Defense!

I’ve been ranting about offensive statistics, so now it’s time to show some love for the defense.  On your standard box score you can see steals and blocks, which are pretty much the only basic statistics that give any merit to good defense.  It’s cool to see which players are the leaders of those simple figures, but let’s take it a step further.  To start, let’s look at the WBBL teams’ overall defensive performance.  The table below ranks the teams defense based on how many points the team allows the opponent to score.  Nottingham leads the pack with only giving up an average of 55.7 points a game.  I included several more columns that show some other noteworthy numbers, too.
team overall defense
Poss / Game = (Total Possessions for opponent) / (Games Played)
PPP = (Points the opponent scores per game) / (Total possessions for opponent per game)
FG% = field goal percentage
%TO = (Total Turnovers) / (Total Possessions) = percentage of time the opponent turnovers the ball
%FT = (Total Possessions with FTA) / (Total Possessions) = percentage of time opponent is awarded free-throws
%SF = (Total Shooting Fouls Drawn) / (Total Possessions) = percentage of time opponent is fouled in act of shooting
%Score = (Total Possessions with  1 point) / (Total Possessions) = percentage of time opponent scores at least 1 point

Now I wanted to find out which players on a particular team were great defenders beyond just looking at steals and blocks.  It’s those players who’s impact in the game may not be seen on the stat sheet.  My attempt at quantifying these “not so quantifiable” impact players was to compare a player’s minutes played per game with their opposition’s offensive efficiency.  By doing this we should be able to see which players on the court may have a positive or negative effect on their opponents performance.

For example, let’s says a player plays 20-25 minutes and their opponents perform quite poorly in those particular games.  Now lets say that same player plays 10-15 minutes in other games and in those games the opponent performs better.  If this were the case, then it can be safe to say that this player had a positive impact in the game from a defensive standpoint.  In short, this is what I’m saying:  The more minutes that player plays, the worse their opponent performs.  The less minutes that player plays, the corr min pppbetter the opponent performs.

What I did now was used Nottingham Wildcats data and compared the correlation outcomes.  For sake of these numbers being used to this team’s advantage, I’m not posting the actual names of the players.  Player #1 on this team can be categorized as the best defensive player based on this analysis.  Why?  Because there is a moderately strong negative relationship (R=-0.58) shown between that player’s minutes played and their opponent’s offensive efficiency (PPP).   This means the more minutes this player plays, the worse their opponent performs.  Player #9 is the polar opposite.  Player #9 may be seen as the worst defender.  The more minutes this player plays, the better their opponent performs.  Those players with values resulting very close to zero just means that theres no relationship between the minutes that player plays and how well their opponent performed offensively.  As a coach, how can this analysis be useful?  Well, what you can do is decide which lineup gives you a greater chance to disrupt your opponents.  You can then look at these correlations and as well as correlations between minutes played and your own team’s offensive efficiency.  A player may have a really good impact on their opponent’s efficiency, but they be a liability on the offensive end.

It’s important to note that this correlation by no means is the best way to quantify defensive performances.  It’s merely a single attempt to analyze and compare individual defensive impact on games.  There are so many other factors that can impact your opponent’s performance more than just how many minutes certain players play.  I just shared this way because I found it rather shocking which players had higher correlations than others.  What would be even more cool though would to be able to compare certain combinations of players on the court and how efficient those players perform together.

Until next time,

“The idea is not to block every shot. The idea is to make your opponent believe that you might block every shot.” – Bill Russell

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s