Eight Conference Games In: what drives Vandy's losses

Before the season really got under way, I posted that we'd only go as far as our defensive efficiency would take us.  I held up last years performance in the SEC schedule and what was a relatively poor defense (106 points per 100 possessions in SEC play). 

Guess what?  Our defensive efficiency in conference is allowing 104.6 points per game (using stats from Ken Pomeroy)!  Improvement! But still pretty bad.  So that's it.  That's why we're losing, right?  Not so fast...

All stats from Ken Pomeroy here.  And in the rest of this post.  Go to his website. http://kenpom.com

In conference, we have lost 2 of our best four overall defensive performances (UT/UF on the road... what a waste).   So, we know that half our in-conference losses, and a third of our total losses came with reasonable defensive performances.  So it isn't just "overall defensive efficiency."

More sleuthing post jump.

It could be playing away from Memorial.  Our record is 2-1 at neutral sites (with wins over the 11th (UNC) and 46th (Neb) best teams and a loss to the 19th (WVU) best), 2-4on the road (wins over 196th (MTSU) and 165th (MS St) best; losses to the 23rd (Mizzou), 106th (SCar), 45th (UT) and 31st (UF) best), and a record at home of 12-1 (best wins are 26th Marquette, 34th St Marys, 35th Belmont, and 59th UGa).

As of right now, looking at the schedule we have faced, I think our road record has more to do with the quality of teams we played on the road, versus at home.  We have had an ambitious road schedule thus far; we only have two difficult road trips left (UGa and UK).  In our road losses, our defensive efficiency ranks from as good as 93.6 at UT to an abysmal 116.4 at SCar (both games are also our extremes for offensive efficiency, with 89.4 and 105.2 respectively).  We may play worse on the road, but with two OT losses and two 3 point losses, we aren't playing a ton worse on the road.  We have good defensive performances (UT, UF) and we have good offensive performances (105 against SCar, 103 against Mizzou) in our losses.  And it doesn't explain Arkansas.

So, to identify what ails us, I'll move to KenPom's four factors (eFG%, TO%, OR%, and FTR).  And this is, from what I can tell, where the problem lies.

If you notice KenPom's evaluation of our defense, you'll see that we excel at holding opponents to a poor eFG%, where we are 15th in the nation (primarily by smothering 3pt shooters; we're 11th in the nation at 3pt fg%).  Second, we don't foul opposing teams; we have the 19th lowest free throw rate for opposing teams.  To do this, we sacrifice crashing the defensive glass somewhat (we're 133rd in preventing offensive rebounds) and we are not a team that gambles for steals (we are 292nd in turnover percentage).  These are the central tenets of our defense this year.

HOWEVER: for a team that doesn't bank on forcing turnovers to win, we sure do suffer when we produce them at less than a 16.9% clip (ie. about 17 to's per 100 possessions... the D-1average is 20.5).  To wit:

This shows clearly that there is reason for concern whenever we aren't hitting at least 17% (ok, 16.9%) in the opp TO rate. No loss is an exception to this rule.  If we won't force turnovers at 17 to 100 possessions, we will very likely lose.  It wasn't until the latter quarter of the game against Miss State that we pulled out the press, and that move may have been more necessary than we thought. 

Now, why didn't we lose to Grambling and Ole Miss then? I'm glad you asked.  We held them to offensive rebounding rates of 20.6% and 22.2% respectively, our 2nd and 5th best efforts on the defensive glass all year.  We may not have force them to give us the ball, but we didn't allow second chances, either.  They shot, and the ball was ours one way or the other.

This brings us to our second factor: 

These are our 10 worst games at keeping the defensive glass clean. You'll notice, it doesn't appear quite as strong a correlation (rather than 6 of the bottom 8, it's 6 of the bottom 10), but we do appear to suffer when we let teams get on the offensive boards. In the four instances in which we won, you can see the areas of the defense that compensated: Davidson connected on less than a third of their shots, and took almost no free throws (a perfect game by our defensive standards),  UNC gave us the ball nearly 1 out of every 3 possessions,  SCar round 2 turned the ball over almost twice as much as they did in the first game with them, and UGa shot horribly against us.  I think it's important to note that we seem to be somewhat better at masking a bad game on the defensive glass than we are at masking a bad game (by our low standards) when it comes to forcing turnovers.

So, after all of this, what have we learned?  It's not just defensive efficiency overall, or playing on the road that is killing us.  The simple fact is that when we don't force enough turnovers, we are much more likely to lose; this is particularly true when coupled with a bad job on our defensive glass.  Unfortunately, our defense is geared towards producing bad shots and avoiding fouls, NOT producing turnovers.  Perhaps we will see more of the press, or some jumping of the passing lanes... I'm not a basketball coach, so I don't know.  It may not be addressed at all.  But, right now, if you want to tell if we'll win or lose a game, check the turnovers.  Hope for something in the range of 12+ (depending on pace of the game).  It seems to have the strongest correlation at the moment.

Note: There is also some correlation with how well we shoot: four of the five worst eFG% performances are losses.  Oddly, so are the 6th and 11th best shooting performances. There is little correlation with how much we turn it over, which shocked me... of the seven best Vandy TO% performances, three are losses.