Our new stat tools

Welcome to the new Thinking Basketball dot net! In addition to a facelift, we’ve also made major upgrades to our stats tools that reflect how we use data internally to research trends around the league for podcasts and videos.

During the season, our player stats and team stats update daily, and you can browse under “metrics” for our historical databases as well. In the screen shot below, I’m currently in player stats with visualizations open (“hide visualization” will close the graph). I’m currently hovering over Joel Embiid, so he highlights in green on the graph:

As you’ll below, we currently have player stats that include our in-house metrics for scoring, playmaking, and overall effectiveness (our Box Plus-Minus, or BPM model), along with popular public impact metrics, offensive and defensive stats, stats for effectiveness and frequency around the rim, midrange and outside shooting and team-level stats.

1. Player cards

My favorite feature is probably the player summary card, where you can see everything at once compared to the league. Just click on a player’s name, (or on a player’s dot on the graph), and it’ll bring up a summary card with every stat we use and the player’s percentiles relative to the rest of the league, organized by category:

2. Team search

Underneath filters, you can view a specific team. But one of my favorite things to do is compare all the players on a team to the league, and for that you can use the search box in the upper right. Just type in the team’s shorthand name (eg “BOS” for the Celtics”), and all the players on the Celtics will be highlighted. (Note, this will also scoop up players who aren’t on the team who have “BOS” in their name.) In this case, I’ve used the “select stats” dropdown in the upper right to make team offense the X-axis and a player’s on-court offensive rating the Y-axis. Then I used a comma to add another search term (“LAL”)  and voila! It’s easy to see the Laker with the best on-court rating is below the Celtic with the worst on-court rating. The larger dots mean more minutes played:

3. Percentile view

You’ll see a button that says “hide visualization” (or show), which allows you to hide the graph and look at a larger table view. When you’re looking at the table, you can also flip all of the numbers to league-wide percentiles just by clicking on any statistic in the table. Sometimes I browse around in this view when trying to contextualize a bunch of stats for a team or player:

4. A team’s Last Month

All stats for players and teams are organized by category, and I’ll often select (and de-select) entire categories to focus on skills like outside shooting and rim finishing. One of my favorite areas to keep an eye on is a team’s previous month, where it’s easy to see who is playing well, and really easy to see who is hot (or cold) relative to the rest of their season. Here, the Grizzlies, Nets, and especially Knicks have had a great month. Milwaukee and Phoenix have struggled:

5. Player trends

Finally, I like to use our free trends tools to spot changes in style of play, or to identify shooting slumps and hot streaks. Here is Steph Curry since 2016, where it’s easy to quickly see he’s using the midrange more — something I thought was key heading into last year’s playoffs — and you can see his shooting last year looked more like a (relatively) normal slump versus a large decline in form:

Here’s a teaser video of the new site when we launched: