Positions of Power: Basketball Data Project

Data sourced from Basketball Reference (https://www.basketball-reference.com/) and Stat Muse (https://www.statmuse.com/)

Introduction:

There are five positions in basketball. Point guard, shooting guard, small forward, power forward and center. Unlike in other sports, the positions are not rigid in terms of role. In soccer, the strikers are the ones that score, in cricket batters and bowlers rarely exchange responsibilities. However in basketball scoring can come from anyone on the court, as can defense, passing and rebounding.

In the past, the different positions were mostly defined by skill based characteristics. While this still comes into consideration, in the last ten years positions are more often defined by the size of the player and the position they are able to guard on the opposing team.

Within a basketball team there will never be an equal distribution of importance, capability or skill. This project was born out of the idea that there has to be a superior positional make up for a winning team. Which position sees the most team success as the dominant player? Which position is most suited to be the ‘second fiddle’? And what do the side players contribute to the team’s triumph?

Overall I’m attempting to answer the question, What does a championship basketball team look like?

Method:

I collected statistics from NBA championship winning teams from the 2000-2001 Lakers to the 2020-2021 Bucks. I chose the players based on the most common lineups the team used throughout that season.

I selected statistical categories that I felt were the most useful to tell the story I wanted to tell. I made a table for each team using those statistical categories. I omitted unnecessary details, like names, that might distract from the project.

	Offensive Rating	Defensive Rating	Win Shares	Scoring Rank (1-5)	3 Pointers Attempted
Point Guard
Shooting Guard
Small Forward
Power Forward
Center

^Example of table

Usage:

An estimate of the percentage of team plays used by a player while they were on the floor (Basketball Reference).

Because usage percentage was the foremost statistical category, I decided to display the data in multiple ways. The above graph shows how many times each position was the most used player in the lineup and second most etc. I used rankings to show, not only the player’s usage amongst others of their position, but also in relation to the rest of the team. For example, we can see that Power Forwards are the most likely to be the most ball dominant and Centers are the least likely.

An alternate way to read the data is to look at high vs low usage (below). The average UR for all players was 21.5, so I collected the URs that were below 15 and above 25. From this data we can see that high usage shooting guards are the most represented group. Additionally they have the least number of low usage players. Meaning they are the most likely position to be used in a play.

Win Shares:

An estimate of the number of wins contributed by a player (Basketball Reference).

The Others:

To represent those that aren’t the superstar of their team or even the second or third best player, I have chosen two skill-specific statistical categories. Defensive rating and 3 pointers attempted per game. I wanted to determine what archetypes fill out the line up after the top two or three. Elite defenders and three-point shooters are able to thrive without the ball.

Defensive Rating:

An estimate of points allowed per 100 possessions (Basketball Reference). It is better to have a low defensive rating.

Out of the 105 players I used data from, 24 of them have a defensive rating below 100. It’s safe to call these players elite defenders.

3 pointers attempted per game

I chose to compare the players 3-pointers made per-game to the 2009-10 season because 3 pointers are becoming more and more common. 2009-10 was the median season.

Results:

So, according to my research, what would the most successful composition of a basketball team be? I will describe the ultimate championship team position by position.

Point Guard:

His usage is high, but his scoring is relatively low. This suggests he is a passer, someone to set up the real luminaries of the team. He is the worst defender, but shoots a high number of 3-pointers.

Shooting Guard:

He is offensively the most reliable. He is the one of the main scorers. He isn’t an unbelievable defender, but definitely can make it hard for the opposing teams best guard. He is the second best scorer on the team.

Small Forward:

He is a 3-point specialist. Unlike the SG, he is unwilling or potentially unable to create his own shot. He almost always shoot off of passes. He acts as a relief for when the stars get double teamed.

Power Forward:

The Prometheus of the team. Their success lives and dies with him. Ideally he would be considered to be one of the all time greats. He is dominant both offensively and defensively. He isn’t a high volume deep range shooter, but makes up for it with his post presence.

Center:

He is mostly a utility player. Pulls down rebounds, sets screens, creates space for his teammates. He is useful defensively, his height allows him to easily deter shots close to the rim. He can score but certainly not to be depended on in that department.

Leon's notes:

This was an assignment for my data journalism class, a class I throughly didn't enjoy. Too mathy. However I did see a path of less trouble, that would make the assignments a breeze. My undying obsession with basketball.

In the best and worst way, basketball, its history and sports in general can be dumbed down into numbers. I remember telling my data journalism classmates that if you weren’t doing your assignments on sports you were making it harder for yourself on purpose. The statistics are all there, so clear, so accessible. "No!" They cried "I must make it about politics!". Kicking and screaming, hypnotised by a wall of polling data.

-

I had had this concept on my mind for a really long time. 2014 finals for example. I remember recognising that Danny Green and Tiago Splitter (SG and C) were the two least consequential Spurs starters. While for Miami, it was Mario Chalmers and Rashard Lewis (PG and PF). I then thought every team must have an observable positional make up like these. So which is most successful?

It was probably my first sophisticated NBA thought.

I wrote an impassioned pitch letter to my professor, I stopped just short of saying I was put on this earth to write this piece. He isn't really a sports guy, but he saw my vision.