Data Analytics

Chicago Bulls Reddit Analysis

An independent behavioral analytics project built using Python to analyze fan discussion patterns within the Chicago Bulls subreddit.
The Context Using live game and post-game discussion data collected throughout December, January, and February, this project explored how online fan communities shape player perception, emotional investment, and long-term narrative around athletes and teams.
The Work Built an independent behavioral analytics system using Python to collect, structure, and analyze discussion patterns within the Chicago Bulls subreddit.
The Approach Used Reddit scraping workflows, sentiment classification logic, and structured data pipelines to transform raw fan conversation into measurable narrative and player perception trends over time.
The Outcome Created a system capable of identifying recurring emotional patterns, narrative concentration, and long-term belief structures within digitally engaged sports communities.
The analysis revealed that fan loyalty was increasingly concentrated around a small number of players rather than team outcomes themselves.

Development narratives continued preserving emotional investment even as institutional trust declined. Instead of simply tracking positive or negative sentiment, the system focused on identifying durable belief structures and recurring narrative patterns within community conversation.

Research Focus

The project analyzed live game and post-game Reddit threads to understand how fan communities form long-term player narratives independent of wins and losses.

The Goal

The broader goal was to explore how online communities shape emotional investment, trust, and public narrative around athletes and teams over time.

The Approach

Built using Python, Reddit JSON scraping workflows, CSV pipelines, rules-based classification logic, and Streamlit dashboard development to convert unstructured fan discussion into measurable behavioral data.

System Build
Folder Structure
bulls-narrative-analysis/
├── data/
├── scripts/
├── dashboard/
├── exports/
└── config/
Workflow Logic
for thread in game_threads:
    comments = fetch_reddit_comments(thread.url)
    parsed = parse_comment_metadata(comments)
    tagged = tag_players(parsed)
    scored = classify_sentiment(tagged)
    export_to_csv(scored, game_id)
Narrative Classification
if contains_phrase(comment, growth_terms):
    theme = "developmental optimism"

elif contains_phrase(comment, trust_terms):
    theme = "brand trust"

elif contains_phrase(comment, frustration_terms):
    theme = "performance frustration"

Examples of Player Narrative Profiles

Matas Buzelis
Very High Narrative Concentration
Strong Positive Belief Signal
Functioned as the primary future-facing belief anchor. Fans projected long-term meaning onto his development regardless of immediate outcomes.
Coby White
High Narrative Volatility
Mixed / Net Negative Signal
Occupied a stress-testing role within the fan base. Conversation reflected tension between optimism, ceiling concerns, and uncertainty around long-term fit.
Nikola Vučević
Medium Narrative Presence
Structural Frustration Signal
Absorbed broader organizational frustration rather than isolated game-specific blame, signaling emotional disengagement already in progress.

What the data revealed

Live game threads were significantly more emotionally reactive than post-game discussions.

Real-time conversation patterns were dominated by short-term emotional swings, while post-game threads reflected more stable long-term narratives and memory formation.

Fan trust was increasingly player-centric rather than institution-centric.

Development narratives around specific players continued preserving engagement even as broader organizational trust declined.

Narrative absence itself became a measurable signal.

Low mention frequency combined with negative framing suggested emotional disengagement and declining narrative leverage around certain players.

Cyclistic Bike Share

Due to privacy concerns, I cannot show data or internal documents from previous employers.  Thankfully, Google’s Data Analytics Certification allowed me to produce assets that showcase my understanding of best practices. 

Data Analyst Duties Performed:

  • Clean and organize data to prepare for analysis.

  • Complete analysis and calculations using spreadsheets, SQL, and R programming.

  • Use data-driven decision-making to better understand consumer behavior.

  • Create visualizations of important data using Tabelu.

  • Address the key question by suggesting 5 potential action steps supported by the data.

  • Summarize findings in a presentation for stakeholders.

View Deck
Previous

Content Marketing