When I have time, I like to watch Iowa MBB games with an Excel spreadsheet open and track each possession to see how we do when certain players are in the game, when we try zone vs. man, when our offense is in transition vs. the half court. I'll try to write some posts and share my findings with you as the season goes along. Today, I will show you data regarding our main rotation players and how our team performs when they are on the floor. My data is from the Creighton, Arkansas State, Oklahoma, Seton Hall and North Florida games, roughly 355 possessions altogether (I often don't track possessions at the very end of the game when the scrubs are in or when everyone is going all out for steals/fouling).
Some definitions first. OPP = Points per possession that we score when a player is in the game. DPP = Points per possession that we allow when a player is in the game. O Diff = How many more points do we score (per possession) when a player is in the game vs when that player is out of the game (positive is good). D Diff = How many more points do we allow when a player is in the game vs when the player is out of the game (negative is good).
Total OPP = 1.17, DPP = 1.08
Bowen OPP = 1.19, DPP = 1.16, O Diff = 0.03, D Diff = 0.18
Perkins OPP = 1.22, DPP = 1.07, O Diff = 0.15, D Diff = -0.02
Payton Sandfort OPP = 1.16, DPP = 1.11, O Diff = -0.03, D Diff = 0.09
Pat McCaffery OPP = 1.22, DPP = 1.08, O Diff = 0.15, D Diff = 0.02
Krikke OPP = 1.21, DPP = 1.09, O Diff = 0.17, D Diff = 0.07
Dix OPP = 1.18, DPP = 1.07, O Diff = 0.01, D Diff = -0.02
Pryce Sandfort OPP = 1.07, DPP = 1.04, O Diff = -0.14, D Diff = -0.05
Freeman OPP = 1.16, DPP = 0.91, O Diff = -0.02, D Diff = -0.28
Harding OPP = 1.10, DPP = 1.12, O Diff = -0.10, D Diff = 0.06
Dembele OPP = 1.02, DPP = 1.04, O Diff = -0.23, D Diff = -0.06
I went ahead and put my main point in bold. We allow a whopping 0.28 fewer points per possession when Owen Freeman is in the game. For a frame of reference, a defensive improvement of 0.28 DPP would result in 21 fewer points allowed over a 75 possession game! The offense stays more or less the same when Freeman is in but the defense seems to improve dramatically. This passes the eye test, Freeman is by far our best shot blocker and also gets a surprisingly high number of steals. He is quick for a guy his size.
Naturally, there are caveats to my analysis. My samples size still isn't huge. I don't have a good way to control for the specific lineups that our opponents have out there (i.e. maybe our backups are playing against their backups etc.) If we tend to use certain combinations of players together it can be hard to disentangle the impact of individual players. For example, you might also notice that the defense appears to be way worse when Bowen is in the game. I think that is an artifact of Bowen and Freeman having only 8 total possessions together across my five game sample (we have allowed only 4 points in those 8 possessions). So it's not so much the presence of Bowen as much as it is the absence of Freeman, IMO, that makes Bowen's D Diff look bad.
Freeman is currently playing slightly less than 15 minutes per game. That needs to increase. All three of our starting front court players average 24 mpg or higher. Let's chip a couple minutes each from Ben, Pat and Payton and get Freeman 20+ minutes per game. In case folks are wondering if playing Krikke and Freeman simultaneously will cause problems, I should point out that they have about 60 possessions together. So far, the Freeman/Krikke combination is scoring 1.31 points per possession while only allowing 0.89 points per possession over those 60 possessions.
This seems like the quickest fix for our (very) leaky defense.
Some definitions first. OPP = Points per possession that we score when a player is in the game. DPP = Points per possession that we allow when a player is in the game. O Diff = How many more points do we score (per possession) when a player is in the game vs when that player is out of the game (positive is good). D Diff = How many more points do we allow when a player is in the game vs when the player is out of the game (negative is good).
Total OPP = 1.17, DPP = 1.08
Bowen OPP = 1.19, DPP = 1.16, O Diff = 0.03, D Diff = 0.18
Perkins OPP = 1.22, DPP = 1.07, O Diff = 0.15, D Diff = -0.02
Payton Sandfort OPP = 1.16, DPP = 1.11, O Diff = -0.03, D Diff = 0.09
Pat McCaffery OPP = 1.22, DPP = 1.08, O Diff = 0.15, D Diff = 0.02
Krikke OPP = 1.21, DPP = 1.09, O Diff = 0.17, D Diff = 0.07
Dix OPP = 1.18, DPP = 1.07, O Diff = 0.01, D Diff = -0.02
Pryce Sandfort OPP = 1.07, DPP = 1.04, O Diff = -0.14, D Diff = -0.05
Freeman OPP = 1.16, DPP = 0.91, O Diff = -0.02, D Diff = -0.28
Harding OPP = 1.10, DPP = 1.12, O Diff = -0.10, D Diff = 0.06
Dembele OPP = 1.02, DPP = 1.04, O Diff = -0.23, D Diff = -0.06
I went ahead and put my main point in bold. We allow a whopping 0.28 fewer points per possession when Owen Freeman is in the game. For a frame of reference, a defensive improvement of 0.28 DPP would result in 21 fewer points allowed over a 75 possession game! The offense stays more or less the same when Freeman is in but the defense seems to improve dramatically. This passes the eye test, Freeman is by far our best shot blocker and also gets a surprisingly high number of steals. He is quick for a guy his size.
Naturally, there are caveats to my analysis. My samples size still isn't huge. I don't have a good way to control for the specific lineups that our opponents have out there (i.e. maybe our backups are playing against their backups etc.) If we tend to use certain combinations of players together it can be hard to disentangle the impact of individual players. For example, you might also notice that the defense appears to be way worse when Bowen is in the game. I think that is an artifact of Bowen and Freeman having only 8 total possessions together across my five game sample (we have allowed only 4 points in those 8 possessions). So it's not so much the presence of Bowen as much as it is the absence of Freeman, IMO, that makes Bowen's D Diff look bad.
Freeman is currently playing slightly less than 15 minutes per game. That needs to increase. All three of our starting front court players average 24 mpg or higher. Let's chip a couple minutes each from Ben, Pat and Payton and get Freeman 20+ minutes per game. In case folks are wondering if playing Krikke and Freeman simultaneously will cause problems, I should point out that they have about 60 possessions together. So far, the Freeman/Krikke combination is scoring 1.31 points per possession while only allowing 0.89 points per possession over those 60 possessions.
This seems like the quickest fix for our (very) leaky defense.