Project Defensive Score Sheet: Western Michigan

By Tommy D — There is a growing movement among college basketball analysts to try to tackle an issue which has traditionally been very difficult to quantify and synthesize:  defense.  Much as sabermetrics has influenced the way baseball is analyzed and changed the nature and types of statistics that many baseball observers believe are relevant, so too is the goal here in college basketball.  The idea is to use actual data to evaluate defensive performance, rather than relying solely on the eyeball test.

Have defensive statistics been kept for decades?  Of course.  But the reality is that defense involves much more than an individual player’s steals and blocked shots.  How often is a player beaten for a basket?  How many shots does he defend well and force a miss?  How many turnovers does he force, including taking charges?  How many defensive rebounds does he grab?  For what is the point of a good defensive play forcing a tough shot if someone on the defensive team doesn’t snare the rebound?  What about the overall performance of the team defensively?  How much impact does each player have on the team’s defensive performance?  Don’t all those questions sound more interesting than simply, “how many steals did he get?”

So the idea is to really assess what is happening on each defensive possession, and to give “credit” and assess “blame” for each play — missed shots, made shots, turnovers forced, missed and made free throws.  That takes time to review each game in detail, but the result is a much more complete and much more fulfilling understanding of a player, a team, and a basketball game.  All of the data for each player, and for the team, is synthesized into a “Defensive Rating” for each game.

One of the pioneers in the field has been Dean Oliver, who wrote a book called Basketball on Paper, in which he devised a number of formulas that aid in coming up with various measures of defensive efficiency for both teams and individual players.  Another has been a gentleman named David Hess, who writes a blog called The Audacity of Hoops: Thoughts on Recalculating an American Game and is a full-time contributor at TeamRankings.com

Others interested in the topic have begun to drill down into this area with emphasis on a particular team.  We here thought it was time use these methods to look at the Blue Devils and see what we could learn, especially given the concerns that many Duke fans have expressed about the defense being played by this year’s squad.

So what does this “charting” look like?  Check it out.  A chart of the first game I analyzed, which was Duke vs. Western Michigan, is below.  Then below the chart, I’m going to copy and paste David Hess’s explanation of the abbreviations used in the charting system that is becoming a standard in the field.  Then I’ll analyze what the chart is showing.  More Duke games to follow.

OK so here’s what a chart looks like:

Again, copying from David Hess’s blog, with a few tweaks, here is how you read this chart:

These numbers come from four sources.

Taken from the traditional box score:
• Min – Minutes played
• DREB – Defensive Rebounds

Tracked directly by the charter:
• FM – Forced field goal Miss – when a defender forces an offensive player to miss a shot from the field. This includes blocked shots.
• FTO – Forced TurnOver – when a defender forces an offensive player to turn the ball over. This includes steals. One thing to note here is that a player who draws an offensive foul is always credited with a FTO, even if it’s just a moving screen.
• FFTA – Forced missed Free Throw Attempt – missed foul shots resulting from a defender’s foul
• DFGM – allowed Defensive Field Goal Made – when a defender’s error or poor play leads to an offensive player scoring a field goal (intentional fouls at end of game excluded)
• DFTM – allowed Free Throw Made – made free throws resulting from a defender’s foul (intentional fouls at end of game excluded)

Calculated Tallies:
• Stops – the credit a defensive player gets for actions that contributed to ending an opponent possession. This isn’t as simple as adding FM + FTO + 0.4*FFTA, because the credit for a missed shot has to be shared with the defensive player who rebounds it. The formula is more complex than you might think, and includes a sliding weight for FM vs. DREB, based on how difficult those actions seem to be in each particular game. For full details, see Appendix 3 of Basketball On Paper.
• ScPos – Scoring Possessions allowed by a player. This is essentially just DFGM plus a FT-related factor. Again, see Basketball On Paper for the full formula.
• DPoss – [Stops + ScPos] – total Defensive Possessions that were credited to (or blamed on) a player.

Calculated Metrics:
• Stop% – Stop Percentage — [Stops/DPoss] – the fraction of an individual player’s credited defensive possessions that ended with zero points. Essentially the inverse of offensive Floor%.
• %DPoss – Defensive Possession Percentage — the percentage of team defensive possessions faced by an individual defender. Analogous to %Poss on offense.
• DRtg — Defensive Rating – individual Defensive Rating. Gives a player credit for stops and scoring possessions he was directly involved in, then assumes a nebulous team-average performance in the other possessions. This is the analog of offensive rating.  The lower a player’s DRtg, the better he played defensively.

So what does the Western Michigan chart tell us?  As will become a theme for this year’s Devils, Miles Plumlee was excellent.  His DRtg was a full 5 points below the overall team’s rating of 93.5.  That’s a ton.  Why were his numbers so good?  Because he forced 10.5 missed shots from WMU, plus 2 free throws and a turnover in only 25 minutes of action.  And he grabbed 10 — 10! defensive boards in just those 25 minutes.  All of that more than offset the four hoops he gave up.  The team got stops on 65% of the possessions that Miles was on the floor.  That is excellent.  Not that this charting shows it, but another big reason why Miles graded out so well is the excellent help defense he provided all night long.

On the perimeter, Seth Curry had the best numbers.  He pitched a shutout personally, giving up no points.  And a 100% stop percentage.  His DRtg was also almost 5 points below the team’s overall.  Seth should lock up a DVD of this game in his safe deposit box, because numbers like these, even against an opponent the level of Western Michigan, are rare indeed. Now, Western MIchigan’s guards rarely challenged him, as a close view of the game disclosed that Seth wasn’t actually involved in the outcome of all that many plays, but nevertheless, he played 24 minutes and was never beaten.  Impressive.

This was Duke’s first game after the holiday break, and it was a 40 point win, so there’s only so much to be taken from it.  If the idea was to brush some rust off, get everybody involved, and work on a few new wrinkles, it was mission accomplished.  Western didn’t have the talent to challenge our defense much, but it was good to see that we rarely let up either physically or mentally, and just took care of business at both ends.

You can see some other efforts in this project done by Chris Mackinder for Michigan State, Adrian Atkinson for UNC, and David Hess for many teams.  Tommy D already has charts done for a number of other Duke games and we’ll be rolling those out over the next few weeks.