Welcome to UofL's 2018-19 season with Hoops Insight!
Coming this season from Hoops Insight: Stats Profile dashboards!
I have spent some time developing a one-page stats profile dashboard that I can share with my subscribers in order to summarize some of the key stats that I like to look at when analyzing Louisville's performances. This season, I'll be sharing these regularly with my subscribers via email in order to provide context around my analysis and insights about Louisville basketball.
I try to stay away from just dumping stats and numbers out, and expecting my subscribers to be able to figure out what is meaningful. I prefer to discuss what I'm seeing in the data, and trace the impacts in the data back to things that are happening on the court. I view data as the results of choices and actions, and I'm less interested in the data itself than what it's telling me about what is happening, what's going well, and what needs to change. To that end, I've avoided just posting a lot of data, but I've decided to develop this stats profile dashboard so that my subscribers get more context than just my interpretation of the data.
I'm planning on regularly including these stats profiles with my emails. The profile may cover one game, a selection of games, or the full season. In the corresponding email, I'll walk the reader through what I'm seeing in the profile that is interesting. I'll supplement this with other newsletters about more specific topics, but I want my subscribers to get into a rhythm of seeing a consistent type of analysis using the stats profile. For you, the subscriber, this will probably mean a bit more frequent emails during the season that are more focused, mixed in with occasional editions of the newsletter which go more indepth on specific topics.
How do I read the stats profile?
As a dry run, here's an example using Louisville's stats profile for the 2017-18 season. (click link to view).
The stats profile is organized into tables. At the top of the stats profile will always be a list of the included games; this version covers UofL's full 2017-18 season.
Table A lists key team stats for the selected game, and a comparison of that performance to the full season. Since this profile is for the full season, the comparison isn't useful, but for subsets of games it may reveal important trends.
Table B is an infographic showing the types of shots UofL gets and allows its opponents to get, categorized by Putbacks, 1-10 seconds into shot clock, 11-20 seconds into shot clock, and 21+ seconds into shot clock. For each category, the infographic displays what percentage that category is of the total shots, and the effective field goal percentage. Each category is color-coded to display the eFG%; green is better, red is worse. In this example, you can see that UofL took many more shots in the first 10 seconds of the shot clock than opponents, and excelled at limiting opponents' shooting late int he shot clock. UofL's tendency to get quicker shots on offense was a notable change last year under David Padgett, but the Cards have been very strong for years at forcing opponents into bad shots late in the shot clock.
Table C shows how UofL is performing in key stats as the number of opponent starters varies. UofL's table from last year is interesting, in that UofL doesn't have any stats in which they clearly improved as opponents went deeper into their bench. This indicates that UofL tended to go to their own bench to match opponents, as the Cards went 10 deep pretty consistently last season. UofL actually performed slightly better (per adjusted +/- per possessions) against 5 starters last season than bench lineups.
Table D shows the 5 lineups who played the most possessions. This will update to show the 5 most played lineups for whichever games are included. Looking at this example, you can see that UofL leaned very heavily on their usual starters, who played 389 possessions. That's more than twice as many as the 2nd most played lineup. The best lineup among this group featured Ray Spalding as the lone big man, and had an adjusted scoring margin of +0.38 points per possession. This lineup outshot opponents comfortable, rebounded extremely well, and drew free throws at an above-average rate. No other frequent UofL lineup was so well-rounded.
Table E is a stat I created which tracks what I call "same four" impact. This is a measure of how lineups perform with a given player, compared to how lineups with the same four teammates perform with anyone else on the team substituted for that key player. This is a better way of approximating player impact than simple on/off court stats, since I'm controlling for changes in the rest of the lineup. For ease of reading, any stat where the player has a positive impact is shaded green, and those where the player has a negative impact are shaded red. For these tables, I'm focusing on the impact of two of UofL's key returnees this season: VJ King and Darius Perry. This table is not meant to provide a simple evaluation of how good a player is, but rather it highlights how a player contributes compared to the other options the team has. So, players who lack a viable backup are going to look great here. King struggled somewhat last season, and this table indicates that UofL was usually better off when he replaced by other options such as Dwayne Sutton, Darius Perry, and Ryan McMahon. Perry's table shows that he had a largely positive impact when he replaced other players in the lineup. In particular, UofL's turnover rate increased when he came in to pressure the ball, although this aggressive defense did result in more free throws for opponents. Perry posted a higher steal rate than starting guards Quentin Snider or VJ King last season, but he also fouled more frequently than any other Cardinal.
Lastly, Table F summarizes UofL's best and worst performances in various stat categories. This data is less insightful, but is interesting to track during the season. As you can see, the range of performances during a season is huge. It's important to keep that in perspective during the season there will be some very dramatics highs and lows.
Whenever I post a stats profile, I will strive to tell the reader what the data says, and what that means. If you see anything interesting you'd like to point out, feel free to reach out to sean@hoopsinsight.com or on Twitter @hoopsinsights. Also, if you have any questions during the season, feel free to reach out as well. I hope you enjoy subscribing this season, and I will try to make every email worth reading.
Thanks for reading my newsletter. If you have any questions, want to argue a point, or have some feedback, feel free to reach out via email at sean@hoopsinsight.com, or on Twitter @hoopsinsights. If you liked this, let me know as well, and tell your friends to subscribe at www.hoopsinsight.com.
You can forward this to others, but please ask them to subscribe as well so I can keep track of who's enjoying my insight and analysis