League of Legends: Top Lane Meta Analysis

Top Lane Champion Performance
and Meta Analysis

Game

League of Legends

Goal

This project provides a comprehensive analysis of top lane champion performance in League of Legends. By evaluating key performance metrics such as win rates, pick rates, and other relevant data, this analysis aims to identify trends within the current meta and provide insights for balancing champions. The analysis draws on data from 100 top-ranked matches using the Riot Games API.

Tools used

Programming Language: Python
Data Collection: League of Legends API (Riot Games API)
Data Analysis: Pandas, NumPy, Scikit-learn,
Visualization: Tableau, Matplotlib, Seaborn, Plotly
Models: Logistic Regression, RandomForest

Dashboard

This dashboard provides a comprehensive analysis of top lane champions in League of Legends, focusing on champion performance across multiple metrics including a custom performance metric that evaluates a champion based on a standardization of the features within data.

The tiers have been calculated based on a champion's performance score and are broken up as:

S-Tier

At or above the 90th percentile

(top 10% performers)

A-Tier

Between the 75th and 90th percentiles

(top 25% performers)

B-Tier

Between the 30th and 75th percentiles

(moderate performers)

C-Tier

Between the 10th and 30th percentiles

(below-average performers)

D-Tier

Below the 10th percentile (lowest performers)

Key Analysis and Takeaways

Champion Performance vs. Tier Ranking

Certain champions, like Gragas, Monkey King, and Rumble, exhibit strong win rates despite not being in the highest tier, making them valuable picks.

Champions like Dr. Mundo and Monkey King provide a balance between high win rates and top-tier performance, making them solid and consistent choices.

Conversely, champions like Heimerdinger show lower win rates, suggesting they may require more skill or favorable conditions to succeed, making them riskier picks.

Win Rate and Tier Misalignment

There is a noticeable misalignment between win rates and tier rankings for some champions. For example, Kled is an S-tier champion, but does not necessarily guarantee the highest win rate.

Champions in B- and C-tier are spread across a wide range of win rates, indicating that player skill, favorable matchups, or specific game conditions may influence their performance more than tier rank.

Popularity vs. Effectiveness

Champions like Garen, Aatrox, and Jayce are popular despite being in the B-tier, potentially because they are accessible and widely recognized, even though their win rates are only average.

Fiora and Kled, categorized in the S-tier, have lower pick rates, likely due to their higher difficulty or specialized roles.

There is a balance between game counts across tiers, but some champions with high play rates (like Garen and Riven) are in B/C-tier, implying that while they are reliable, they may not be the strongest choices based on win rate.

Meta Considerations

Win rates do not always directly correlate with tier rankings. Champions that perform well in terms of win rate might still be classified as average or below average in tier rankings due to other factors like versatility, consistency, or their impact in professional play.

Balancing Popularity and Performance

Despite champions like Garen, Aatrox, and Aurora being popular, they may not be the most effective in the current meta and players should consider lesser-picked champions like Fiora and Kled, who show strong performance despite lower popularity.

Visualizations

Queries

Find Champions Who Performed Above the Average Performance of Their Tier

WITH avg_tier_score_cte AS (
-- Calculate the average champ performance score for each tier
SELECT
tier,
ROUND(AVG(champ_performance_score::NUMERIC), 2) AS avg_tier_score
FROM
champions_tiers
GROUP BY
tier
)
-- Select each champion only once, comparing their performance score to the average for their tier
SELECT DISTINCT ON (c.champion_name)
c.tier,
c.champion_name,
c.champ_performance_score,
a.avg_tier_score
FROM
champions_tiers c
JOIN
avg_tier_score_cte a
ON c.tier = a.tier
WHERE
c.champ_performance_score > a.avg_tier_score
ORDER BY
c.champion_name, -- Ensure each champion is distinct
c.champ_performance_score DESC; -- Return the highest performance score for each champion

----------------------------------------------------------------------------------------------------------------------------

"B-Tier" "Aurora" 9.48 1.38
"B-Tier" "Cassiopeia" 1.57 1.38
"B-Tier" "Chogath" 11.71 1.38
"S-Tier" "DrMundo" 42.17 39.63
"B-Tier" "Gragas" 6.2 1.38

......

Find the average match performance score for champions who won their matches

WITH avg_match_performance AS (
-- Calculate the average match performance score for each champion who won
SELECT
champion_name,
AVG(match_performance_score) AS avg_match_performance_score
FROM
champions_tiers
WHERE
win = TRUE
GROUP BY
champion_name
)
-- Calculate the overall average of match performance scores and round to 2 decimal places
SELECT
ROUND(CAST(AVG(avg_match_performance_score) AS NUMERIC), 2) AS average_match_performance_score
FROM
avg_match_performance;

------------------------------------------------------------------------------------------------------------------------

16.17

Which champion has the highest assist-to-death ratio (assists per death) for matches they won?

WITH assist_to_death_cte AS (
-- Calculate the assist-to-death ratio for each champion
SELECT
champion_name,
ROUND(assists::NUMERIC / deaths, 2) AS assist_to_death
FROM
champions_tiers
WHERE
assists > 0
AND deaths > 0 -- Ensures no division by zero

AND win = True
)
-- Fetch the top 5 champions by assist-to-death ratio
SELECT
champion_name,
assist_to_death
FROM
assist_to_death_cte
ORDER BY
assist_to_death DESC
LIMIT 1;

------------------------------------------------------------------------------------------------------------------------

"Shen" 11.00

Find Champions Who Performed Above the Average Performance of Their Tier

-- Calculate the average damage dealt to objectives per tier and order the result by the highest average
SELECT
tier,
ROUND(AVG(damage_dealt_to_objectives), 2) AS avg_objectives_dmg
FROM
champions_tiers
GROUP BY
tier
ORDER BY
avg_objectives_dmg DESC

------------------------------------------------------------------------------------------------------------------------

"S-Tier" 13354.39
"A-Tier" 9947.37
"B-Tier" 7720.35
"C-Tier" 4799.86
"D-Tier" 3651.41

​Top Lane Champion Performance and Meta Analysis

Dashboard

S-Tier

A-Tier

B-Tier

C-Tier

D-Tier

Key Analysis and Takeaways

Champion Performance vs. Tier Ranking

Win Rate and Tier Misalignment

Popularity vs. Effectiveness

Meta Considerations

Balancing Popularity and Performance

Visualizations

Queries

Top Lane Champion Performance
and Meta Analysis