A3-StephenChu

From CS 294-10 Visualization Sp10

Jump to: navigation, search

Domain: I'm a big basketball fan, so I decided to focus on NBA related data.

Initial question: Kobe Bryant is the the face of the Los Angeles Lakers. Whether you're a fan of his or not, no one can debate his importance to the organization. I want to understand the following questions: What is Kobe's career contribution to the LA Lakers? What is his worth to the franchise?

Data sources: 1) Basketball-Reference.com, http://www.basketball-reference.com/players/b/bryanko01.html, http://www.basketball-reference.com/teams/LAL/

2) "Top 10 most (and least) popular jerseys from the past 10 years", http://sports.yahoo.com/nba/blog/ball_dont_lie/post/Top-10-most-and-least-popular-jerseys-from-the?urn=nba,73743

3) "Kobe Bryant's Jersey is Number One Selling Jersey in Europe", http://www.examiner.com/x-24394-Kobe-Bryant-Examiner~y2009m10d12-Kobe-Bryants-Jersey-is-Number-One-Selling-Jersey-in-Europe

4) "Kobe remains with Lakers", http://sports.espn.go.com/nba/news/story?id=1840336

Tools: Tableau


Process

My first step is to collect all the data that shows Kobe's importance to the Lakers. An easy place to start is his game statistics. Much of the emphasized numbers in basketball are the points/game, rebounds/game, and assists/game averages. For most fans, much of a player's ability can be judged with these three numbers. NBA.com reflects this opinion by highlighting these 3 figures in each of its player profiles. My first step was to figure out how to value Kobe's contribution as a player, so I began exploring his points/game by season.

Here is a simple graph for Kobe's pts/game by season.

Image:Kobe_ptsavg.jpg

What does this visual tell us about Kobe's worth as a basketball player? Fans of the game realize that he is a prolific scorer because they have prior knowledge that 20+ pts/g is an impressive figure. However, if the reader is not familiar with basketball numbers, s/he will not know what constitutes a high level scorer. Furthermore, how valuable is Kobe's 25 points a night to the Lakers? What if he has three other teammates that score over 20 pts/g? In that case, surely Kobe's value to the Lakers would not be as amazing. The problem with using only game averages is that these numbers are not put into the context of the team.

To understand why the Lakers have paid Kobe nearly $200 million dollars, we need to incorporate team-related statistics involving Kobe. We cannot just focus on individual scoring, rebounding, assists, etc. Ideally, we want to understand the size of Kobe's contribution in the context of the team.

Another complication I had during the design was figuring out how to plot all the different game stats together. I wanted to include points, rebounds, assists, turnovers in one plot, but these attributes scale differently. 20 assists is more impressive than 20 points. Furthermore, 5 blocks per game is more difficult to achieve than than 20 points per game. However, if placed everything in the same graph with the same scale used above, an interesting stat like 5 blocks/game would be overshadowed by the length of the points/game column.

Here is a graph of Kobe's pts/game, rebounds/game, and assists/game by season using the same scale.

Image:Kobe_3statsavg.jpg

Kobe's rebounds and assists contributions are grossly overshadowed by scoring. Only readers who know that points are generally easier to come by than rebounds or assists will understand that Kobe has a significant impact on rebounding and assisting. The great difference in length and position of the columns is visually distracting and detracts from part of Kobe's value. I needed to find another way to show Kobe's game stats.


To solve the problem of relating raw numbers to Kobe's contribution to the team, I have decided to transform the data into percentages. For example, instead of focusing on pts/game, I will use the percentage of Laker points scored by Kobe. Instead of listing his yearly salary, I will show the percentage of the Lakers player salary budget rewarded to Kobe. Using the percentage scale for the y-axis, all the statistics can now be placed on one graph, and it is now easy to understand how much Kobe Bryant means to the Lakers.

Data transformation

I switched focus from Kobe's averages to Kobe's total contribution per season. I also used data from a spreadsheet of the Laker's season totals (total team points, rebounds, assists, etc). From these two datasets, I calculated Kobe's percentage of contribution to the Lakers in each of the categories I thought was important.

Here is a link to the compiled stats of Kobe's % contribution to the team. I included points, rebounds, assists, and steals.

[Kobe % Contribution: http://vis.berkeley.edu/courses/cs294-10-sp10/wiki/images/5/5f/Kobe_percentage.txt]

Here is a bar chart visual of the percent contribution:

Image:Kobe_percentages.jpg

Transforming the data allowed us to get a deeper sense of Kobe's worth. However, there are disadvantages to this graph. This graph is a bit packed and it is hard to focus on just 1 attribute at a time. It is difficult to find trends because of the competing bars and clashing colors. For example, to find trends in the points category, you have to look at one of the points bars and then skip the next 3 bars to look at the next year's points bar. This does not make for an effective tool to focus on single attributes. You just get a feel of Kobe's overall production.

To make the visual more effective, I turned it into a line graph:

Image:Kobe_percentages_line.jpg

This visual is much less cluttered, allowing users quicker, easier focus. Furthermore, it reveals trends much better. This graph also raises some interesting questions. I wonder why Kobe had such a significant spike in 2005-2006 season. Looking back on his stats, I see that he had his best stats ever that year. Also, the Lakers had a very rare losing season in 2004-2005, which probably motivated Kobe to not repeat such a disappointing performance. Also, Phil Jackson, a Hall of Fame coach, came back to the Lakers in 2005-2006.

It's also interesting to notice the dip in production in 2003-2004. The Lakers had a great year (though disappointing to their fans because of hyped up expectations to win the championship). How did the Lakers do so well with Kobe not producing as much? In that single season, the Lakers acquired Karl Malone and Gary Payton, two Hall of Fame players. With 4 HOF players in the starting lineup, Kobe did not have to carry as much of a burden as he did in other seasons.

Compensation

I found myself wondering about Kobe's monetary value. How much money has the Lakers given him over the course of his career? What is his current worth? What is his current contract? I discovered that in 13 years he has earned a total of $148,347,990 from the Lakers. The 2008-2009 season was his 5th year of a 7-year contract worth approximately $136.4 million. He was paid $21,262,500 in 2008-2009. And his current salary is $23,034,375 in 2009-2010.

I decided I want to include data about Kobe's yearly earnings in the visualization because it is an important metric of Kobe's value according to the Lakers. I wanted to keep everything in one graph, but the salary numbers did not scale well with the game stats. I also did not want to use different scales on the same axis because this can cause confusion, so I tried plotting game stats on one graph and salary on another. However, having two graphs creates competition between them, and I don't want one of the plots to detract from the other. Furthermore, listing Kobe's salary provides his "market value" as a basketball player, but those who are not familiar with NBA salaries may not understand the magnitude of his contract.

I decided to stick to the percentages. I found data for Kobe's salary by year, and calculated the percentage of the team's player salary budget allocated to Kobe.

Final visualization

Image:Kobe_lakers.jpg

Using this graph, we can see the game contributions Kobe makes to the team and track trends of individual stats. The dips and peaks are related to circumstances of the corresponding years such as the adding/removing of players. This graph gives us more information than a graph of averages because the viewer now understands the makeup of the team. The fact that many NBA teams can have 13 players makes Kobe's effort even more impressive. Admittedly, it is hard to compare salary with game stats because there are many intangibles in the sport, and Kobe's popularity makes him command an enormous contract. However, it is beneficial to have the salary line for extra information.









I wrote the following below before I realized that I couldn't visualize/manipulate this data effectively using the visualization software I had. The part below is no longer part my homework.

It is easy to look at Kobe's game statistics and recognize that he is a talented player. However, many players have great stats and awards/titles differentiate between good and great players. These awards should be shown in the visual to fully understand Kobe's impact. Furthermore, these titles are difficult to graph, so I had to come up with another way to visualize them. There are pros and cons to using these achievements as data. Although these awards have been voted on by basketball "experts," some would say that they are too subjective and based on popularity.

Career Achievements/Awards

Kobe has played 14 seasons for the Lakers from the 1996-1997 to 2009-2010 seasons. Kobe is the All-Time Leading Scorer for the LA Lakers. He was the 2007-2008 MVP and 2008-2009 Finals MVP. He has contributed to 4 NBA Championships and has finished in the Top 5 of MVP Voting 7 Times. He also has been a Two-Time NBA Scoring Champion, 11-Time All-NBA Team, 9-Time All-NBA Defensive Team. He has represented the LA Lakers as a NBA All-Star 12 times and won the NBA All-Star MVP 3 times.

Popularity and Exposure

Now, I know how much the Lakers are willing to pay to retain Kobe, but what does he give back to the team besides great playing ability? What is his draw, from a fan and consumer perspective? His popularity as a basketball player? I decided to focus on jersey sales because they give some indication of the popularity of players. According to "Top 10 most (and least) popular jerseys from the past 10 years", Kobe's jersey was the 2nd highest selling jersey from the NBA store over the last 10 years (from 2008). According to ESPN, Kobe had the highest selling NBA Jerseys in 2009 and 2007. Furthermore, his jersey was the highest selling basketball player jersey in China, Europe, and the US in 2009. This shows his popularity and the enormous fan following that he brings to the Lakers.

I cannot use a graph to effectively visualize Kobe's career achievements and his popularity. Although I could possibly chart Kobe's yearly jersey sales, I do not have that information. I think the best way to visualize the data is through text. This leads to another question: What are Kobe's most important achievements and awards? Surely this is subjective, but writing Kobe's accomplishments in the same size/type text will make it unclear what the most important awards are.

Here is a simple, completely text visualization of Kobe's awards without any variation in text font, size, or color.

Image:Kobe_achievements_sametext.png

The viewer is left confused about the order of importance. If we were to guess about the ordering, most would instinctively assume that the awards are most to least impressive going from top to bottom. To make it more clear to the viewer the significance of each line, I will vary the font style, size, and color. Larger size means (generally) more important. Colors are used to differentiate between the awards. Style, such as bolding, can also be used for emphasis.



[add comment]
Personal tools