HOME arrow ANALYTICS 101
NBA Moneyball

Search This Category

Begins with Contains Exactly matches
View Glossary

All | | A | B | C | F | G | H | J | L | M | N | P | R | S | T


All
There are 27 entries in the glossary.
Pages: 1
Term Definition
Big Data & Advanced Analytics Summit

IE. Big Data & Analytics is a global community and think-tank for senior level Big Data executives, data engineers, architects, data scientists & advanced analytics executives.

IE. Summits held on fields that include Sports Analytics, Predictive Analytics, Business Intelligence and Web&Social Media Analytics.

 
Advanced Scout

Advanced Scout was developed by IBM during the mid 1990's as a data mining and knowledge management tool.

Advanced Scout reveals hidden patterns in NBA play-by-play data and provide additional insights to coaches and other related organizations.

Advanced Scout not only collects in-game structured stats, but also unstructured multimedia footage. NBA teams have access to Advanced Scout, so that coaches and players can use this tool to prepare for upcoming opponents and study them using historical footage.

 
APBRmetrics

APBRmetrics takes its name from the acronym APBR, which stands for the Association for Professional Basketball Research.

 

 
Basketball On Paper

A book which is written by Dean Oliver. In his book , Oliver highlights general strategies for teams when they're winning or losing and what aspects should be the focus in either situation. He describes and quantifies the jobs of team leaders and role players, then discusses the interactions between players and how to achieve the best fit.

Oliver conceptualizes the meaning of teamwork and how to quantify the value of different types of players working together. He examines historically successful NBA teams and identifies what made them so successful: individual talent, a system of putting players together, or good coaching.

 
Bayesian Network

A Bayesian network is a probabilistic graphical model that represents a set of variables and their probabilistic independencies.

Bayesian Networks are used in predicting the outcome of sports events

 
Bill James Revolution

George William “Bill” James is a baseball writer, historian, and statistician whose work has been widely influential. Since 1977, James has written more than two dozen books devoted to baseball history and statistics. His approach, which he termed sabermetrics in reference to the Society for American Baseball Research (SABR), scientifically analyzes and studies baseball, often through the use of statistical data, in an attempt to determine why teams win and lose. In 2006, Time named him in the Time 100 as one of the most influential people in the world.

In an essay published in the 1984 Abstract, James vented his frustration about Major League Baseball's refusal to publish play-by-play data of every game. James proposed the creation of Project Scoresheet, a network of fans that would work together to collect and distribute this information.

 

 
Box Score

Download historical NBA Box Score Stats in Excel

A box score is a structured summary of the results from a sport competition. The box score lists the game score as well as individual and team achievements in the game.

1986-87 season is the earliest season available with complete box score stats.


Comments: In basketball, we need to account for intangible things such as setting the pick, making the right pass, positioning in the corner to spread the floor, clearing space by moving down the lane and etc. These kind of data are not available in the box scores, that's why it's often been misleading. Metrics derived from play-by-play (PBP) data, not 100% perfect though, might help more than box score stats.
 
Competitive Balance Ratio (CBR)

A metric that reflects team-specific variation in winning percentage over time and league-specific variation. Based on estimation of a model of the determination of annual attendance in professional baseball during the past 100 years, variation in the CBR explains more of the observed variation in attendance than other alternatives measures of competitive balance, suggesting that CBR is a useful metric.

 
Front Office Staff of NBA Teams

Looking for a dream job in the NBA or getting curious about the organizational structure of NBA franchises? The list below compiles links to front office staff of NBA teams.



Comments: Contribute to this list by sending your tips / corrections to

 
Game Flow Chart

A unique way of illustrating a revealing summary of a basketball game with a chart showing the point innovative plots which has been presented by Peter H. Westfall in 1990.


Check out some NBA research tools with links to sources which offer game flow charts.


Comments: In basketball the outcome is difficult to summarize in a simple boxscore. Game flow charts is a very helpful tool if you didn't watch the game.
 
Herfindahl Index

The Herfindahl index, also known as Herfindahl-Hirschman Index or HHI, is a measure of the size of firms in relationship to the industry and an indicator of the amount of competition among them.

The Herfindahl Index can be used to measure scoring balances for NBA teams.

To calculate HHI for each team: (1) divide each player's points scored by the total number of points scored by the team, (2) square that result for each player, (3) sum those squares.

 
Home Court Advantage

The home-court advantage is the net effect of several factors that may have an (generally positive) effect on the play of the home team and an (generally negative) effect on the play of the road team.

Possible Sources of Home Court Advantage in Basketball:

  • Psychological support of the fans.

  • Comfort of being at home, rather than traveling.

  • Referees give home teams the benefit of the doubt?

  • Teams are familiar with particulars/eccentricities of their home court.

  • Different distributions of rest between home and road teams

Calculating The Home Court Advantage:
Subtracting one team’s power rating from another can help predicting point diffential between the two teams. Unless the game is being played at a neutral site the "home court advantage" factor needs to be incorporated for more accurately predicting the point differential of a game.

Knowing the distribution of the teams' home and road performances provides predictive information on estimating home court advantage. The probability of a "random point is picked in the home team's point distribution being greater than the random point picked in road team's point distribution" gives an estimate of the home court advantage.


Comments: Home Court Advantage is usually valued at 3-5 points.
 
How the NBA Schedule is Made

Download NBA schedules in excel including rest days to build your own strategy!

Matt Winick, who has been the architect of the NBA schedule for more than 20 seasons, is the NBA's vice president of scheduling and game operations. Winick starts it in February. 6 months later, in first week of August, the final schedule is completed. His interview with ESPN, unveils how he responds to complaints about strength of schedule and common questions on building the NBA schedule.

According to Matt Winick, the NBA sets the league schedule to accomplish both competitive balance and a reduction of costs. The goal of the NBA schedule, as it is constructed, is to be efficient from a competitive standpoint with an indirect consideration of travel costs.

Matt Winick has a complicated system that assigns a point value to each date or series of dates a team makes available. The point system rewards a team for making several consecutive dates available instead of insisting on a particular date. Each time team must amass at least 50 points. Generally, teams play 3.5 games in a week and those 82 games take roughly 165 days through the end of regular season.

Factors that have an impact on setting NBA schedule can be summarized as follows:

1. NBA SCHEDULING FORMULA

Each team have to play:
  • 4 games against the other 4 division opponents, [4x4=16 games]
  • 4 games against 6 (out-of-division) conference opponents, [4x6=24 games]
  • 3 games against the remaining 4 conference teams, [3x4=12 games]
  • 2 games against teams in the opposing conference. [2x15=30 games]

A five year rotation determines which out-of-division conference teams are played only 3 times.

2. COURT AVAILABILITY

All teams, about a month before the end of the preceding regular season, have to submit to the NBA office a list of:
  • at least 50 dates on which their home court will be available,
  • 4 Mondays
  • 4 Thursdays (to help TNT plan its telecasts).

3. OFFICIAL BREAKS (on which no games are played)

  • Christmas eve,
  • The all-star game,
  • NCAA championship game,

4. CONFLICTS

The conflicts such as NHL games on the same court have to be resolved.

5. BROADCASTERS

Games can be moved to satisfy the NBA's TV partners (ABC, ESPN and TNT). Game times can be tweaked.

 
Journal of Quantitative Analysis

Journal of Quantitative Analysis in Sports is being edited by Ben Alamar who works for the Thunder as senior quantitative analyst.

 
Logistic Regression Markov Chain (LRMC)

LRMC (Logistic Regression Markov Chain) is a college basketball rankings system designed to use only basic scoreboard data, including which teams played, which team had home court advantage and the margin of victory.

It was originally designed by Joel Sokol and Paul Kvam and has been maintained and improved by Sokol and George Nemhauser, all three optimization and statistics professors in the Stewart School of Industrial and Systems Engineering at Georgia Tech.


Comments: The system objectively measures each team’s performance in every game it plays, and mathematically balances all of those outcomes to determine an overall ranking.
 
MIT Sloan Sports Analytics Conference

In 2004, MIT Sloan and Daryl Morey initiated one of the first MBA programs with a sports analytics business class.

This began a focus in sports business that was extended when Daryl Morey and Jessica Gelman founded the MIT Sloan Sports Analytics Conference in the winter of 2006.

The inaugural conference in 2007 was highlighted by keynote speakers JP Ricciardi and Jamie McCourt. Panel topics included baseball analytics featuring Bill James, sponsorship across all leagues, league management/expansion, and careers in sports.

The 2008 conference doubled in attendance as Wyc Grousbeck was the keynote speaker. Featured panels included Defending the Title, which included General Managers or Decision-makers from the then reigning champions of the four major sports leagues; Bill Polian (President, Indianapolis Colts), RC Buford (GM, San Antonio Spurs), Brian Burke (then-GM, Anaheim Ducks), and Jed Hoyer (Assistant GM, Boston Red Sox).

Building on continued success and growth, the 2009 conference transitioned to a featured panel format and again doubled in attendance. The first featured panel was Evolution of the Fan Experience, which was moderated by Bill Simmons and included Jeff Van Gundy and Brian Burke looking at how new technology, stadium design, game innovations, and customer initiatives are taking the fan experience to the next level. The second featured panel, Value of Icon Players, included Carla Christofferson and all-star guard Ray Allen discussing how to quantify the value icon players bring to a team or city. Other featured speakers included Adam Silver talking about the evolving value of new media and sports, Jonathan Kraft speaking on the globalization of sports, and Mark Cuban debating basketball analytics with some of the leagues top analytics users.

The 2010 conference was again a tremendous success, attracting over 1,000 attendees with another 400 on the waitlist. It also marked the first time the conference was held away from the MIT campus, moving to the Boston Convention & Exhibition Center. The feature panel was What Geeks Don’t Get: The Limits of Moneyball, and included Mark Cuban, Jonathan Kraft, Daryl Morey, Bill Polian, and Bill Simmons, and was moderated by Michael Lewis. The panel explored the decision making processes general managers and owners go through beyond the numbers. Beyond panel discussions, 2010 also saw the introduction of the research paper track, an extremely popular addition.

Watch the panels presented at MIT Sloan Sports Analytics Conference.

 
Moneyball Approach

A research-driven approach that relies heavily on empirical analysis of player performance.

Oakland Athletics' general manager Billy Beane built successful baseball teams year after year, rather than relying on the gut instincts of old-time scouts, as was standard practice for decades. Writer Micheal Lewis realized another investment game was being played out in baseball, notably by the A's. He gained inside access to A's general manager Billy Beane and got a look at how Beane's value players differently than other teams. In 2003, he wrote a book called Moneyball: The Art of Winning an Unfair Game

The Oakland organization assesses offensive production differently than others, stressing on-base percentage and power, de-emphasizing stolen bases and putting the ball in play. It has engendered an approach to acquiring talent based as much on statistical achievement as on traditional tools, an approach that has gripped some franchises and galled many traditionalists.

 
NBA Teams That Have Analytics Department

The list includes the NBA teams using advanced stats by either employing basketball analytics professionals or working with statistical consultants.



Comments: Contribute to this list by sending your tips / corrections to

NBAstuffer.com does not guarantee the accuracy or timeliness of any information on this list.

 
Neural Networks

Neural networks are one of the machine learning systems in sports. By the help of neural networks, datasets are learned by the system and hidden trends can be revealed for creating a competitive edge.

Simulations and machine learning systems means a lot for sports analytics. The ability to apply statistics and rigorous mathematical models to provide instantaneous results is quickly becoming an invaluable commodity. Systems of this type vary from simulations that model an entire upcoming season’s worth of data to identify the best chance of winning, to simulations that identify weaknesses in motion and offer advice for correcting them.

Other than statistical predictive algorithms, neural networks can be used as sports betting and fantasy league tools in following ways;
* [STAT MINING ]: Locate common characteristics in large amounts of data,
* [ODDS SETTING]: Predict probability of an outcome,
* [MOMENTUM MENTALITY]: Forecast trends based on previous data,
* [SIMILAR PLAYERS]: Group potential assets or players based on similarities.

Other machine learning techniques include;
* genetic algorithm,
* the ID3 decision tree algorithm,
* a regression-based variant of the Support Vector Machine (SVM) classifier, called Support Vector Regression (SVR).

 
New England Statistics Symposium NESSIS

NESSIS is an event in which statisticians from all over the world come together and present academical studies. NBA and other sports analytics community has been benefiting from NESSIS that many studies on statistics regarding professional sports are being introduced first time at NESSIS just like MIT Sloan Sports Analytics Conference.

The first New England Statistics Symposium was held at the University of Connecticut, which traditionally hosts the Symposium on alternate years. Now, The Department of Statistics of Harvard University hosts the New England Statistics Symposium.

 
Play by Play

Historical play-by-play data sets are available for download in CSV format. Data sets include a separate CSV file for each game.

Traditional box score shows per-game totals for players and for teams and reveals only a fraction of what happens in a game and that the information therein is often misleading, especially at defensive stats. At this point, Play-by-Play (PBP) data has been the main source of many advanced stats such as adjusted plus-minus.

Play-by-play provides a transcript of the game in a format of individual events.
A typical play-by-play data should have following informations:
* The time of the possession,
* The player who initiated the possession (in the case of a steal or defensive rebound,
* The opposing player who initiated the possession (in case of a missed shot or turnover) including the location on the floor the shot was taken from, and some other unique identifiers we use to classify the the type of possession

Play-by-play data is being tracked since the 1996-1997 season.

 
Pro Basketball Forecast

With the Pro Basketball Forecast, John Hollinger takes an in-depth and insightful look at the game. Downplayed are all the per-game statistics; in their place are points, rebounds, and assists per forty minutes. Hollinger also examines how many possessions each player uses and what percentage of his team's rebounds he collects.

 
Regression To The Mean

Regression to the mean refers to the fact that those with extreme scores on any measure at one point in time will, for purely statistical reasons, probably have less extreme scores the next time they are tested. Scores always involve a little bit of luck. Real situations fall between these two extremes: scores are a combination of skill and luck.

How it applies to basketball statistics is, any athlete who posts a significant outlier, whether as a rookie or particularly after their prime years can be expected to perform more in line with their established standards of performance.

Statistical analysts have long recognized the effect of regression to the mean in sports; they even have a special name for it: the "Sophomore Slump". For example, Carmelo Anthony had an outstanding rookie season in 2004. It was so outstanding, in fact, that he couldn't possibly be expected to repeat it in 2005. Anthony's numbers had slightly dropped from his torrid rookie season. John Hollinger has an alternate name for the law of regression to the mean: the "fluke rule", while Bill James calls it the "Plexiglass Principle".

Regression to the mean in sports performance produced the "Sports Illustrated Jinx" superstition, in all probability. Athletes believe that being on the cover of Sports Illustrated jinxes their future performance, where this apparent jinx was an artifact of regression.

 
SportVu

SportVu Player Tracking Stats are made available at NBA.com by the 2013-2014 NBA season

SportVU is an automated ID and tracking technology that has the ability to collect positioning data of the ball, players and referees during a game. SportVu, which is founded in Israel in 2005, was acquired by STATS LLC in 2008 December.

SportVU tells us defensive and offensive alignment relative to ball location, shot trajectory and the number of dribbles and passes made by a team and player.

Players, referees and the ball with an exact location in x,y coordinates are tracked. 25 times a second, software analyzes the video, and stores information about where everyone is and what is occurring. 1,000,000 entries per game is added to databases. Without having to chart and track a game by hand and eye, player position and defender proximity to a player are now available. In another words, defense is more quantifiable with what SportVu brings to the table. Here's an article on how the Raptors use SportVu technology.

* The first five teams to have SportVu cameras are Dallas, Houston, Oklahoma City, San Antonio and Golden State.
Starting from 2013-14 season, all 30 teams have installed SportVu cameras thanks to the NBA & Stats LLC partnership.

 
StatsCube

StatsCube is a new data warehouse tool which the NBA has been developing over the last few years.
Starting by 2009-10 season, StatsCube is available to every team. It provides instantaneous statistical analysis, calculations and produce adjustable notifications and customized reports.

StatsCube has in it every point, rebound, assist, steal, block, turnover, missed shot, foul and substitution since the 1996-97 season, when play-by-play data first started being tracked courtside. The point in the game when each occurred, and what players were on the floor at the time, is recorded. Best of all, StatsCube can slice and dice the data so teams can analyze it instantaneously.

Say a team's power forward grabs only X rebounds per minute. Is the team better when he's on the floor? What players shoot the best in the last three minutes of a close game? Does a center block as many shots after he's picked up his fourth foul as he does when he has fewer than four?
With a little bit of training, StatsCube can answer all of those questions within seconds. The data always has been available to teams, but they've never before been able to access it so quickly and easily with StatsCube.

What's Next: During Game 4 of the 2009 Finals, the NBA did a demo of a system in development that tracks player and ball movement through the use of six HD cameras placed around the arena. This technology, called SportVu, gives the precise location of every one on the floor so that the play that is run, all of the the ball touches and the positioning for anybody on the court in terms of x,y coordinates is a matter of few clicks. Even the trajectory of the ball as it went to the basket will be available.

 
The Hot Hand Myth

The "hot hand" describes the belief that the performance of a player, temporarily improves following a string of successes.

Players, coaches, commentators and fans believe in streaky shooting, but the academic studies against this conventional wisdom suggest that there is no player with the hot hand. Analysis through play-by-play data strongly considers streaks of made shots are some sort of natural variation.

Making 10 consecutive shots does not prove that a player is hot. NBA players tend to become significantly overconfident after making consecutive shots. After making one shot, a player's shooting percentage actually drops for the next field goal attempt. As if the player and his teammates believed him to be the team's best scoring option. Behaving as though the hot hand existed might actually be detrimental and cost an average team about four victories over one season!

See the analysis of "a hot hand", breakdown of Kobe's 81 points.

 
The Price of Anarchy
Author: Brian Skinner

It’s a suboptimal arrangement used in "traffic networks" that can be applied to Basketball as well.

ESPN.com sportswriter Bill Simmons calles it the “Ewing Theory”. The idea that a team could improve after losing one of its best players may in fact have a network-based justification, and not just a psychological one.

Optimizing the performance of a basketball offense may be viewed as a network problem, wherein each play represents a "pathway" through which the ball and players may move from origin (the in-bounds pass) to goal (the basket). Effective field goal percentages from the resulting shot attempts can be used to characterize the efficiency of each pathway. The analysis suggests that there may be a significant difference between taking the highest-percentage shot each time down the court and playing the most efficient possible game. There may also be an analogue of Braess's Paradox in basketball, such that removing a key player from a team can result in the improvement of the team's offensive efficiency.

 


All | | A | B | C | F | G | H | J | L | M | N | P | R | S | T


Glossary V2.0