I chose to collect and analyze data from the 2003-2004 season of the NBA. The data was downloaded with the R SportsAnalytics package, which sources the data from http://www.dougstats.com. This data includes the players name, team, position, number of games played, total minutes played and 19 other columns pertaining to their performance throughout the season.
Prior to performing any analysis, I chose to review the dataset to look for dirty data and to generate a quick summary of a few stats. I did find and correct three players who had a typo in their team name and also decided to add three calculated columns, the total number of two-point field goals, the points from those goals and the total points from three point shots.
In the 2003-2004 season there were 29 unique teams and 442 unique players. The average number of minutes a player played throughout the season was 1,300, over 21 hours! The season saw a total of just over 222,000 points scored, over 100,000 rebounds and 12,000 blocked shots.
For an individual team, I decided to analyze the Timberwolves as they had one of the best records at 58/24 and Kevin Garnet, who was one of the top overall players during the season. When looking at the top three players from their team for total points, rebounds and blocks, you can see that Garnet was the best on the team for all three. Latrell Sprewell was also a star player for the team, but trailed Garnet in rebounds by over 800!
The top 5 teams by record was the Pacers, which was the only team in the top 5 to come from the Eastern conference. Second and third was the Timberwolves and the Spurs out of the Midwest division, followed by the Lakers and the Kings out of the Pacific division. The 6th place team in this category was the Pistons with a record of 54 and 28 and while they may have not made the top 5, they went on to win the championship against the Lakers.
Looking at the top 10 teams by total points we see a pretty close spread and oddly enough only the Kings and Lakers were in the top 5 for total wins. The Mavericks was the highest overall for points, but placed 7th in the season for number of wins, followed by the Kings, Supersonics and the Bucks.
When comparing the point distribution across the Timberwolves, we can see three players contributed the most in this category. Kevin Garnett provided just under 2,000 points, 368 from free throws, 33 from threes and 1,586 from field goals. Sam Cassell provided 1,092 field goals, 222 threes and 289 free throws. Finally, Latrell Sprewell shot 838 field goals, 297 threes and 240 free throws.
This boxplot focuses on the point distribution across all players on each of the highest scoring teams. I found it interesting to note that Michael Redd from the Bucks and Predr Stojakovic from the Kings both scored so many points, 1,776 and 1,964 respectively, they were an outlier, which is when a value is greater than 1.5 times the interquartile range, or the range within the box. The Lakers actually had two outliers, Shaquille O’Neal and Kobe Bryant, who scored 1,439 and 1,557 total points.
Reviewing the same data as the above box plot, but instead looking at it by type of goal, we see a pretty even distribution. The Kings lead in the three-point category with 1,803, but are actually in last for regular two pointers with 5,004. The Mavericks lead this category with 5,732, but are in second to last for free throws with 1,484, which the Bucks lead at 1,788. The Mavericks lead overall with 8,743 points, with the Grizzlies pulling in at 10th with 8,005 total points.
When we size up the top 10 point scorers for the season, we see an interesting distribution. Kevin Garnett, who we looked at before, was the season leader with 1,987 total points. From the previous plot we again see Predr Stojakovic and Michael Redd placing 2nd and 5th, but neither Kobe or Shaq appear on this list, both falling just outside of the top 10. While Garnett clearly dominated in the two pointer category, Stojakovic had 200 more three pointers than Tracy McGrady, who was second best and almost 22 times as many as Garnet.
Looking back from 2018 to 1999 there have only been nine different teams to win the championship. The Lakers had the most wins with five in 1999, 2000, 2001, 2008 and 2009. The Spurs won four in 2002, 2004, 2006, and 2013. The Heat won three 2005, 2011 and 2012. The Warriors also won three championships in 2014, 2016 and 2017. The Pistons, Celtics, Mavericks, Cavaliers and Raptors each only had one win, which was in 2003, 2007, 2010, 2015 and 2018 respectively.
<!DOCTYPE html PUBLIC “-//W3C//DTD XHTML 1.0 Strict//EN” “https://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd”> In conclusion I learned a lot about the 2003-2004 NBA season from this project. While the Pistons may not have made the top 5 for overall record, they went on to defeat the Lakers, who had both Kobe and Shaq in their primes. Kevin Garnet had his career high for total points and Predr Stojakovic had his career high for number of three-pointers in a season.