'21-'22 NBA Statistics Analysis (Tableau)

Introduction

This project consisted of a quick case study visualizing player/team statistics from the 2021-2022 NBA season. The goal was to create visuals in Tableau and explore using the "Stories" feature, while finding out if there were any insights to make note of along the way. The data for this project can be found online here, and was available to download as a .csv file.

Analysis and Insights

Prior to any analysis, one thing I noticed was the data was a little dirty - if a player had been traded mid-season, this created two rows (one for each team they played on) and a total row summing their statistics from the season. This was important to take note of and impacted how I implemented filters in Tableau.

I created 4 charts on separate pages, then added each into a story to search for any insights - the data story is laid out below.

First, I created a bubble chart that plotted assists vs points scored, and the size of the bubble varied depending on the player's total rebounds. As one would expect, the majority of centers (C) have lower assist numbers, whereas point guards (PG) typically have much more assists - this is due to the nature of the positions, where the PG is typically handling the ball more and "running" the offense. The key takeaway however, is the domination by Nikola Jokic, who is a center but has a tremendous amount of assists, as well as a being a top-5 scorer and an excellent rebounder.

Next, I created a heatmap that showed average 3-point shooting percentage by team and position. The intention for this chart is that someone (an assistant coach for example) could quickly look and easily be able to pick out where the outliers are. This could drastically change the defensive game plan that a coach would implement for a given game - for example, the huge contrast between 3-point percentage for the center position between UTA and SAS would definitely have an impact on how an opposing team would plan to defend.

The chart above is a stacked bar chart that shows each team's total points, as well as individual player point tally's. It is definitely interesting to see which teams have one main scorer vs two or three main scorers - that distinction I'm sure has implications on an opposing teams defensive game plan. One thing that stands out is that IND and POR do not seem to have any high-volume scorer (no player scored over a 1,000 points). While POR is towards the bottom of the league in points scored, IND actually hovered right around the middle of the league in points scored and did so without a clear scoring threat - this would indicate to me a system more reliant on an offensive scheme rather than one or a few individual player(s) and that the bench (players that do not start) is relied on more to contribute to scoring than that of other teams.

Finally, I created a treemap that categorized assists by position. For the most part it is what you would expect, the point guards (PG) have higher assist numbers than the other positions The shooting guard (SG) and two forward positions (PF and SF) have less than that of the point guards, and finally the centers (C) have the lowest assist number. All of that makes sense when taking in the context of the game and the typical roles that each position has. There is not a great deal of separation between the top 10 of each position besides the one glaring exception - Nikola Jokic! This chart, similar to the bubble chart from earlier, clearly shows how Jokic is a standout player who not only is an outlier for his position, but that his assist statistics are actually more in line with a top-performing point guard! These number clearly indicate why he was selected as the NBA's Most Valuable Player (MVP) for this season

.