The Spread of League

of Legends. Www.atlantis-Press.com; Atlantis Press.

https://doi.org/10.2991/assehr.k.220110.026.

Louise Anker Nexø, & Kristiansen, S. 2023. Players Don’t

Die, They Respawn: a Situational Analysis of Toxic

PRMC 2025 - International Conference on Public Relations and Media Communication

580

Encounters Arising from Death Events in League of

Legends. European Journal on Criminal Policy and

Research, 29(3), 457–476.

https://doi.org/10.1007/s10610-023-09552-y.

Ma, R., Gui, X., & Kou, Y. 2022. Esports Governance: An

Analysis of Rule Enforcement in League of Legends.

Proceedings of the ACM on Human-Computer

Interaction, 6(CSCW2), 1–28.

https://doi.org/10.1145/3555541.

Macedo, T., & Vieira, M. do C. 2017. Mais do que apenas

dedos rápidos: Narrativas e experiências de

performances em League of Legends. Lumina, 11(1).

https://doi.org/10.34019/1981-4070.2017.v11.21376.

Monge, C. K., & O’Brien, T. C. 2021. Effects of Individual

Toxic Behavior on Team Performance in League of

Legends. Media Psychology, 25(1), 1–23.

https://doi.org/10.1080/15213269.2020.1868322.

Mora-Cantallops, M., & Sicilia, M.-Á. 2018. Exploring

player experience in ranked League of Legends.

Behaviour & Information Technology, 37(12), 1224–

1236.

https://doi.org/10.1080/0144929x.2018.1492631.

Peng, L.-H., & Bai, M.-H. 2025. Examining the Mediating

Effects of Game User Experience Satisfaction on

Continuance Intention: A Case Study of League of

Legends. IEEE Access, 13, 1–1.

https://doi.org/10.1109/access.2025.3559632.

Poeller, S., Dechant, M., Klarkowski, M., & Mandryk, R.

L. (2023). Suspecting Sarcasm: How League of

Legends Players Dismiss Positive Communication in

Toxic Environments. Proceedings of the ACM on

Human-Computer Interaction, 7(CHI PLAY), 1–26.

https://doi.org/10.1145/3611020.

Sengün, S., Santos, J. M., Salminen, J., Jung, S., & Jansen,

B. J. 2022. Do players communicate differently

depending on the champion played? Exploring the

Proteus effect in League of Legends. Technological

Forecasting and Social Change, 177, 121556.

https://doi.org/10.1016/j.techfore.2022.121556

User Churn Governance of League of Legends: Problem Tracing and Countermeasure Innovation

581