Measuring the Knicks’ shot quality with PBP stats

The other day while I was drinking my morning coffee before I began work, I was perusing the internet and came across a Nylon Calculus article as like any cool, hip person who isn’t a total NBA math herb and absolutely does not smoke a stoge every now and then when first-world problems add stress to one’s life does on a regular basis. The article asked the following question: why aren’t you using pbpstats.com?

Great question, Todd. Why wasn’t I using phpstats.com? Is that actually a website? Sure enough, it is a website and it’s a pretty cool one if you’re an NBA stat dork who certainly doesn’t just look at box scores, VORP, and defensive RPM and conclude that Nikola Jokic is a better defender than Kristaps Porzingis. I mean, it’s irresponsible to actually believe Jokic is a better defender than Porzingis.

Back to the subject. As I continued reading the article it briefly discussed a shot quality metric that exists on PBP Stats as well as a Tableau dashboard highlighting the shot quality differentials of a list of players. You’re more than welcomed to click the link to read the methodology on this shot quality metric if you like reading about log loss analyses. The quick-and-dirty understanding of the metric is that it measures the expected effective field goal percentage of a shot based on court location and play context via play-by-play data. For a free, publically available statistic, it’s rather sound.

After finishing the article and exploring the website, I thought to myself, “Drew, you should make a New York Knicks shot quality dashboard for Posting & Toasting because the readers totally love metrics based off of statistical analyses.” And guess what I did? Look below

[In Mike Breen’s voice] BANG! An interactive New York Knickerbockers dashboard detailing different shot quality differential data. Here is the link to go to the Tableau page to get a better, full-screen experience if the view in the article is constricting. Reading and using this dashboard on your phone is even more constricting than on a laptop or desktop. Using it is impossible when your phone is vertical, so turn it horizontally to get that full-screen stretch to make it easier. The dashboard works much better on a computer, so that’s what I recommend, but it can function on a phone when turned sideways.

It’s fairly simple to understand the data. Let’s use Knicks legend Mindaugas Kuzminskas’ 2016–17 and definitely not a Knicks’ legend Derrick Rose’s seasons as an example. According to the data, when Kuzminskas was on the court, the Knicks’ shot quality increased by 2.2 percent; when Rose was on the court, the Knicks’ shot quality decreased by 0.5 percent.

The Tableau Dashboard has three tables with different data.

1. Knickerbocker Single-Season Shot Quality Differential (top left): This table captures year-over-year shot quality differential data for all Knicks who played between the 2010–11 season through this past one. Players who played less than 100 minutes on the court were removed. Sorry Chris Smith. Use the “Knickerbocker” drop down menu in the top right corner to select which player you want to view.
2. Knicks Point Guard Shot Quality 2017–18 (top right): This table captures shot quality differentials for each Knick when Emmanuel Mudiay, Frank Ntilikina, Jarrett Jack, or Trey Burke were on the court. The more blue, the better, and the more orange, the worse (as outlined in the “Shot Quality Differential” gauge in the top right corner. You can select the point guards via the “Point Guard” drop down menu. Players with less than 100 minutes of shared playing time were removed.
3. Knickerbocker Career Shot Quality Differential (below the first two): The data here is cumulative, meaning shot quality differential totals over the course of a player’s career with the Knicks. The data is sorted from best to worse, darker colors representing players who played more career minutes. Same as the previous two, the Chris Smith’s and Jimmer Fredette’s of the world were removed.

PBP Stats has a lot more to offer, so absolutely go ahead and play around with the data. It’s quite intuitive and detailed. Explore the dashboard as much as you want too as there are some fun memory lane players like Jeremy Lin, Landry Fields, and Rasheed Wallace. And let’s hope that players like Porzingis can utilize his scoring gravity to help his teammates get better quality shots when he gets back.