Chapter 16 Quiz

Instructions: Choose the best answer for each question. Some questions may have multiple plausible answers; select the most correct or complete option. A score of 70% or higher is required to pass.

Question 1. Which of the following is NOT one of the six primary team style dimensions described in Section 16.1?

(a) Possession orientation (b) Pressing intensity (c) Individual dribbling frequency (d) Defensive line height

Question 2. A team's style fingerprint is represented mathematically as:

(a) A scalar value between 0 and 1 (b) A vector in $\mathbb{R}^d$ where $d$ is the number of style dimensions (c) A binary classification (attacking vs. defensive) (d) A 2D coordinate on a pitch map

Question 3. When computing style fingerprints, why is it important to normalize defensive metrics by opposition possession time?

(a) To make the numbers smaller and easier to interpret (b) Because teams with more possession mechanically have fewer defensive actions, regardless of defensive quality (c) Because defensive metrics are unreliable without normalization (d) To account for referee bias in recording defensive events

Question 4. In the expected points model using the Skellam distribution, what does the Skellam distribution represent?

(a) The total number of goals in a match (b) The difference between home goals and away goals (c) The probability of a clean sheet (d) The expected goal difference over a season

Question 5. A team has 1.6 xG and their opponent has 1.6 xG. Under the independent Poisson model, the expected points for this team are closest to:

(a) 0.5 (b) 1.0 (c) 1.2 (d) 1.5

Question 6. What does a positive "Luck Index" (Actual Points minus Expected Points) indicate?

(a) The team has elite finishing ability (b) The team has scored more goals than expected from their xG, which may include both skill and randomness (c) The team has been fortunate and may regress in future matches (d) The team's coach is tactically superior

Question 7. In the Dixon-Coles model, the correlation parameter $\rho$ is typically:

(a) Large and positive (around +0.5) (b) Small and negative (around -0.03 to -0.10) (c) Exactly zero (d) Large and negative (around -0.5)

Question 8. What does the Dixon-Coles correction factor $\tau$ adjust for?

(a) Home advantage (b) The tendency for low-scoring draws to be more common than independent Poisson predicts (c) Seasonal trends in goal scoring (d) Referee bias in awarding penalties

Question 9. The Squad Depth Index (SDI) has a value of 1.0 when:

(a) All players in the squad are equally talented (b) The backup players perfectly replicate first-choice quality at that position (c) The team has exactly 25 players in the squad (d) Every position has at least 3 competent players

Question 10. A Wage Gini Coefficient above 0.55 for a squad is associated with:

(a) Better team chemistry (b) Higher league positions (c) Dressing-room tension and underperformance relative to total wage bill (d) More equitable playing time distribution

Question 11. The Lineup Stability Index (LSI) is computed by:

(a) Counting the total number of unique players used in a season (b) Averaging the fraction of shared starters between consecutive matches (c) Measuring the standard deviation of squad rotation (d) Counting the number of matches where the same XI started

Question 12. In the team chemistry framework, pairwise passing chemistry $C_{ij}$ incorporates all of the following EXCEPT:

(a) Pass accuracy between the two players (b) Pass frequency between the two players (c) The market value of the two players (d) Expected threat (xT) added by passes between the two players

Question 13. The integration curve for a new signing follows $C_{ij}(t) = C_{ij}^{\infty}(1 - e^{-\kappa t})$. A higher value of $\kappa$ means:

(a) The player reaches peak chemistry more slowly (b) The player reaches peak chemistry more quickly (c) The player's ultimate chemistry level is higher (d) The player is more likely to be sold

Question 14. Press trigger alignment measures:

(a) How often a team wins the ball back from pressing (b) The percentage of pressing sequences where the trigger player initiates and teammates join within 2 seconds (c) The speed at which a team transitions from defense to attack (d) How high up the pitch a team begins pressing

Question 15. In score-state analysis, the "tactical flexibility" metric $\Delta_{\text{flex}}$ is computed as:

(a) The difference in points earned when winning vs. losing (b) The Euclidean distance between style vectors in the winning and losing states (c) The number of formation changes made during a match (d) The variance of possession percentage across different score states

Question 16. A team with fewer than 4 days of recovery between matches typically shows approximately what decline in high-intensity running distance?

(a) 0-2% (b) 5-8% (c) 15-20% (d) 25-30%

Question 17. In the Elo rating system for soccer, home advantage is typically modeled by:

(a) Doubling the home team's Elo rating (b) Adding approximately 65 points to the home team's rating for expectation calculations (c) Subtracting 100 points from the away team's rating (d) Using a separate Elo rating for home and away performance

Question 18. The Fixture Difficulty Rating (FDR) for a team's upcoming matches is primarily based on:

(a) The team's own recent form (b) Historical head-to-head results (c) The Elo ratings of upcoming opponents, adjusted for home/away (d) The goal difference of upcoming opponents

Question 19. In Monte Carlo season simulation, why are 10,000+ iterations typically used?

(a) To ensure the simulation runs for at least one hour (b) To produce stable probability estimates with low sampling error (c) Because fewer iterations would violate the central limit theorem (d) To account for leap years and scheduling irregularities

Question 20. The Dixon-Coles model decomposes team strength into:

(a) A single overall rating per team (b) Separate attack ($\alpha$) and defense ($\beta$) parameters with a home advantage parameter (c) Offensive, defensive, and transitional ratings (d) First-half and second-half performance ratings

Question 21. In a state-space model for dynamic team strength, the Kalman filter is used to:

(a) Remove outlier matches from the dataset (b) Model team strength as a latent variable that evolves smoothly over time (c) Compute the exact number of goals a team will score (d) Filter out the effects of referee decisions

Question 22. The Brier score for evaluating prediction calibration is defined as:

(a) The average absolute difference between predicted and actual outcomes (b) The average squared difference between predicted probabilities and actual outcomes (c) The logarithm of the predicted probability of the actual outcome (d) The correlation between predicted and actual points

Question 23. Which of the following is a valid use of season simulation for decision support?

(a) Determining the exact final league table (b) Evaluating whether a potential signing would improve top-4 probability enough to justify the transfer fee (c) Predicting the exact scoreline of every remaining match (d) Guaranteeing a team will avoid relegation

Question 24. Style drift is measured as:

(a) The change in league position over consecutive matchweeks (b) The Euclidean distance between style vectors at consecutive time windows (c) The number of tactical formation changes in a season (d) The difference between expected and actual goals

Question 25. When validating a season simulation model, a well-calibrated model is one where:

(a) It always predicts the correct league winner (b) Events predicted to occur with probability $p$ actually occur with frequency approximately $p$ (c) The Brier score equals exactly zero (d) The predicted points total matches the actual points total for every team

Answer Key

(c) Individual dribbling frequency --- style dimensions focus on collective behaviors.
(b) A vector in $\mathbb{R}^d$.
(b) Teams with more possession have fewer defensive actions mechanically.
(b) The difference between home goals and away goals.
(c) With equal xG, the draw probability is high, yielding about 1.2 xPts.
(c) A positive Luck Index suggests over-performance that may regress.
(b) Small and negative.
(b) Low-scoring draws being more common than independent Poisson predicts.
(b) Backup players perfectly replicate first-choice quality.
(c) Dressing-room tension and underperformance relative to wage bill.
(b) Averaging the fraction of shared starters between consecutive matches.
(c) Market value is not a component of pairwise passing chemistry.
(b) Higher $\kappa$ means faster integration.
(b) Percentage of pressing sequences where trigger player initiates and teammates join within 2 seconds.
(b) Euclidean distance between style vectors in winning and losing states.
(b) 5-8%.
(b) Adding approximately 65 points to the home team's rating.
(c) Elo ratings of upcoming opponents, adjusted for home/away.
(b) To produce stable probability estimates with low sampling error.
(b) Separate attack and defense parameters with home advantage.
(b) Model team strength as a latent variable that evolves over time.
(b) Average squared difference between predicted probabilities and actual outcomes.
(b) Evaluating whether a signing would improve top-4 probability.
(b) Euclidean distance between style vectors at consecutive time windows.
(b) Events predicted with probability $p$ actually occur with frequency $p$.