Defensive Score Sheet: Kentucky Wildcats at South Carolina

If you haven't already, make sure you read Glenn's piece below where he discusses the outstanding defensive effort the Cats put on the floor against South Carolina. Also scroll down for the Big Blue Links.

One of the great voids in basketball is the lack of accounting for the defensive contributions by players during a game. Box scores only keep track of three defensive plays: blocks, steals, and defensive rebounds. It doesn't take long to see that there is a lot more to good defense than those three things, yet because they are the only defensive measures available they tend to dominate the conversation when talking about great defenders. To put it another way, we all know that Liggins is a fantastic defender, but if you haven't watched many UK games you probably are unaware of just how good he is. On the other hand, you don't have to watch many games to know that Jared Sullinger is a really good offensive player or that Doron Lamb is an outstanding shooter. Watching the games helps you appreciate how good those guys are, but it's also pretty easy to see based on their offensive numbers.

Enter Dean Oliver, author of the book Basketball on Paper. Oliver now works for the Denver Nuggets, but several years ago he was a consultant for the Charlotte Sting of the WNBA and as part of that job organized an effort to collect better information about the defensive side of the ball during league games. Oliver did this by enlisting the aid of a host of volunteers who would attend games and track defensive plays using a special score sheet. The effort was aptly named Project Defensive Score Sheet and Oliver wrote about it in the afore mentioned book.

When I read about this last fall I was very intrigued. I've been working to gain a better appreciation for defense for several years now and this concept appealed to me as a way to continue that effort. I thought about making these sheets for UK games this season but didn't follow through for a variety of reasons (general laziness mostly).  As it turns out, David Hess had the same idea except that he actually went out and did it and has been posting his score sheets on his Audacity of Hoops blog.

That was enough to get me off my lazy behind so I tracked the defensive possessions for UK's game against South Carolina.  Follow the jump to see the results as well as a bit more detail about what actually gets tracked.

From the official box score:

• minutes played (Min)
  
• defensive rebounds (Def Rb)
  
• Blocks (Blks)
  
• Steals (Stls)
  

What I track during the game for Player X:

• forced missed shots (F. Miss): field goal attempts that are missed when Player X defends the shot
  
• defended made shots (Def FGM): field goal attempts that are made when Player X defends the shot
  
• missed Free Throws (FT Miss): free throws that are missed when Player X commits a foul
  
• made Free Throws (Def FTM): free throws that are made when Player X commits a foul
  
• forced turnovers (F.TO): turnovers forced by the defense of Player X (not just steals)
  
• deflections (Defl.): basically anytime Player X does something to deflect the ball, but doesn't result in a turnover or missed shot.  I would also add jump balls that don't change the possession here.
  

Deflections are not in Oliver's original score sheet, but I added them because they are something I've heard/read that many coaches like to chart during games as a way of gauging how active their players are on defense and hey- I'm keeping track of this kind of thing already so why not?

What gets calculated for each individual player:

• Stops: The number of times Player X does something to end the opponent's possession and get the ball back for his team without giving up any points (grab a defensive rebound, force a turnover, etc)
  
• Stop Percentage (Stop%): The percentage of personal possessions used that Player X forces a stop (as opposed to the opponent scoring)
  
• Percentage of Team Defensive Possessions Used (%TmDPoss): The percentage of a team's defensive possessions that Player X uses.  You can think of this as the percentage of time that Player X is the one defending the ball (that's not completely accurate but will give you the general idea).
  
• Individual Defensive Rating (Def Rtg): How efficient the opposing offense was at scoring while being defended by Player X.  Lower is better, but keep in mind that this in particular is judgment-neutral.  That is, just because a player has a high Defensive Rating it doesn't necessarily mean he played poor defense - it could mean that he was guarding a very good offensive player.

I'm going to refer you to the book for the specifics in calculating these things because otherwise this is going to be about four times as long.

Basically how this works is that when a shot is attempted, I mark down which player - if any - is defending it and the appropriate category gets marked for the result of the play.  If two or more defenders are involved on a play then credit gets divided evenly between them.  For example, if Knight and Lamb put a double team on the ball handler and he travels, Knight and Lamb would each get a 0.5 F.TO.  Finally, if an offensive player isn't guarded then the result gets marked down in a "Team" category.  There are some other ways that plays can get listed in the Team category and I'll discuss one of those after the charts.

Obviously a lot of this is subjective and depends on the attentiveness and experience of the observer, but I think overall most people will arrive at largely the same results.  Enough with the talk though - here are the defensive score sheets for UK's victory over South Carolina.

I've included the official box score totals so you can compare what I came up with against the official record.  I highlighted a few areas that caught my eye, but since this is already getting long I'll only make a couple of brief comments:

• You'll notice that Jones, Knight, and Liggins have a ".03" in their forced TO count.  This comes from a play in the second half when Knight and Liggins double teamed the SC ball handler (I don't remember who it was) who picked up his dribble.  The play occurred near the baseline and the SC guy was under a lot of pressure.  When he tried to pass out of the double team to get the ball back out, Jones was there to pick it off.  In the box score, Jones gets the steal, but Liggins and Knight were critically important for making that play work, so they split the credit and each got 0.33 of a Forced Turnover for that play.
• Slawson's block came in the first half when he blocked Miller's shot attempt off Darius for a turnover.  On that play, Muldrow came over and got between Miller and the basket which gave Slawson the chance for the block, so Muldrow and Slawson split the credit for the miss and Slawson got full credit for the turnover.  I think this play is indicative of the value of these score sheets as otherwise Muldrow's contribution to that play would exist only in the memory of the people who actually noticed it.  This is also why Slawson has a block but only 0.5 of a Forced Miss.
• There was a play in the second half where Jones bit on a Muldrow ball fake and Muldrow drove past him and scored over Harrellson.  I split the Def FGM between Josh and Terrence because Josh wasn't able to stop the score but it was Terrence's mistake that put him (Josh) in that position.
• There are 5 free throw attempts listed for SC's Team category.  These were UK's free throws in the last 41 seconds of the game where it was clear that the fouls were being committed by the "closest guy available" and were not a reflection of the particular defense played by the individual but rather by that particular team strategy.
• You'll see there are some discrepancies between my counts and the official box score.  Frankly, I'm surprised there weren't more given that this was my first try at this sort of thing.  I don't know exactly where in my sheets they occurred so I just left them in, but if you compare the calculations from my tracked totals to those from the official box score you'll see that they didn't cause a huge change.

Whew!  That's a lot of words.  I plan to continue doing these for the rest of the season as my television viewing allows.  I have some ideas for how I can make them more accurate as well as more detailed (such as dividing shot attempts into 2's and 3's).  If anyone out there is interested in having a go at this I'll happily email you the sheets I used.

Update (2-07-2011): I discovered a mistake in the calculations portion of the tables.  Posted below are the correct tables.