How Duke’s Improved Defense Has Led Them to the Final Four

An examination of how Duke’s defense has improved leading up to -- and during -- the NCAA Tournament.

Coming into the NCAA Tournament the fact that Duke had one of the best offenses in the nation was widely known. In that respect they have not been disappointing.

After four tournament games, the Blue Devils’ offensive percentile rank of 99.7% is second best in the country behind fellow Final Four contender Wisconsin. What was more of a question was how the Duke defense would hold up against tournament competition.

In short, they have been nothing short of excellent.

Coming into the tournament the Duke defense ranked in the 76th percentile nationally, but after only four tournament games they have improved this rank to the 86th percentile, which means they are better on defense than 86 percent of teams in the country. They have not allowed any of their tournament opponents to break 60 points, despite giving up a season average of 64.27 points per game.

Here, we look at what has led to this known offensive powerhouse to play like a defensive juggernaut leading up to, and during, the NCAA Tournament

The Insertion of Matt Jones Into the Starting Lineup

Matt Jones played the majority of the season as Duke’s seventh man. Following a loss to Notre Dame, on January 29, however, Junior guard Rasheed Sulaimon was dismissed from the team, and Jones moved up in the rotation. Then, on February 21, against Clemson with Freshman Jahlil Okafor sidelined with an injury, Jones was thrust into the starting lineup. Despite the return of Okafor to the starting five, Jones never left.

Jones entered the lineup as a wing player and perimeter defender, which shifted Freshman Justise Winslow to defending the power forward position in Duke’s man-to-man defensive system, and Junior Amile Jefferson was moved to the bench. While it is difficult to analyze the relative contributions of Jefferson and Jones because they play different positions, it is important to compare their defensive statistical averages to determine the net gain to the team since the switch has taken place.

Prior to being replaced in the starting lineup, Jefferson averaged .69 steals, .65 blocks, and was called for 2.19 fouls per game. After being inserted in to the starting lineup Jones has averaged 1.36 steals, .18 blocks, and 1.38 fouls per game. Thus, just this transition has benefited Duke to the tune of being called for nearly half a foul less per game and improving their steals by .67 per game but reduced their block totals by around one every two games.

This does not end the analysis, however, as we must also consider any net gain or loss to the defense that occurred due to Winslow being moved to defending the power forward position. Since the move, Winslow has maintained a nearly identical foul rate, while blocking .36 more shots per game, and averaging an incredible 1.54 more steals. When this new role for Winslow is factored in to the swap of Jefferson and Jones, the change translates to .47 less fouls per game, and losing approximately 1 block every 10 games, while increasing the team’s steals per game to an average of 2.21 steals per game.

Moreover, it is possible that this move may have improved other aspects of Duke’s defense as well.

For example, because Winslow -- and his team-best Defensive Rating of 93.1 -- is absurdly athletic, he is able to switch off and cover guards when his defender sets a screen, rather than forcing a teammate to try to work around the screen to stay on his man every possession. Overall, Duke is giving up 2.5 fewer points per game, committing 2.23 fewer fouls per game, and drawing 4.49 fouls more per game since moving Jones into the starting rotation.

While this accounts for a good portion of Duke’s defensive improvement toward the end of the season, it does not entirely explain the tournament performance, as Duke has given up, on average, 9.68 fewer points per game than they did on the season even after Jones was moved into the starting lineups.

Thus, let’s examine some other improvements and potential explanations.

Increased Blocks

The most noteworthy statistical improvement in the Duke defense during the tournament has been an increase in blocks by their two main post players. Winslow is averaging 1.75 blocks during the tournament compared to his average of 1.09 during the regular season after the insertion of Jones. Similarly, Okafor’s block average is up to 2.0 per game during the tournament as compared to the overall average of 1.3 since Jones has been a starter.

Together this equates to nearly five and a half (5.46) more blocks over the four-game tournament stretch as compared to their season average in games when Duke used the current starting five.

Winning the Foul Battle

During the course of the season, Duke has drawn 20.5% more fouls from opponents than they have committed themselves. During the NCAA tournament this differential has increased to 27.1%.

Even more noteworthy is the key opposing personnel that Duke has drawn those fouls against. For example, against Utah, Utes superstar Delon Wright had to sit out the last 4:59 of the first half after committing his third personal foul. He committed his fourth with 7:58 left to go in the game and had to finish with the knowledge that one more foul would remove him from the contest.

This is particularly noteworthy due to the fact that during the course of the season, Wright averaged only 2.7 fouls per 40 minutes. Similarly, Duke was able to get Gonzaga’s center Przemek Karnowski in early foul trouble, causing him to miss more than eight minutes of the first half of that game. By contrast, all of the Blue Devils’ starters have committed 3 fouls or fewer in each of the four tournament games -- with the exception of Quinn Cook’s four fouls versus Utah.

With the insertion of Matt Jones into the starting lineup in February, the increased rim protection, and better job of drawing fouls in the tournament, Duke has emerged as a highly improved and dangerous defensive team.

If this performance continues, this could prove difficult for a Michigan State team whose offense has been impressive this year (89th percentile) but not as impressive as the Gonzaga (99th percentile) and Utah teams (95th percentile) that Duke has already controlled. Among other things, keys for Michigan State will be to take smart shots and avoid getting their stars into foul trouble early. If they can’t, then this new-look Duke squad could be destined for a national title game appearance.