At the Quarter Mark: Some Defensive Charting Stats and Observations

We’re 10 games into a season that optimistic Tar Heel fans hope will go the full 40 (31 regular season games + 3 ACCT + 6 NCAAT). Through the season’s opening quarter, Carolina’s adjusted defensive efficiency places 11th in the country– a slight drop from last year’s final ranking of 6th. This post will try to shine some light on which UNC defenders are already in mid-season form, and which ones are still trying to shake off that early-season rust.

Kendall Marshall

2011 (all)
2012 (1st 10 games)
FG% All 37.6 40.0
3Pt% All. 26.5 35.3
FTA Rate 18.8 13.3
TS% All. 47.0 51.3
Pts. All. / 40 10.7 14.6
Deflections / 40 4.17 6.13
Forced TOs /40 2.56 2.30
Off. Fouls Drawn / 40 0.22 0.00
Denies / 40 0.91 0.77
DR% 7.6 7.5
Stop % 58.9 53.8
%Possessions 15.5 17.5
Def. On-Court/Off-Court +2.1 -8.3

Marshall, now firmly entrenched as a starter, is being game-planned against and attacked like one. Teams are challenging him more on the defensive end, hoping to exploit his relative lack of lateral quickness. After being involved (from a defensive charting perspective) in 15.5% of defensive possessions while on the court last season, that number has jumped to 17.5% this year. That, in conjunction with a drop in Stop% from 58.9 to 53.8, has resulted in Marshall allowing 14.6 points / 40– up from 10.7 as a freshman. Part of the reason for Marshall’s lower Stop% might involve some tactical decisions by Roy Williams. Marshall saw significant defensive minutes against both Jordan Taylor and Casper Ware– the types of assignments that might be increasingly handled by Strickland as the games get more and more important.

Marshall’s decline in Stop% has corresponded to a reduction in on-court/off-court defensive efficiency. This season, the Heels have a defensive efficiency of 94.2 with Marshall on the floor. It improves to 85.9 with Marshall on the bench. While he can be exposed as an on-ball defender, Marshall has excelled as a help defender this season. He ranks second on the team in deflections / 40, with many of those coming on help-side reach-ins. Marshall’s quick hands and keen instincts help to overcome some of his other physical shortcomings.

Dexter Strickland

2011 (all) 2012 (1st 10 games)
FG% All 46.2 46.2
3Pt% All. 32.8 25.5
FTA Rate 24.0 20.5
TS% All. 55.3 54.9
Pts. All. / 40 11.1 10.5
Deflections / 40 6.16 6.57
Forced TOs /40 3.54 3.43
Off. Fouls Drawn / 40 0.16 0.45
Denies / 40 2.05 2.54
DR% 8.2 6.8
Stop % 58.5 57.9
%Possessions 16.5 15.4
Def. On-Court/Off-Court -2.5 -2.4

Through 10 games, Strickland, Carolina’s best on-ball defender, has eerily similar defensive charting numbers to what he posted last season. After a tough defensive showing against Long Beach State (in which he was repeatedly beaten on back cuts), Strickland’s Stop% fell from the low 60s (where it had resided for most of the season) down to the high 50s. He is at his best as a shot preventer/ball denier, and has made high-usage guards work hard to find their scoring opportunities. Strickland is also quite disruptive, leading the team in deflections / 40 for the third straight season.

He’s good defensively right now, but certainly not elite (like Manuel in ’05, for example). Still, he does play hard consistently and, in a must-stop possession, he’s the Heel you’d want defending the opponent’s top perimeter scorer.

Harrison Barnes

2011 (all) 2012 (1st 10 games)
FG% All 40.5 51.2
3Pt% All. 38.2 49.0
FTA Rate 31.9 22.1
TS% All. 51.8 60.9
Pts. All. / 40 10.8 12.5
Deflections / 40 4.27 4.53
Forced TOs /40 2.37 2.19
Off. Fouls Drawn / 40 0.30 0.44
Denies / 40 0.89 1.75
DR% 12.6 7.4
Stop % 59.5 50.6
%Possessions 15.8 14.7
Def. On-Court/Off-Court +1.5 +4.0

Of the six returning rotation players, two have made big defensive charting strides (Bullock and Henson), two have maintained their 2011 level (Zeller and Strickland), and two have suffered precipitous declines through 10 games (Marshall and Barnes). After ending his freshman season with a Stop% of 59.5, Barnes is currently at 50.6 in that category. His 3-point defense– a weakness last season– has continued to be poor as a sophomore. Barnes’s defensive rebounding rate has also plummeted from 12.6% to 7.4%. Continuing a trend from the second half of last season, Barnes has limited his fouling this year (after a foul-prone start to his collegiate career). He’s also nearly doubled his rate of forced resets/denies, doing a better job of overplaying the wings and preventing easy catches/scoring opportunities.

Barnes can be maddeningly inconsistent on the defensive end, interspersing inspired, dominant possessions with mental lapses and breakdowns in effort (especially related to transition defense and boxing out). While I think physical conditioning/stamina is sometimes an issue, he must do a better job of eliminating mental mistakes and competing hard on every defensive possession. Barnes has shown flashes of being a brilliant defensive wing, but he needs to become more committed and consistent on that end to make it a reality.

John Henson

2011 (all) 2012 (1st 10 games)
FG% All 28.7 24.3
3Pt% All. 33.3 13.5
FTA Rate 14.2 10.5
TS% All. 35.5 28.8
Pts. All. / 40 14.0 11.2
Deflections / 40 3.82 2.35
Forced TOs /40 2.43 1.80
Off. Fouls Drawn / 40 0.12 0.14
Denies / 40 0.96 1.25
DR% 24.8 25.0
Stop % 66.4 69.7
%Possessions 25.2 23.6
Def. On-Court/Off-Court +2.9 +8.0

After an ACC DPOY season as a sophomore, Henson’s bar for defensive improvement had been raised pretty high. Even so, he’s managed to exceed his lofty defensive charting numbers of ’11 with a brilliant start to his junior campaign. Henson is allowing opponents to shoot just 24.3% against him, all while maintaining a miniscule foul rate (just 10.5 free throw attempts allowed for every 100 field goals he defends). His block rate (11.1% last year, 10.5% this year) and DR% have both plateaued– albeit it All-American-caliber levels. Most of his defensive improvement has been fueled by his improvement against 3-point attempts (both as a hedger and a close-out defender)– after allowing 33.3% last year, opponents are shooting just 13.5% (5-37) from behind the arc against Henson after 10 games. His hedges are noticeably improved this season, although he’s still often a step slow on recovering/closing out on perimeter-oriented 4s. But you can’t teach length, of course, and Henson’s wingspan has proved troublesome for opponents both in the paint and on the perimeter (although I certainly expect some regression to the mean on his 3-point defense– despite his length, he’s been lucky on some of his lightly-contested forced misses from behind the arc).

Henson is still not a perfect fundamental defender (a reality underscored by the fact that Henson has won none of the staff’s first 9 defensive player of the game awards– an honor that rewards fundamental execution rather than defensive results). He’ll gamble for blocks when he should either wall or box out, he’ll allow some easy/deep entries in the post, he’s susceptible to face-up post moves and penetration from the perimeter, and his help rotations aren’t as consistently timely as Zeller’s. Despite these warts, his game-changing defensive presence is undeniable. Like last season, Henson has the team’s biggest +/- defensive impact (90.1 with him on the court, 98.1 with him on the bench)– an accurate reflection of his defensive dominance.

Tyler Zeller

2011 (all) 2012 (1st 10 games)
FG% All 31.9 33.7
3Pt% All. 32.7 25.0
FTA Rate 31.3 19.6
TS% All. 38.1 40.0
Pts. All. / 40 10.8 11.8
Deflections / 40 3.29 3.51
Forced TOs /40 3.46 3.81
Off. Fouls Drawn / 40 1.11 1.61
Denies / 40 0.80 1.61
DR% 15.2 16.4
Stop % 65.7 65.9
%Possessions 20.2 21.6
Def. On-Court/Off-Court +2.2 +0.6

Zeller, the perfect complementary post defender to Henson, leads the team in forced turnovers / 40 (3.81) and drawn offensive fouls / 40 (1.61). Unlike Henson, he is terrific at fronting the post and denying the entry pass. Though not a great shot-blocker, he uses his length effectively to set textbook walls and contest shots in the paint. Like Strickland, Zeller’s defensive charting numbers through 10 games look remarkably similar to his 2011 figures. Still not a great defensive rebounder, he’s making strides in that department since a rough opening game versus Michigan State. He’s unquestionably Carolina’s most dependable and consistent defender– the steadying force who anchors the entire defensive system with the Swiss-watch precision of his help rotations.

He can still be exposed as a hedge defender on the high ball screen– he’ll sometimes cheat back to his man at the expense of providing a hard/effective hedge. Alternatively (when he does hedge out hard), he can be a step late in recovering (which allows teams to effectively create some clean perimeter looks/deep post entries by using high ball screen action). Of course, well-executed pick-and-roll basketball by skilled teams is hard to defend at any level.

Reggie Bullock

2011 (all)
2012 (1st 10 games)
FG% All 49.6 29.2
3Pt% All. 29.2 16.1
FTA Rate 25.5 30.6
TS% All. 56.3 37.0
Pts. All. / 40 8.9 7.1
Deflections / 40 5.52 4.88
Forced TOs /40 2.66 3.60
Off. Fouls Drawn / 40 0.10 0.23
Denies / 40 1.40 2.79
DR% 14.9 11.8
Stop % 60.8 71.8
%Possessions 14.4 15.6
Def. On-Court/Off-Court -3.5 +6.1

After a slow defensive start to his Carolina career (Stop% of 48.1 in his first 10 games as a freshman), Bullock eliminated some mental mistakes, caught up to the speed of the collegiate game, and quickly became a fantastic wing defender. Over his final 17 games in 2011, Bullock posted a Stop% of 66.2%. So far as a sophomore, he’s improved that to (a team-leading) 71.8% through 10 games. Like Strickland, Bullock is capable of fantastic ball denial defense (a team-high 2.79 denies / 40). He uses his length and quick feet to put pressure on the ball and be extremely disruptive as a one-pass-away overplay defender. He’s also a tremendous defensive-rebounding wing (his slightly lower DR% this season reflects his increased minutes at the 2) with a nose for the ball. Best of all, perhaps, is Bullock’s competitive spirit and appetite to defend hard on every possession.

What has sent Bullock to a new defensive level this year is his ability to contest shots/close out of shooters. After allowing opponents to shoot 49.6% last season, he’s is forcing foes to a paltry 29.2% from the field through 10 games. Like with Henson, some of that is unsustainable and due to good luck/randomness (specifically the 16.1% that opponents are shooting from behind the arc versus Bullock). Teams are due to start making a few more of their lightly-contested perimeter attempts against Bullock. But what won’t change is Bullock’s fantastic defensive length as a help-and-recover close-out defender. While he still gets caught ball watching on occasion, or is a step late to locate his man in transition (two of the mistakes that plagued Bullock as a freshman– especially early in the season), he’s made huge strides within the mental/team aspect of Roy Williams’s defensive system. Still a little foul-prone, a slightly high FTA Rate is just something the Heels might have to live with to reap the rewards of Bullock’s aggressive/disruptive style of defense. He won’t maintain a 70+% Stop% all season; luck will catch up to him, and teams will game-plan/attack him a little more as his minutes continue to increase. But it’s clear to me that Bullock will continue being a very valuable defensive asset if he maintains (or improves upon) his current level of play.

P.J Hairston/James Michael McAdoo

Hairston McAdoo
FG% All 38.0 36.1
3Pt% All. 18.8 26.3
FTA Rate 36.6 37.8
TS% All. 45.6 47.5
Pts. All. / 40 12.6 11.8
Deflections / 40 2.98 5.25
Forced TOs /40 1.49 3.79
Off. Fouls Drawn / 40 0.99 0.58
Denies / 40 1.32 1.75
DR% 10.2 18.7
Stop % 52.6 63.7
%Possessions 16.5 20.6
Def. On-Court/Off-Court -13.6 -2.2

Not surprisingly, Carolina’s two freshmen have the highest defensive FTA Rates on the team. The combination of learning the speed of the college game, absorbing a complicated defensive system, and receiving no benefit of the doubt yet from officials (Hairston’s attempted some drawn charges that almost certainly would been called had the more experienced Zeller been taking them) is almost always one that results in (relatively) high foul numbers for Carolina freshmen. Plus, Hairston has a very bad habit of grabbing/holding to defend off-ball movement.

Hairston plays/competes hard, is willing to sacrifice his body (his 0.99 offensive fouls drawn / 40 is easily second on the team), and has good physical gifts defensively. Once he gets some more experience, he’ll be a very valuable wing defender in UNC’s system (even though he lacks the length of Barnes/Bullock or the lateral quickness of Strickland/Bullock). By the end of the year, I expect his disruption numbers (deflections, forced TOs, and denies/forced resets) to increase dramatically (they’re easily the lowest among Carolina’s guards/wings right now). That, along with a decrease in his foul rate, will allow Hairston to earn more minutes on the court by February/March. His defensive on-court/off-court of -13.6 (defensive efficiency of 101.8 with him on the floor, 88.2 with him on the bench) is the worst on the team. While I’m not surprised he ranks at the bottom in this metric through 10 games, he’s not been nearly as bad as his defensive +/- numbers might suggest.

McAdoo’s also been foul-prone to start his collegiate career. Much of these troubles have stemmed from his late defensive rotations, forcing him to foul rather than properly wall off the driver. And, on some egregiously late help rotations, he hasn’t even been in position to foul, but simply to take the ball out of the net after a made lay-up. With McAdoo’s superior lateral quickness and athleticism, I have no doubts that he’ll develop into a very good help defender. The game just needs to slow down for him a little right now. That superior quickness/athleticism is already playing dividends in some areas, though. McAdoo’s length and anticipation in the passing lanes have been remarkable for a 6’9″ player, and he leads the Heels in Steal% through 10 games (while just fractionally trailing Zeller for the team lead in forced TOs / 40). He’s also third on the team in deflections / 40, proving to be a very disruptive defensive force. McAdoo has shown a penchant for gambling for steals so far– often taking bad/low-percentage/unnecessary risks that have led to easy penetration opportunities for opponents. Once he learns when and when not to gamble (this was an issue for Danny Green early in his career, too), McAdoo’s disruptive style will pay even bigger dividends for UNC. On the defensive boards, McAdoo has been very solid (his DR% of 18.7 ranks between those of Zeller (16.4%) and Henson (25.0%)). He’s shown the ability to grab rebounds both out of area and above the rim, although he’s been overpowered on (or just flat-out missed) some box-out assignments so far.

Speaking of defensive rebounding, let’s check out the following table to see how UNC compares to last season on the defensive backboards. The numbers in this table represent a player’s individual DR%– defined as defensive rebounds / (defensive rebounds + offensive rebounds allowed).

Player Ind. DR% (2011) Ind. DR% (2012) Change
Marshall 84.5 80.7 -3.8
Strickland 71.3 79.2 +7.9
Barnes 81.5 66.7 -14.8
Henson 76.1 74.5 -1.6
Zeller 65.7 72.7 +7.0
Bullock 76.1 73.3 -2.8
Watts 60.0 57.4 -2.6
McDonald (’11)/Hairston (’12) 86.0 81.3 -4.7
Knox (’11)/McAdoo (’12) 62.4 69.4 +7.0
Team Offensive Rebounds Allowed / Game (Team DR%) 2.6 (36.7) 3.5 (25.5) -11.2
Overall DR% 70.1 69.4 -0.7

As seen in the final row, UNC has been slightly worse on the defensive glass in 2012 than it was in 2011. This is despite an upgrade (at least so far) from Knox to McAdoo, and a pronounced improvement from Zeller on the defensive glass (he’s allowing 2.6 offensive boards / 40 this year after 3.2 / 40 in ’11). The biggest problem has been Barnes’s thoroughly mediocre work on the defensive glass: he’s grabbing far fewer DRs per 40 than a year ago (3.2 in ’12 vs. 5.2 in ’11) while allowing more ORs (1.6 in ’12 vs. 1.2 in ’11). He simply needs to get better as a defensive rebounder for UNC to reach its defensive potential. Henson’s giving up slightly more ORs / 40, too (3.7 in ’12 vs. 3.2 in ’11), but has still been a very good overall defensive rebounder. The team has also given up an extra offensive rebound / game– these generally occur when a shot is blocked out of bounds (or directly to an opponent, etc.), or in transition. Tightening up the transition defense and doing a better job of containing dribble penetration (which pulls help defenders out of defensive rebounding position) should help the Heels to improve a little in the “team rebounding” area.

About Adrian

I'm the editor of Maple Street Press's Tar Heel Tip-off, and live in Raleigh with my wife and 2-year-old daughter. I grew up along the banks of the Allegheny River, and terrorized the WPIAL as a pass-first point guard. Follow me on Twitter @FreeportKid.
This entry was posted in UNC and tagged , , . Bookmark the permalink.

3 Responses to At the Quarter Mark: Some Defensive Charting Stats and Observations

Register |

  1. C. Michael says:

    Great work, Adrian. I know that this is a lot easier said than done, but have you thought about measuring Stop% as a function of % of minutes played against an opponents starters/bench? I’d be curious to see if a part of Bullock’s tremendous stop% stems from a higher % of minutes played against opposition bench players, who are generally going to be much more limited offensively than is UNC’s bench.

    1. Adrian says:

      Yeah, I’ve fooled around with different ways to adjust for this. Marshall was definitely much better last year from a Stop% perspective as a reserve– it does make a difference. I think Bullock as a starter would drop into the mid-60s or so in terms of Stop% (he might as it is when some of his luck levels off)– he’s still be very, very good, though.

      The other key thing that remains unaccounted for in this analysis is quality of opponent. Guys like Strickland/Ginyard/Manuel (who are usually covering the best perimeter or wing scorer) have a tougher average assignment than other players. But I don’t adjust for this in the Stop%/DRatings– at least not yet.

  2. brock matthews says:

    Great work. Keep it up.

Leave a Reply

Your email address will not be published. Required fields are marked *