Looking At Basketball Data In a New Way with Vantage
By Ed Isaacson
For those who are familiar with me and my thoughts on what has become a statistical “revolution” in basketball, you know that I have not been a big fan. I have nothing against numbers and they have their usefulness, but my largest problem has always been the way data was being manipulated and analyzed. Often, the conclusions being made either gave little to no real insight, or were, at worst, inconsequential.
Now, I do use numbers a lot in my scouting work, but it is often just used as a check against what I have already watched. Much more often than not, it checks out. But where I find the problems are that it is easy to tell when conclusions about a player are just being made off of numbers alone, and in scouting, that provides little value.
A few weeks ago, someone on Twitter had passed along a blog dealing with basketball analytics, and in checking it out, I was stunned with what I was reading. Numbers were being applied in a way that made real sense and told a deeper story.
The company behind the blog, Vantage, is now looking to bring their unique statistical take to a much broader audience – the public. I spoke to Vantage’s CEO, Brett McDonald, about how his company, and it’s take on analytics came to be, and about what they are looking to offer.
“Publicly reported stats just weren’t answering the questions people and teams had,” said McDonald when discussing how Vantage came to be.
To ensure that Vantage was going to try and fill that void in the best way possible, they brought in Ryan Blake, the Senior Director of Basketball Operations for the NBA, to help them focus on what was really important in all the data.
“He came to us with a wish list of what he, and others, wanted to see from the numbers, and we looked to provide it with a mixture of technology and basketball analysts,” McDonald said.
After just a month as working as a consultant, McDonald said Blake was so convinced that Vantage was going to revolutionize the game that he joined the company as President of Basketball Operations.
From there, the company worked with Blake, as well as other NBA executives and professionals, to refine the data and create the perfect data set.
According to McDonald, the challenge was now trying to figure out how to capture the amount of data they needed and in an efficient way. To handle that, Vantage hired technologist Cameron Tangney from Google.
The result has been Vantage being able to capture over 16,000 unique data points from every game.
“From the beginning, we have put lots of focus into the accuracy of our data,” explained McDonald.
An NBA study of Vantage’s data over the summer concluded that the data was 99.6% accurate, an unheard of number in basketball analytics.
I had the opportunity to look at Vantage’s work and what they were doing with the data, and the results were staggering to me. In many ways, I felt that someone had been listening to my complaints all along.
Looking at the amount of ways Vantage breaks down a game is too much to detail here, but you can see a brief example here in this look at how a simple missed shot by Chris Paul is broken down:
Not only is Vantage breaking down the data (I counted 7 different levels of shot defense they look for) in a way that hasn’t been done before, but they are analyzing and presenting results, and even new statistics which make more sense than what is available.
One of the major traditional stats I have always had a problem with is “Assists.” The use of “Assist Percentage” was a step in the right direction, but still wasn’t very reliable in determining a player’s true distributing capabilities. Vantage has closed this gap considerably with the use of what they call “Assist +.”
“Assist +” not only tracks traditional assists, but also gives credit to a passer when a pass results in a shooting foul, a pass results in a missed open shot, and when a pass is deemed crucial to an assist or that pass leads to a shooting foul (also known as the “hockey assist”.)
There is much more work like this that Vantage is currently doing, and as I mentioned in the beginning, they want to bring their work to the public through ProScout. In order to do this, a Kickstarter campaign launched yesterday in order to provide the funding needed to accomplish this.
Here is a look at the Vantage ProScout product
They are bringing meaningful data in a visually appealing, and easy to manipulate, way.
You can find out much, much more about the company, their data, work, and ProScout at their Kickstarter page.
Technology and data will never replace actually spending the time and watching a player for scouting purposes. It can confirm what you’ve seen, or make you think about what you’ve watched in a different way.
If you go to Vantage’s blog you can see some analysis they did before this past June’s NBA Draft. The results were fascinating and I can already vision how this tool can be used by NBA front offices for college scouting purposes when Vantage gets to that point.
As I stated at the beginning, I have never been the biggest fan of how statistics have been used for scouting purposes. Vantage is the first company to make me re-think my dislike. I recommend checking out the links provided for a much more comprehensive look at their work, and if you are a fan, consider contributing to the Kickstarter campaign.
Vantage Links
Follow me on Twitter – @NBADraftBlog