Exercises: Home Field Advantage

Exercise 1: Basic HFA Calculation

A season had 272 games with these results: - Home wins: 148 - Away wins: 120 - Ties: 4

Calculate: - Home win percentage - Away win percentage - Implied HFA in points (use 0.5% ≈ 0.35 points approximation)

Exercise 2: Point Margin HFA

Season point data: - Total home points: 6,120 - Total away points: 5,850 - Games: 272

Calculate: - Average home points per game - Average away points per game - HFA in points

Exercise 3: Era Comparison

Compare HFA across eras:

Calculate home win percentages and identify the trend.

Exercise 4: COVID Analysis

2020 season with limited fans: - Home wins: 128 - Away wins: 126 - Total games: 256

Compare to 2019 (pre-COVID): - Home wins: 148 - Away wins: 108

What does this suggest about the role of fans?

Exercise 5: Travel Distance Effect

Away team performance by travel distance:

Calculate away win percentage for each category. Is there a pattern?

Exercise 6: Timezone Analysis

Games by timezone direction:

Which direction of travel is most difficult for away teams?

Exercise 7: Team-Specific HFA

Team A's results: - Home: 7-2 (avg margin +8.5) - Away: 4-5 (avg margin -2.1)

Team B's results: - Home: 6-3 (avg margin +4.2) - Away: 5-4 (avg margin +1.5)

Calculate HFA for each team. Which has stronger home field advantage?

Exercise 8: Penalty Differential

Season penalty data: - Home team false starts: 285 - Away team false starts: 345 - Home team delays: 120 - Away team delays: 155

Calculate the ratio differences. What does this suggest about crowd noise effect?

Exercise 9: Altitude Effect

Denver's home results vs away: - Home: 8-1, avg margin +12.5 - Away: 5-4, avg margin +1.8

League average HFA: +2.5 points Calculate Denver's HFA. How much might be attributed to altitude?

Exercise 10: Spread Prediction

Build a spread prediction for KC hosting BUF: - KC power rating: +4.5 - BUF power rating: +3.0 - KC home field advantage: 3.0

What's the predicted spread?

Exercise 11: Playoff Value

If home field is worth 2.5 points per game, calculate the total value of the #1 seed if they: - Play 3 home playoff games - Win probability = 50% + (spread × 3%)

How much does home field increase championship probability?

Exercise 12: Weekly HFA Variation

HFA by game type:

Is there significant variation? Which games favor home teams most?

Exercise 13: Dome vs Outdoor

Compare HFA: - Dome stadiums: 58% home win rate - Outdoor stadiums: 53% home win rate

Calculate the difference in implied points. Why might domes have higher HFA?

Exercise 14: Weather Impact

Cold weather games (< 40°F): - Home wins: 65 - Away wins: 45

Moderate weather (40-70°F): - Home wins: 85 - Away wins: 75

Calculate HFA for each. Does cold weather amplify home advantage?

Exercise 15: Model Calibration

Your model predicts without HFA that Team A beats Team B 52% of the time on neutral field. The game is at Team A's home.

If HFA = 2.5 points and each point ≈ 3% win probability: - What's the adjusted win probability for Team A?

Programming Exercises

Exercise 16: HFA Calculator

Write a function to calculate comprehensive HFA:

def calculate_hfa_metrics(schedules: pd.DataFrame, season: int = None) -> dict:
    """
    Calculate HFA metrics.

    Returns:
    - home_win_pct
    - home_margin
    - implied_points
    - penalty_differential (if PBP available)
    """
    pass

Exercise 17: Team HFA Ranker

Build a team-specific HFA calculator:

def rank_team_hfa(schedules: pd.DataFrame, min_games: int = 16) -> pd.DataFrame:
    """
    Rank teams by home field advantage.

    Returns DataFrame with:
    - team
    - home_record
    - away_record
    - hfa_win_diff
    - hfa_margin_diff
    """
    pass

Exercise 18: Travel Effect Analyzer

Create a travel distance analysis function:

def analyze_travel_effect(schedules: pd.DataFrame, team_locations: dict) -> pd.DataFrame:
    """
    Analyze away team performance by travel distance.

    Returns:
    - distance_bucket
    - games
    - away_win_pct
    - avg_away_margin
    """
    pass

Exercise 19: HFA Trend Visualizer

Build a visualization of HFA over time:

def plot_hfa_trend(schedules: pd.DataFrame, start_year: int, end_year: int) -> None:
    """
    Plot HFA trends over multiple seasons.

    Include:
    - Home win percentage by season
    - Home point margin by season
    - Trend line
    - COVID year marker
    """
    pass

Exercise 20: Prediction Model with HFA

Build a complete prediction model:

def predict_game(
    home_team: str,
    away_team: str,
    team_ratings: dict,
    team_hfa: dict = None,
    default_hfa: float = 2.5
) -> dict:
    """
    Predict game outcome including HFA.

    Returns:
    - predicted_spread
    - home_win_prob
    - confidence
    """
    pass

Challenge Exercises

Challenge 1: HFA Decomposition

Decompose HFA into its components: 1. Calculate overall HFA 2. Estimate crowd effect (compare to COVID) 3. Estimate travel effect (by distance) 4. Estimate timezone effect 5. Calculate residual (unexplained HFA)

Challenge 2: Dynamic HFA Model

Build a model where HFA varies by: 1. Time of season (early vs late) 2. Weather conditions 3. Playoff vs regular season 4. Prime time vs day games 5. Division vs non-division opponent

Challenge 3: HFA Prediction Value

Test whether incorporating HFA improves predictions: 1. Build baseline model (no HFA) 2. Add fixed HFA (2.5 points) 3. Add team-specific HFA 4. Compare accuracy (ATS, straight up) 5. Determine optimal HFA values

Challenge 4: Venue-Specific Analysis

For a specific stadium (e.g., Seattle): 1. Calculate historical HFA 2. Analyze false start rates 3. Track overtime noise levels 4. Compare to league average 5. Estimate value in points

Challenge 5: Future HFA Projection

Project HFA for next season: 1. Analyze recent 5-year trend 2. Account for any stadium changes 3. Consider roster factors 4. Build confidence interval 5. Compare to betting market HFA

Solutions

Solutions are available in code/exercise-solutions.py