Sorry for the mini-hiatus everyone, I was busy wrestling my Final Exams into submission these last couple of weeks, and simply couldn't find the time to post anything. Things have just settled down, so I thought I'd do a little something on SportVu Rebounding.
Specifically for today, I thought I'd look at the correlations between Offensive Rebounding Rate (ORB%) and Defensive Rebounding Rate (DRB%), vs. Contested Rebound Percentage. I shouldn't have to do this, but the SportVu data at NBA.com doesn't have Offensive/Defensive Rebounding Splits, so I decided to run a couple regressions. Observe the key below, and note that correlation numbers are the R2 term *100.
ORB% = Offensive Rebounding Percentage
DRB% = Defensive Rebounding Percentage
CRB% = Contested Rebounding Percentage
ORB/DRB = Offensive/Defensive Rebounding Ratio
All = All Players
Rebounds = Top 100 Players in Total Rebounds
Minutes = Top 100 Players in Total Minutes
Specifically for today, I thought I'd look at the correlations between Offensive Rebounding Rate (ORB%) and Defensive Rebounding Rate (DRB%), vs. Contested Rebound Percentage. I shouldn't have to do this, but the SportVu data at NBA.com doesn't have Offensive/Defensive Rebounding Splits, so I decided to run a couple regressions. Observe the key below, and note that correlation numbers are the R2 term *100.
ORB% = Offensive Rebounding Percentage
DRB% = Defensive Rebounding Percentage
CRB% = Contested Rebounding Percentage
ORB/DRB = Offensive/Defensive Rebounding Ratio
All = All Players
Rebounds = Top 100 Players in Total Rebounds
Minutes = Top 100 Players in Total Minutes
- Predictably, we see the strongest Contested Rebounding correlations with Offensive Rebounding, and visa-versa with Defensive Rebounding. This seems to jive with many NBA metrics, which place a larger importance on Offensive Rebounding then Defensive.
- The correlations when including "All" players in the dataset seem relatively weak, although the ratios still seem alright. This will probably correct itself when we have a larger sample to work with.
- The "Minutes" group showed quite a strong DRB%:CRB% correlation. This could be due to more perimeter players making their way into this dataset (compared to the Big-Men dominated "Rebounds" category), who snag a greater proportion of Defensive Rebounds (many of them contested).
- Jordan Hill leads all players in Contested Rebounds per 48 minutes. (Minimum 12 games and 20 minutes per game to qualify).
- Thus far, about 33% of all Rebounds collected in this NBA season have been "contested". This is a remarkably low number, which we should keep an eye on.
- That's all for now. There are clearly many more layers that can be added to this study, such as factoring in position, height, consistency after changing teams, and so on. It just seems like a waste of time to tabulate all that data, when the season/sample is barely at an respectable level.
- The correlations when including "All" players in the dataset seem relatively weak, although the ratios still seem alright. This will probably correct itself when we have a larger sample to work with.
- The "Minutes" group showed quite a strong DRB%:CRB% correlation. This could be due to more perimeter players making their way into this dataset (compared to the Big-Men dominated "Rebounds" category), who snag a greater proportion of Defensive Rebounds (many of them contested).
- Jordan Hill leads all players in Contested Rebounds per 48 minutes. (Minimum 12 games and 20 minutes per game to qualify).
- Thus far, about 33% of all Rebounds collected in this NBA season have been "contested". This is a remarkably low number, which we should keep an eye on.
- That's all for now. There are clearly many more layers that can be added to this study, such as factoring in position, height, consistency after changing teams, and so on. It just seems like a waste of time to tabulate all that data, when the season/sample is barely at an respectable level.