Coaches know the importance of defense. Teams spend hours practicing defense. But is defense actually valued when it comes to allocating playing time? As I’ve written previously, the best way to measure an individual player’s impact on both offense and defense is to use offensive and defensive ratings. The difference between these two ratings is the player’s net rating. Offensive and defensive ratings found on sites like Sports-Reference are estimates. These estimates are less accurate than play-by-play data. Ratings from play-by-play data are the best measure of value as they account for all areas of the college game that affect the score, but are not tracked by an official stat such as charges, deflections, 50-50 balls won, communication, etc. Simply looking at a player’s box score stats doesn’t tell the whole story of the player’s true impact. Play-by-play based ratings are scarcely available but are tracked by DribbleHandoff. So how do these ratings relate to playing time? The scatterplot below shows a modest relationship between minutes per game and offensive rating, suggesting that a player’s offensive impact plays a role in their minutes per game allocation. However, looking at defensive data, the scatterplot below shows that a player’s defensive impact has little to no role in how many minutes per game he will play. Furthermore, an analysis of the top offensive and defensive players provides an alarming insight after adjusting each player’s ratings for team, conference, and opponents.
The top 20% of players in offensive rating have, on average, equal net ratings to the top 20% of players in defensive rating. Despite being equally efficient overall, the top offensive players receive an average of 4 more minutes per game. Coaches are undervaluing defensive abilities, and therefore, teams aren't playing their most efficient lineups. Coaches are just playing their most efficient offensive players, without consideration for their defensive value. Undervaluing defense affects all aspects of the game. Teams who properly value defense not only are rewarded by an increased scoring margin over the course of the season, but also find additional value from a player acquisition perspective. Valuing defense appropriately should completely alter recruiting and the transfer market. The relationship between playing time and these ratings suggests that coaches are either actively choosing to value offense over defense, or there is too much to consider defensively with very little data to assist in the judgment. Box scores are almost entirely offensively-based dashboards. They track points, shooting percentages, assists, turnovers, number of shots, and offensive rebounds while offering just defensive rebounds, steals, and blocks on the defensive end. Even looking at advanced metrics, the offense is still heavily favored in the number of metrics tracked. Most coaches use these box score stats and their advanced versions, such as assist percentage, to inform their lineup and playing time decisions. These metrics help serve as a somewhat unconscious proxy to offensive rating, while the lack of metrics defensively provides little insight into defensive rating. Offensive stats, whether using box score or advanced, explain about four times more of the change in offensive rating than the defensive metrics explain in defensive rating. If anything should be overvalued, it should be defense. Even though the top 20% of offensive players and top 20% of defensive players have equal adjusted net ratings, a top defensive player with limited offense is arguably more valuable than a top offensive player with limited defense. The data suggest that the offensively talented player would see more minutes, but strategically, these players can be exposed with different strategies. Coaches have the ability to find ways for a sub-par offensive player to provide value. Whereas on defense, depending on the scheme, the offense can target and attack a weaker defender every trip down the floor. For example, if a defense is switching, the offense can target the weaker defender by incorporating him in a ball screen or a dribble handoff. On the other side of the ball, weaker offensive players can be used as screeners and can crash the offensive boards. Regardless of the root cause, defense continues to be undervalued. The game is being managed at a sub-optimal level. This large-scale market inefficiency offers tremendous returns for teams willing to invest in analytically studying defensive possessions.
3 Comments
9/4/2019 10:43:39 pm
Basketball is not a sport that you can only play one way. Just like other sports, basketball requires a team to play both defense and offense at a high level. Sure, there are teams who have made it work by investing in one of these aspects, however, it rarely makes a championship team. If you were to ask me, I would say that a championship team is built both on defense and offense. It is important to invest in both of them.
Reply
11/21/2019 11:18:10 am
Most points win- Offense will always be more valued, because that is how the game is scored. You earn points when u score, you don’t get points for stopping a team from scoring. That’s not how the winner is determined.
Reply
Leave a Reply. |
Categories
All
Archives
January 2020
|