Live Ball vs Dead Ball Turnovers
Last season, I added a new feature to the site which allowed for a deeper dive into NBA team per possession efficiency. Per possession efficiency has long been a staple of advanced metrics. It allows for a more precise measure of offensive and defensive skill. While traditionally a box score stat, I used the detailed play by play data to break out efficiency based on how each possession began. With rare exceptions, a possession can begin in one of the following ways:
Possessions off of dead ball turnovers are slightly less efficient than possessions off of made shots, and clearly less efficient than live ball turnover possessions. This is all intuitive, given that a made shot creates a quasi-dead ball situation due to the inbounding process.
For that reason, I will be making a change to my per possession statistics tool. Dead ball turnovers will now be grouped with "After Made Shot" possessions. I could create a separate category, but I'm running out of real estate on the page, and would prefer to keep things simple where I can. And the efficiency graph above supports grouping dead ball turnovers with made shots.
In case you are curious, here is a summary, by season, of the % of turnovers that are live ball.
Note the drop in percentage for the 2015-16 season. Does this mean teams are getting more careful with the ball? Are referees calling more offensive fouls? The actual answer is more mundane. For the 2015-16 season, the NBA has added an additional turnover category to their play by play data. Prior to this season, "bad pass" turnovers were a single category, and could include bad passes that go out of bounds (i.e. a dead ball turnover). There is now a separate category for out of bounds bad pass turnovers, and the 10% dip in live ball turnovers is almost wholly attributable to this new split.
- After an opponent made shot
- After a defensive rebound
- After a turnover
As one would expect, possessions after a turnover are the most efficient, due to their opportunities for fast breaks. However, several people pointed out that the turnover bucket includes both "live ball" turnovers (bad passes, steals, etc.) as well as "dead ball" turnovers (out of bounds, offensive fouls, etc.). If turnover efficiency is driven by fast break opportunities, then possessions off of dead ball turnovers should have a much lower efficiency than their live ball counterparts. Here are the results, aggregated over the past four full NBA seasons (2011-2014):
Possessions off of dead ball turnovers are slightly less efficient than possessions off of made shots, and clearly less efficient than live ball turnover possessions. This is all intuitive, given that a made shot creates a quasi-dead ball situation due to the inbounding process.
For that reason, I will be making a change to my per possession statistics tool. Dead ball turnovers will now be grouped with "After Made Shot" possessions. I could create a separate category, but I'm running out of real estate on the page, and would prefer to keep things simple where I can. And the efficiency graph above supports grouping dead ball turnovers with made shots.
In case you are curious, here is a summary, by season, of the % of turnovers that are live ball.
Leave a Comment