Weekly FPL Lesson Plan: Teaching Data Literacy Using Premier League Statistics

Hook: Turn students' love of Fantasy Football into a focused data literacy course

Teachers and students are overwhelmed by scattered tutorials, messy datasets, and too many abstract statistics that don’t connect to real work. This week-long lesson plan uses live FPL and Premier League statistics to teach practical data literacy: cleaning messy data, analyzing trends, and crafting compelling data stories that communicate insight — all using a familiar, motivating sports context.

Why this matters in 2026

By 2026 classrooms increasingly blend human instruction with AI tools, and sports datasets are a natural playground for learning core data skills. Recent educator reports from late 2025 and early 2026 show growing adoption of automated data-cleaning helpers and interactive visualization platforms in secondary and university courses. Meanwhile, public FPL JSON endpoints and community GitHub mirrors still provide rich, real-world data students can access without complicated licensing (always check provider rules before publishing). This lesson plan reflects those trends and emphasizes ethical data use, reproducibility, and clear storytelling.

Learning outcomes (week summary)

• Data cleaning: Identify and fix missing values, inconsistent player names, duplicate records, and merge disparate sources (FPL stats + team news).
• Interpretation: Compute meaningful metrics (points per million, form-adjusted expected points) and explain limitations.
• Data storytelling: Produce a short report and a one-slide interactive visualization that answers a specific question about player or team performance.
• Ethics & provenance: Document data sources, assumptions, and privacy considerations.

Materials and data sources

• Classroom devices with internet (Chromebooks, laptops)
• Spreadsheet software (Google Sheets or Excel) and access to Python (pandas) or R (tidyverse) for advanced groups
• Visualization tools: Google Data Studio, Tableau Public, Flourish, or Plotly
• Primary datasets: FPL JSON endpoints (bootstrap-static, fixtures, element_stats) or curated CSVs from public GitHub mirrors
• Supplemental qualitative data: BBC team news (injuries, rotation announcements) and official club updates for context

Week-at-a-glance plan

Five consecutive lessons, 50–90 minutes each. Each day builds on the last so students progress from messy raw inputs to a polished data story.

Day 1 — Context, questions, and dataset reconnaissance (50–60 mins)

1. Hook (10 mins): Show a short clip or screenshot from the FPL interface and BBC team-news headlines (e.g., injuries before a derby). Ask: what would you need to recommend a transfer?
2. Learning goal (5 mins): Explain that the week’s task is to answer a focused question using real FPL data — e.g., "Which midfielders under £7.5m are most likely to outperform their price over the next 4 gameweeks?"
3. Data walkthrough (20 mins): Demonstrate how to pull the FPL bootstrap-static JSON (or open the CSV). Identify fields: element_type (position), total_points, now_cost, minutes, chance_of_playing_next_round, form, and team.
4. Quick group activity (15 mins): Students inspect raw files to find anomalies: empty cells, inconsistent names, and duplicate rows. Record examples.

Day 2 — Cleaning and merging (60–90 mins)

Focus: transform raw files into analysis-ready tables and merge qualitative team-news items.

• Mini-lecture: common cleaning steps — typographic fixes, normalization, handling NA, date parsing, and duplicate removal.
• Guided worksheet: step-by-step cleaning tasks. For spreadsheet users: filter blanks, use VLOOKUP/XLOOKUP or INDEX/MATCH to join team names. For code users: run the sample pandas script below.
• Merge example: attach BBC team-news flags (injury/suspension) to players when available for the upcoming gameweek. Discuss why qualitative context matters.

Sample Python (pandas) snippet for class:

<code>import pandas as pd

players = pd.read_json('bootstrap-static.json')['elements']
players_df = pd.json_normalize(players)

# Normalize cost to £m
players_df['now_cost_m'] = players_df['now_cost'] / 10

# Drop duplicates and fill missing minutes with 0
players_df.drop_duplicates(subset=['id'], inplace=True)
players_df['minutes'] = players_df['minutes'].fillna(0)

# Example: merge team-news (small table with team_id and injury flag)
# team_news = pd.read_csv('team_news.csv')
# merged = players_df.merge(team_news, left_on='team', right_on='team_id', how='left')

print(players_df[['web_name','now_cost_m','total_points','minutes']].head())
</code>

Day 3 — Analysis and metrics (60–90 mins)

Focus: compute and interpret performance metrics.

• Key metrics to teach and compute: points-per-million (total_points / now_cost_m), form-adjusted expected points (use form * minutes-adjustment), minutes-per-point, and rotation-risk flags using team-news.
• Challenge: rank the best value midfielders under a price cap. Have students justify metric choices and note edge cases (e.g., injuries, double gameweeks).
• Stat literacy moment: teach interpretation vs. causation — a player with rising form may be due to easier fixtures, not skill improvement.

Day 4 — Visualization & data storytelling (60–90 mins)

Focus: turning analysis into insight. Teach visual encoding, annotation, and narrative structure.

1. Mini-lecture (15 mins): Story arc: question → evidence → takeaway. Visualization rules: choose the right chart, reduce clutter, annotate anomalies, use color for meaning (not decoration).
2. Workshop (30–45 mins): Students create a dashboard or slide: top 5 bargain players, scatter plot of points-per-million vs minutes, and a small table with injury flags. Encourage interactive tools (Plotly, Flourish) for university groups.
3. Class critique (15–30 mins): Quick peer feedback rounds focusing on clarity and whether the visualization answers the question.

Day 5 — Presentation, assessment, and reflection (50–90 mins)

• Students deliver 5-minute data stories: one key insight, supporting evidence, and recommended action (e.g., "Transfer in X for the next 4 gameweeks").
• Assessment rubric (see below) and reflective discussion on data limitations, ethics, and reproducibility.
• Extension: show how AI tools can suggest headlines or alternative visuals, but insist on human verification of results.

Assessment rubric (practical & clear)

Use a simple 20-point rubric for formative grading (adaptable for secondary or university levels):

• Data preparation (5 pts) — cleaned, merged, documented sources and transformations (2–5 pts).
• Analysis (5 pts) — correct metric selection, sound computations, and transparent assumptions.
• Visualization & communication (5 pts) — clarity, annotation, and one actionable takeaway that answers the original question.
• Ethics & reproducibility (5 pts) — source citation, privacy-awareness, and reproducible steps (scripts or clear spreadsheet formulas).

Sample assignments and exam-style prompts

• Short assignment (for secondary): "Using FPL public data, identify two low-cost forwards (< £6.5m) who have increased expected output this month. Show evidence in a single chart."
• Extended project (for university): "Build a model to predict player points for the next gameweek; compare a baseline linear model with a simple tree model. Evaluate using cross-validation and explain feature importance."
• Exam question: "Explain two data-cleaning steps you would take when combining FPL statistics with weekly team news. Why are these necessary?"

Differentiation: tailoring for grades and skill levels

Not all classes will have programming experience. Use these scaffolds:

• Beginner (secondary): Stay in spreadsheets, focus on sorting, filters, and simple formulas. Provide cleaned CSVs if needed.
• Intermediate: Introduce basic pandas notebooks with template code. Allow students to modify a few lines to compute metrics.
• Advanced (university): Require full reproducibility with notebooks, version control (GitHub Classroom), and a short write-up including limitations and possible biases.

Common classroom challenges & solutions

• Data access blocked: Host datasets on a shared drive or provide CSV snapshots. Encourage offline work if live APIs fail.
• Students over-trust AI helpers: Make a checkpoint where students must explain any AI-suggested code or summary in their own words.
• Time constraints: Offer a "minimum viable product" checklist: cleaned table + one chart + one takeaway.

Ethics, privacy, and classroom policy

Even with public sports data, teach students to:

• Record data provenance: source URLs, download date, and any transformations.
• Check licensing before republishing derived datasets. Community mirrors are useful but may carry licensing restrictions.
• Discuss exposure and potential harms: ranking underage players or using private club data is inappropriate. Emphasize respectful language when interpreting performance.

"Data literacy is less about tools and more about asking the right questions, checking assumptions, and communicating results clearly."

Teaching tips from experienced instructors (2026 practices)

• Leverage AI to scaffold routine tasks (e.g., generating descriptive summaries) but require students to critique and correct outputs.
• Use live fixtures and team-news updates (e.g., BBC or official club channels) to teach how qualitative variables change model inputs rapidly in sports contexts.
• Encourage reproducibility: host a template Jupyter/Observable notebook on GitHub so students can fork and submit via pull request.

Example student project briefs (realistic & motivating)

1. Value hunters: Identify three players who will likely return more points than their price indicates over the next 4 GWs. Deliver: 1-page rationale + interactive scatter with annotations.
2. Rotation risk report: Use injury flags and minutes to create a rotation-risk index for squad selection. Deliver: dashboard and a short memo to a hypothetical FPL manager.
3. Fixture forecasting: Combine fixture difficulty, form, and minutes to produce a team-level forecast for defensive clean sheets. Deliver: 3-slide pitch with methodology appendix.

Assessment: sample marking comments for feedback

• "Good cleaning work — you documented decisions about missing minutes; next time include a short sensitivity test to show how assumptions change your top-5 list."
• "Strong visualization — labels and annotations help. Consider adding a small table of raw values for transparency."
• "Nice linkage to team news — but your claim that player X will start needs an explicit rule (e.g., minutes > 60 in two prior games)."

Extensions & future-proofing (skills for 2026+)

For students ready to go further:

• Introduce simple predictive models (time-series or tree-based) and teach honest evaluation (train/test splits that respect time order).
• Bring in natural language processing: extract sentiment or injury severity from press conferences and merge with performance data.
• Teach deployment basics: publish dashboards with privacy-aware data, or build an API wrapper for classroom use.

Final checklist for teachers

• Prepare datasets and test links at least 48 hours before class.
• Create scaffolded templates (sheets and notebook starters) to lower the technical barrier.
• Decide assessment weight and whether teamwork or individual work is required.
• Plan for a 10–15 minute showcase where students present to peers — public audiences increase rigor and motivation.

Actionable takeaways

• Use sports data to teach real data literacy skills: it’s engaging, plentiful, and naturally lends itself to time-series and event-based reasoning.
• Prioritize reproducibility: require a README that lists data sources and transformation steps.
• Blend quantitative and qualitative sources: team news and injury updates change statistical realities and make the classroom debate richer.
• Keep assessment clear and scaffolded: rubric-driven feedback accelerates learning.

Resources & starter links

• FPL public JSON endpoints (commonly used bootstrap-static and fixtures JSON) — useful for live class demos.
• Public GitHub repositories that snapshot FPL data for reproducible teaching (search for community-maintained mirrors).
• BBC Sport team-news pages for qualitative context — great for merging injury/suspension flags.
• Teaching templates: provide a downloadable Google Sheets and a Jupyter notebook starter (host on your LMS).

Closing: why this lesson plan works

This FPL-based lesson plan leverages students’ intrinsic interest in the Premier League to teach rigorous, transferable data skills: cleaning, interpreting, and telling stories with data. It aligns with 2026 classroom realities — AI-assisted tooling, emphasis on reproducibility, and the need to reason ethically about datasets — while remaining practical and adaptable for secondary and university settings.

Call to action

Try the week in your classroom: download the starter datasets and a ready-to-run notebook from our resource kit, adapt the rubric to your syllabus, and invite students to publish their one-slide data stories publicly (with permission). Share a standout student project with us — we’ll feature it as a case study and provide feedback for classroom improvement.

Related Reading

• Create a City Micro-Series for YouTube: Formats Inspired by BBC Talks and Successful Pilots
• Designing a Small-Space Home Gym: Where to Store Dumbbells, Bikes and Headphones
• Creating Supportive Video Series About Mental Health That Earn Ad Revenue
• Platform Liability 101: Could X Be Sued for AI-Generated Nudes?
• Host a Pre-Season Mix Night: Combining New Releases, Film Trailers and Live Commentary