I decided to investigate the National Basketball Association. More specifically I decided to explore the relationship between the amount an NBA team spends on their entire roster and on individual players to the amount of games they won in the regular season and the playoffs. I wanted to find out if teams had to spend more money as a whole or on individual players to have success. I started my search in the 2002-03 season because this was the first season that the Luxury Tax was instituted in the NBA. The Luxury Tax requires teams to pay an amount to the league if they are over the salary cap. This should hypothetically discourage teams from spending too much money and not going over the cap.
Through the analysis of data, I will explore teams’ relationship with spending and regular season and playoff wins, their relationship with having to pay luxury tax money and winning in the regular season playoffs, and individual players likelihood to contribute to winning teams based on their salaries. Through looking at the analysis, I will show that there is only a small relationship between spending money and win percentage. However, on an individual level we will find that having one of the highest paid players in the league has not been a recipe for individual or team success.
The sources for this project were easy to find. I just had to find the total for salaries, regular season wins, playoff wins and luxury tax payments. This data required little math besides the adjustment for inflation. All of the data that is based in dollars has been adjusted for inflation and represents the value in 2016.
One point of note is my decision to use total wins in the regular season and playoffs, instead of win shares or Player Efficiency Rating, to judge top paid players. I choose this because I felt at the end of the day, the most important thing for a player to do was to contribute highly to a winning team because that is what the fans and management want the most.
As for player’s salaries I decided to focus one section on the teams’ winning percentage and then have one section focusing on the top ten highest paid players in the league. One would assume that this would be a component of a championship team or at least be one more often then not. However, the data will reveal that this is not the case.
The reader should not that the 2004-05 season has been left out because there was not a Luxury Tax during this season. In addition, the 2015-16 season is left out of the discussion when it comes to the playoffs because it has not finished yet.
In addition, keep in mind that the maximum amount of games a team can win in the regular is 82, with the record being 72, and 16 is the maximum amount a team can win the playoffs, which would result in a championship.
One of the most consistent design choices I made throughout each visualization was the use of the colors red and blue. One of the reasons why these colors were used is that it is the same color that the NBA uses in their logo of red, white and blue. This helps to set the theme of the data, so the reader is always reminded that we are viewing data about the NBA.
Another consistent choice I made was the use of scatterplots. I attempted to use some other plots like bar graphs or comparison bar graphs or even comparison scatterplots. However, I did not feel any of these were as visually appealing or easy to follow as the scatter plots.
The scatter plots help me have the reader focus on the colors of the design, so I can focus their attention where I wanted to. In addition, I was able to manipulate the size of data that I wanted to emphasize for the readers. This all allowed me to make clear and conscious choices about each point of data emphasis.
Originally, I thought about doing something to represent what city each team was from with a map of the geography. However, after trying this out I realized this was not the most visually appealing and didn’t reveal the most relevant information to the reader.
Analysis of Data Visualization 1
The first visualization examines the relationship between total wins in the regular season and salary per team per year. As having a great regular season is usually a secondary goal of a team, this section should not get as much emphasis in the readers’ mind as the playoff results.
It is important to note the outliers in this visualization. One of the teams that can be seen over and over again as an outlier on the right side of the graph is the New York Knicks. For too many years one can find the Knicks overpaying for poor results. Two examples of this are in 2006 and 2003 when they won 23 and 37 games respectively.
On the opposite side of the spectrum one can find the Philadelphia 76ers who have been doing the opposite of spending money in the recent past. The 76ers have been “tanking” for the past few years and one can find the evidence in the salary data and win column. For example, in 2016 the 76ers spent about $35,000,000, by far the lowest in the league, and won 10 games.
On the other hand, one can find examples of teams spending money well and being rewarded with regular season wins. The first example of this can be found with the 2011 Bulls and the 2004 Spurs. These two teams spent roughly $57,000,000 and $58,000,000 respectively but also had 62 and 57 wins respectively. This becomes far more impressive when one compares the 2004 Spurs to the 2004 Mavericks. In 2004 the Mavericks spent roughly 99,980,000 on their players but only won 52 games to the Spurs 57.
When examining the correlation between amount spent and total wins, we can see that it is extremely small. The exact amount is measured at 2.9175e-07
The main implication from this graph is that for regular season wins there is a small correlation between spending money and winning. Most teams fell into the middle of the graph while there were some outliers that pulled the graph either way.
Analysis of Data Visualization 2
The next graph examines whether or not it was worth it for teams to pay luxury tax money in the regular season. The first thing to note in this is that a lot of teams are choosing not to pay the luxury tax. A lot of team are steady on the zero line. As one can see however, some of these teams also went on to win a lot of games and championships.
One team to note that went way over the Luxury and didn’t get much result to show for it was the Brooklyn Nets. In 2014 the Brooklyn Nets spent roughly 91,000,000 in Luxury Tax penalties and were only rewarded with 44 wins. We can assume that ownership expected something more in the 50 or 60 range when paying that much money and, as one will see later, expect a lot more in the playoffs as well.
In addition, team that just set the record for most regular season wins, the 2016 Golden State Warriors, did not pay any luxury tax.
However, when I looked closer at the average amount teams paid in luxury tax depending on if they won the championship or not the results were significant. On average, teams that did not win the championship paid around $3,300,000 in Luxury Tax while teams that won championships paid around $6,382,315.
Analysis of Data Visualization 3
The next graph analyzes the relationship between the average salary per year of NBA teams to their total playoff wins. In order to be considered a champion one must win 16 games. Teams that did not make the playoffs or did but did not win a playoff game were not represented on this chart.
While looking at the chart one can see that the lowest amount championship team has spent on their players in a season was roughly $64,000,000 but the lowest a team that made the playoffs in general and at least won one game was the 2012 Pacers who won six playoff games.
On the other hand, one can see that a championship team has never spent more than $100,000,000 on their roster but teams have spent about $120,000,000 and have not won a championship.
When looking at the graph one can see that most championship teams had similar salaries to teams that didn’t do well and they usually fell in the middle when it came to salary.
Another implication from this is that teams don’t have to outspend or worry about cutting back compared to other playoff teams, but there have to be some decisions that championship teams make to separate themselves.
Analysis of Data Visualization 4
This visualization shows the relationship between the amount of Luxury Tax a team has to pay and their amount of total wins in the playoffs.
While looking at this graph one can see a pretty even split between the championship teams. Some teams did not pay any luxury tax while some had to pay some up to the maximum of around $23,000,000.
On the other hand, one can see teams like the 2014 Brooklyn Nets and 2003 Portland Trailblazers who paid large sums in taxes but were only rewarded with 5 and 3 playoff wins respectively.
Most playoff teams in general seem to have to pay some amount of Luxury Tax to make it into the playoffs but it does not seem like they must pay extreme amounts to get the best results.
The significance of this graph is that paying a huge amount in luxury tax does not guarantee you success but more than likely, you will have to pay some to see results in the playoffs.
Analysis of Data Visualization 5
The next two visualization shows the relationship between the percent of the total team salary of one of the top ten highest paid players in the league and their impact on their team’s performance in the regular season and playoff season. This graph specifically looks at its relationship to Regular Season wins. One would assume that having one of the highest paid players on your team would allow you to compete in the regular season.
As we can see the player who had the highest percent of his team’s salary was Kevin Garnett in the 2002-03 season. Fortunately for his team, they also went on to win 51 games.
When we next examine the player that took up the least amount of a team’s salary we find Damon Stoudemire who took up roughly 12 percent of his team’s salary. Of note is that in this year, is that the Trail Blazers have two other players who ranked in the top ten in salary, Scottie Pippen and Rasheed Wallace, and in total these players made up roughly 45 percent of their teams total roster.
The next talking point is the lack of regular season MVPs that were paid one of the top ten salaries in the NBA. Only three players Kobe Bryant in the 2007-08 season, Kevin Garnett in the 2003-04 season and LeBron James 2012-13 season won regular season MVP after being paid a top ten salary. On average these MVP players took up about 30 percent of their team’s total salary compared to the average of the other players coming at roughly 26 percent.
In addition, we can see that most teams won 50 games or less during the regular season when they had at least one top ten paid player on their team. Out of every player, 78 of them played on teams that won less than 58 games while 42 of them plaid on teams that won more than that.
Analysis of Data Visualization Six
This last visualization shows the relationship between players who were paid top ten salaries, and how much of their team’s total salary they take up, and their performances in the playoffs. Again, one would expect individual investments to yield positive results for teams in the playoffs.
The first important note is that only nine players out of the 120 players listed went on to win a championship. On average these nine players took up 25 percent of their team’s total salary which closely compares to the other players, who average out to about 26 percent.
One team that one the championship, the Miami heat in 2012-2013, had three players in the top ten in salary, LeBron James, Dwayne Wade and Chris Bosh, who accounted for about 66 percent of their team’s total salary. This team was a sort of “super team” that united in free agency in Miami and rattled off two championships in four years. If these players were not on this team, they would have probably been paid more.
In addition to just winning the championship, only 16 out of the total 120 players won 12 or more playoff games, meaning they made the finals. When we dig even deeper we can see that only 25 players out of the total 120 won eight playoff games, which meant they made it to their conference finals. Lastly, only 47 of the 120 players won four playoff games, which would equate to winning their first round series. One would assume, that when owners pay players as a top 10 player they expect to get farther than the first round of the playoffs.
Something else that should be pointed out is the dominance of certain players in the playoffs in the time period examined. Since the 2002-03 season the finals has featured either Kobe Bryant, Tim Duncan, LeBron James or Dwayne Wade. During this same time these players have eight out of the 12 finals. This goes to show that really, without one of these players, your team doesn’t have a great chance of making the finals and an equally smaller chance to win once there.
It goes to show that the percent of salary a player takes up is not as important as who the player is.
Limitations of Analysis
One limitation I felt was that I cold not find enough data for the average player salary per year. I wanted to be able to compare how much more the players I choose were being paid but I could not find enough information on this.
In addition, I wanted to examine the issue from a larger context but the salary information was too inconsistent the farther one went back, so the point of implementing Luxury Tax was chosen as the starting point.
Lastly, the 2015-16 season isn’t over yet, so one cannot include that data when it comes to playoff numbers but can when it comes to regular season numbers.
In conclusion, teams definitely have to spend money on a team basis to see success but there is not a huge correlation between win totals in the regular season and spending a lot. However, one can see that on average, championship teams paid more luxury tax money than teams that did not win championships. Investing smart is definitely the preference over investing often.
In addition, on a player level, teams were not rewarded when they gave a player at top 10 contract. As a whole, the team did not do as well in the regular season or the playoffs, which should be used as the ultimate test of a player’s performance, fair or not.