Scribbler Logo

    You have 4 summaries left

    DataFramed

    #141 How Data Science is Transforming the NBA

    Mon Jun 12 2023
    BasketballAnalyticsPlayer EvaluationTracking DataAdvancements in TechnologyOffensive StrategiesCareer PathsData AnalysisNBA Playoffs

    Description

    The episode covers various aspects of basketball analytics, including understanding the subject matter and data, integration of data in decision-making process, player evaluation, tracking data, advancements in technology, offensive strategies, career paths in sports analytics, analyzing data and uncertainty, and uncertainty in the NBA playoffs.

    Insights

    Basketball analytics focuses on measuring scoring volume and efficiency as key indicators of player performance.

    Usage rate measures the number of play-ending efforts a player takes per scoring chance while efficiency measures how many points they generate from those attempts. A player's scoring volume and efficiency can vary depending on their role within a team and the mix of shots they receive.

    The rise of the three-point shot is driven by empirical evidence that it is more efficient than shots closer to the basket.

    The increased distance from the basket creates more space on the court for skilled players to attack the rim.

    Analyzing defense is difficult due to the inherent counterfactual nature and the complexity of measuring choices made by the offense.

    Attributing defensive achievements to individuals is challenging.

    Sports analytics roles are becoming more diverse, including positions like data engineers, analysts, front-end developers, and back-end developers.

    Identifying your competitive advantage is crucial when applying for these jobs. Communication skills are important, including written communication, verbal communication, data visualization, and subject matter knowledge of the sport being analyzed.

    Analyzing data and catching errors can save a lot of pain

    Pick your lane and learn about the thing you want to talk about. Job descriptions in sports analytics have become more accurate.

    The NBA playoffs have shown a divergence between regular season and playoff performance

    The compressed league standings have made the playoffs less predictable. Playoff games can make predictions feel wrong and uncertain. Statistical analysis in the playoffs is challenging due to smaller sample sizes and intense play.

    Learning to say I don't know is powerful.

    The Boston Celtics are superior on paper, but the Miami Heat have been better over the playoffs. It's uncertain which team will hold sway in the Eastern Conference Finals. Unpacking why something happened will be a challenge. The finals could be anyone's game.

    Chapters

    1. Understanding the Subject Matter and Data
    2. Integration of Data in Decision-Making Process
    3. Player Evaluation in Basketball
    4. Tracking Data in Basketball
    5. Advancements in Technology and the Rise of Three-Point Shot
    6. Analyzing Shot Success and Offensive Strategies
    7. Career Paths in Sports Analytics
    8. Analyzing Data and Uncertainty in Sports Analytics
    9. Uncertainty in the NBA Playoffs
    Summary
    Transcript

    Understanding the Subject Matter and Data

    00:00 - 07:44

    • Understanding the subject matter and data is important for accurate analysis.
    • Basketball analytics helps understand player and team decisions in the game.

    Integration of Data in Decision-Making Process

    00:00 - 07:44

    • Teams focus on sports side and business side, with separate data challenges.
    • Data can impact sport science, player recruitment, and game analysis.
    • Integration of data in decision-making process varies among teams.
    • Some teams use data as a resource, while others integrate it at every stage.
    • Most NBA teams do their analytics in-house, with some use of consultants for complex problems.
    • Baseball has a more integrated use of data compared to other American sports.
    • Recruiting players involves using better stats and advanced methods for evaluation.

    Player Evaluation in Basketball

    07:20 - 14:32

    • The evaluation of players in basketball has always included some level of statistical analysis, but now there are better stats and more advanced methods available.
    • Player tracking data has allowed for the application of analytics to on-floor in-game strategy, which is a newer and more directly threatening development to coaching structures.
    • The book 'Mid-Range Theory' is a compilation of topics that the author has extensively researched, thought about, and discussed in basketball analysis.
    • In team sports like basketball, soccer, American football, hockey, rugby, etc., the interaction between players greatly impacts the outcome of the game.
    • Determining attribution for player contributions in basketball is challenging due to factors like shot difficulty and the influence of teammates.
    • Basketball analytics focuses on measuring scoring volume and efficiency as key indicators of player performance.
    • Usage rate measures the number of play-ending efforts a player takes per scoring chance while efficiency measures how many points they generate from those attempts.
    • A player's scoring volume and efficiency can vary depending on their role within a team and the mix of shots they receive.
    • Russell Westbrook is an example of a player with high usage but middling efficiency.

    Tracking Data in Basketball

    14:13 - 21:21

    • There is a trade-off between taking many shots with low chance of success and waiting for high-quality opportunities.
    • Different player types have different offensive roles and shooting efficiencies.
    • Tracking data in the NBA has evolved from box score data to detailed play-by-play to player tracking data.
    • Player tracking data provides more information about shot locations, assists, blocks, fouls, substitutions, etc.
    • Pose data will be added to player tracking data next year, allowing for more detailed analysis.
    • The increase in available data allows for better measurement of shot quality and understanding of possession context.
    • The amount of raw information available has exponentially increased over time.

    Advancements in Technology and the Rise of Three-Point Shot

    21:06 - 28:28

    • Advancements in technology have allowed for the collection of a significantly larger amount of data in basketball games.
    • Analyzing play-by-play data can still be done in Excel, but tracking data requires database solutions.
    • Identifying patterns and events in tracking data requires algorithmic and machine learning techniques.
    • The importance of height in basketball has decreased relative to skill due to rule changes, style of play, and strategic optimization.
    • The rise of the three-point shot is driven by empirical evidence that it is more efficient than shots closer to the basket.
    • The increased distance from the basket creates more space on the court for skilled players to attack the rim.

    Analyzing Shot Success and Offensive Strategies

    28:06 - 35:17

    • There is a drop-off in shot success by distance, with jump shots under NBA conditions being less accurate than closer shots.
    • Analytics has shown certain types of offensive strategies to be more efficient than others.
    • Statistical analysis allows for faster exploration and discovery of new strategies compared to traditional methods.
    • Analyzing defense is difficult due to the inherent counterfactual nature and the complexity of measuring choices made by the offense.
    • Attributing defensive achievements to individuals is challenging.
    • The field of sports analytics offers various career paths, but there are limited defined pathways for entry.
    • The speaker stumbled into sports analytics after working in e-commerce and becoming a lawyer.
    • Playing poker professionally provided applied statistics training and developed statistical intuition.
    • The speaker transitioned into educational consulting while writing about basketball on the side.

    Career Paths in Sports Analytics

    34:59 - 42:03

    • The speaker started by working in educational consulting and writing about basketball on the side.
    • They had a statistical intuition and basketball experience that allowed them to do more with new tracking data than most people.
    • This led to better opportunities to write, do analysis for larger outlets, and get contacted by teams.
    • The Milwaukee Bucks eventually hired them for a role commensurate with someone in the middle of their career.
    • Practical and subject matter training have been more valuable in their career than pure statistical programming or formal statistics.
    • For those interested in sports analytics, it's important to consider whether they want to work in sports analytics or for a team on the sports side.
    • Working for a team requires long hours and sacrifices work-life balance, while working for a data provider offers more stability.
    • Sports analytics roles are becoming more diverse, including positions like data engineers, analysts, front-end developers, and back-end developers.
    • Identifying your competitive advantage is crucial when applying for these jobs.
    • Communication skills are important, including written communication, verbal communication, data visualization, and subject matter knowledge of the sport being analyzed.
    • Understanding the subject matter helps catch errors and ensures accurate analysis of the data being worked with.

    Analyzing Data and Uncertainty in Sports Analytics

    41:39 - 48:42

    • Analyzing data and catching errors can save a lot of pain
    • Pick your lane and learn about the thing you want to talk about
    • Job descriptions in sports analytics have become more accurate
    • The NBA playoffs have shown a divergence between regular season and playoff performance
    • The compressed league standings have made the playoffs less predictable
    • Playoff games can make predictions feel wrong and uncertain
    • Statistical analysis in the playoffs is challenging due to smaller sample sizes and intense play
    • Learning to say 'I don't know' is powerful in data analysis

    Uncertainty in the NBA Playoffs

    48:20 - 48:58

    • Learning to say I don't know is powerful.
    • The Boston Celtics are superior on paper, but the Miami Heat have been better over the playoffs.
    • It's uncertain which team will hold sway in the Eastern Conference Finals.
    • Unpacking why something happened will be a challenge.
    • The finals could be anyone's game.
    1