Hollingizer Mobile 0.4
Between Jay’s school load and my new job there hasn’t been a whole lot of work going on here at Hoopism. Over the last few days I’ve been playing with the Android SDK and ported the Hollingizer over to the mobile world. It’s pretty beta right now but boasts a stat calculator view and up-to-date stats thanks to Dougs Stats. Right now the stats are loaded to a sqlite db from the web and then processed based on the user supplied equation.
I need to make some additional enhancements before daring to release. The performance is a bit slow on the initial load. I think there’s some optimization I can do to get that down. I also want to include the ability to save and load queries. I also hope to get Jay to rework the UI to make it a bit more friendly.
I’ll post some updates as things progress.
NOTE: The data in the images is not accurate. This is pre-Doug Stats.
Player Genome Project: Part I
I’ve been heavily influenced recently by FreeDarko’s book “The Macrophenomenal Pro Basketball Almanac“. If you haven’t heard of it, check out the website. The infostrations (information illustrations?) are some of the most unique and well done I’ve seen in any field.
One of the chapters contains a graph displaying the anomalies that are Gerald Wallace and Josh Smith. It displays the amount of combination of plays over a given time frame. The purpose is to show how diverse and unique these two players are. The simplicity of the volume of data is what really stuck with me.
In an attempt to devise a interactive visual in a similar vain I created this. The more I began to look at it, the more I realized that it resembles a genome chart. And in a strange way, it sort of is a players DNA. The graph displays events (shots, rebounds, fouls…) quarter by quarter for each game in the 07/08 season. This example is Paul Pierce. I’d like to extend this idea to creating genomes for each player and build in a mechanism to compare players for similar sequences.
Move the mouse over the boxes to see the event name. The gray boxes display the game and quarter. Pressing 1,2,3,4 or 5 will hide and display the different events.
1 - Shots - Yellow
2 - Rebounds - Blue
3 - Fouls - White
4 - Turnover - Green
5 - Foul Shot - Light Blue
6 - Foul Given - Purple
NBA Shot Location Visualization Update
We’ve been playing with a few different ideas to allow for users to specify the data set to visualize; in the meantime I threw together a version that allows for each quarter to be shown. The data is hard coded right now, not optimal but easy and fast.
The controls are the same as before (click for reminder) but by pressing 1,2,3 or 4 you can cycle quarter by quarter. There are some obviously interesting contrasts between Q1 and Q4. I’ll be looking to extend this to point differentials as well. I’d imagine the shot selection drastically changes when a team is down 1 and down 15.
Stats Caricatures 2.0
I started messing with some data from Doug stats to see what I could make. This visualization builds caricatures based on NBA player stats for the 07-08 season.
Head Size = Shots taken
Head Aspect Ratio = Percent of shots made
Arm length = Blocked shots
Leg Length = Rebounds
Eye Size = Assists
Eye Aspect Ratio = Turnovers
Here is what I have so far.
NBA Shot Location Visualization
Matt and I collaborated this morning to put together this visualization of shot frequency for the entire 07-08 NBA season. The graphic rotates when you click and drag the mouse and you can zoom in and out with the “S” and “T” keys. We currently have 2 versions, one with cubes and one with bars. The Bars version runs a little slower than the cubes.
I think this is a good example of how location data (where on the court) lends itself to a 3D visualization. I think you may lose something by mapping this data in 2D. You can compare ours to a heat map version that appears to be constructed from the same data. Next step would be making it interactive so you could check by team or player. Thanks to BasketballGeek.com for the data.
Hollingizer Revisited
This is a project I had worked on a while back (circa Celticsblog years). I updated the interface to include the buttons and some javascript and currently use 07/08 season stats. If your unfamiliar with John Hollinger, he’s the brain behind ESPN’s PER, a statistical measure used for player comparisons. This tool allows anyone to create their very own formula to compare players.
Hoopism 1.0
Hoops as a religion. A passion. A way of not doing anything constructive.
Welcome to Hoopism. A collection of musings on all things hoops. The stats, life and times of our favorite past time.
