Duke's Thin Rotation: We've Been Here Before

Mike Krzyzewski has the thinnest bench in his 35 years as Duke's head coach.

If you're going to shop Amazon please start here and help DBR
Drop us a line

Mike Krzyzewski has enough players to play his normal game rotation.

Although those two statements sound contradictory, both are true.

Duke is down to eight scholarship basketball players this season. There are two walk-ons, but neither has played a second in a competitive situation (and is not likely to this season). Coach K also has a promising big man sitting out as a transfer, but while Sean Obi can help in practice, he's not available for games.

So that's it - eight players … and four of them are freshmen.

Has K ever been in that situation before?

Twice, kind of.

The first was in 1995-96. Duke, with K returning from his bad back sabbatical, started the year with nine scholarship players, But senior Tony Moore flunked out after the first semester (seven games), leaving him with eight. Trajan Langdon was also on the team, but missed the entire season with a knee injury.

That team did have some substantial walk-on help. Jay Heaps, a soccer All-American, and Stan Brunson, a 6-7 soccer player, provided some major minutes - even in crucial situations in key games. Late in the season, both were called on to play bigger roles as injuries either sidelined or hobbled regulars. In the NCAA Tournament loss to Eastern Michigan, Brunson got 17 minutes in relief of a foul-plagued Taymon Domzalzki (15 more minutes than sore-kneed freshman center Matt Christensen).

The second eight-man Duke team was 2004-05.

Again, that team had some special help. Reggie Love, a football star who had extensive experience on Duke's 2001 national champs, played 21 games. Two more experienced walk-ons - big man Patrick Johnson and guard Patrick Davidson - played significant minutes in significant games. Both started one game - Davidson famously dueling Chris Paul in the opening minutes of a game against Wake Forest.

The 2005 Devils needed the help as injuries hobbled a number of players over the course of the season - DeMarcus Nelson, David McClure and Shav Randolph all missed time.

That's it - two previous teams with just eight scholarship players.

For 32 of his other 33 years, Krzyzewski had between nine (five times) and 12 (nine times) recruited players eligible. The other exception was 1983, when - astonishingly - K's second worst team had 14! scholarship players (not sure how that worked, but that's how many there were).

The current situation was clearly unplanned. Krzyzewski started the season with 10 recruited players on his roster (not counting Obi). The departure of sophomore Semi Ojeleye was disappointing (especially to me - I was just learning to spell his name), but not surprising - it is the second year in a row that Duke has lost a forward to a mid-season transfer.

Obviously, the dismissal of junior guard Rasheed Sulaimon - a key member of the Duke rotation - was a stunner. And while Ojeleye's departure caused barely a ripple in the Duke rotation (he played just 61 minutes in six games), the loss of Sulaimon, averaging almost 20 minutes a game, had a huge impact.

TRACKING THE MINUTES

Duke fans (at least that post on the DBR message boards) have an unhealthy obsession with minutes played and how they are divided.

The truth is that K rarely overplays his best players. A year ago, Duke did not have a single player ranked in the ACC minutes played top 10. At the moment, he has one - senior Quinn Cook is fifth (35.7 minutes a game), trailing BC's Olivier Hanlon, Syracuse's Trevor Cooney, Notre Dame's Jerian Grant, and N.C. State's Trevor Lacey.

Of course, those season-long stats are misleading, since it's been just seven games since Sulaimon was dismissed. Clearly, his departure has changed the rotation and forced several players to up their minutes. Going into tonight's game at Virginia Tech, the change in minutes played (keep in mind that the last seven games includes an extra five minutes for the UNC overtime):

It's clear that most of Sulaimon's minutes have been divided between Cook (plus-4), Tyus Jones (plus-7.6), Grayson Allen (plus 3) and Matt Jones (plus-10). That adds up to more than Sulaimon's 19.3 mpg. I suspect that's because Matt Jones had also been taking minutes that had previously gone to Jefferson (minus-4) as K has gone more and more to a smaller lineup.

The increase in freshman minutes is not that surprising since it is normal for a freshman to play longer late in the season. Tyus Jones' jump is quite dramatic. And just for the record, Marshall Plumlee's jump is entirely due to Okafor's absence in the Clemson game. He played 24 minutes in that one - his average in the previous six post-Sulaimon games was 8.1 mpg - actually below his pre-Sulaimon average.

So let's plug the current averages into the ACC's minutes-played list. Based on the last seven games, Cook would rank first and Tyus Jones second on the ACC list. No other Duke player would be in the top 10. You get a slightly different result when you look at the ACC's conference-only minutes-played list - Cook is tied for third (with both Grant and former Dukie Michael Gbinije), while Tyus Jones comes in tied for sixth (with Rakeem Christmas).

I guess the point is that while Duke's top players are playing a lot of minutes, it's not an inordinate amount - especially since you consider we're just talking about a seven-game stretch so far.

COMPARING ROTATIONS

Duke's rotation - as it now stands - contains four players getting 30-plus minutes a game, one in the 20-plus minute range (although Jefferson is there for the season and is very close in the last seven games and two players just barely in the 10-plus minute range.

How does that compare to other top teams?

I made this graph to compare Duke's rotation to the ACC's other four ranked teams. In my mind, a player averaging 10 minutes a game is "in the rotation".

The Louisville numbers do not include Chris Jones, who was averaging 34.6 mpg in ACC play. I'm sure their numbers will change as Duke's have post-Sulaimon, Wayne Blackshear will almost certainly increase his average of 27.3 mpg and freshman Quentin Snider's average of 10.0 mpg will skyrocket.

Virginia's numbers include Justin Anderson, who was averaging 31.9 mpg when he went out. It doesn't include Isaiah Wilkins, who is just below the 10 mpg barrier or Devin Hall, whose role has increased in Anderson's absence, but it is still at just 7.4 mpg.

Oddly, the three ACC teams with the most players in the rotation are one middle-of-the-road team and two bottom feeders. Miami, Wake Forest and Virginia Tech have used 10-man rotations.

I guess the point of the chart was to demonstrate that Duke's rotation is right in line with the ACC's other top teams. UNC is playing - on average - one more guy a game. Notre Dame and now Louisville are playing one guy less.

Nationally, let's run the same chart for the non-ACC top teams:

It's clear that Kentucky and Kansas are using more players, but also that an 8-man rotation - that's eight players getting double-figure minutes - is routine - it's probably the national mean.

It's also in line with Coach K's history. Bear with me for one more rotation chart - of Krzyzewski's 11 Final Four teams:

It's obvious from this list that K is comfortable with an eight-man rotation. He's reached the Final Four twice with a nine-man rotation (winning the national title in 1991) and four times with a seven-man rotation (including a title in 2001). Bur eight is the median number - five Final Fours and two titles with an eight-man rotation.

So, it's clear to me that eight IS enough this season. In fact, seven would be enough - even if Plumlee or Allen drop back below the 10 mpg mark.

THE ADVANTAGES OF DEPTH

There is a silly thread on DBR complaining about the team's lack of depth and suggesting that Coach K and his staff should have done more to prevent this problem. To be fair, most of the posters have pointed out just how uniformed that point of view is.

The fact is that Duke went into the season with great depth - depth great enough to withstand the permanent loss of two players and the temporary absence of a third - the team's best player.

There is enough depth left to make a run at the ACC and national championship - if the team falls short, it won't be for lack of depth.

That's not to disparage the advantages of a deep team. Depth gives a coach flexibility - both to build his team early and to the withstand problems later.

Go back to 1991. One reason that championship team used nine players in the rotation was that Krzyzewski struggled to put together his best team. True, he knew he had Christian Laettner, Bobby Hurley and Grant Hill as his core, but who should he fit around them? He had gifted guards Thomas Hill and Billy McCaffrey, who both got a shot to start. He tried Crawford Palmer at center (moving Laettner to forward) in a number of games. He gave freshman Tony Lang eight starts. On the wing, Brian Davis got 11 starts and Greg Koubek 13.

By the Final Four, he was down to a seven-man rotation. Thomas Hill and Koubek joined the core three in the starting lineup, while McCaffrey and Davis got substantial minutes off the bench. That was it.

Duke didn't need its depth in Indianapolis, but K needed it to build that team.

It was a different story in 2001. Through most of the season, K used a six-man rotation - five starters and freshman Chris Duhon off the bench. But when center Carlos Boozer got hurt in the last regular season game, K was lucky to have some depth in the post.

Casey Sanders, who had averaged 4.2 mpg in Duke's first 25 contests, moved off the bench and into the starting lineup. He averaged 16.5 mpg the rest of the way. Walk-on Reggie Love, who had played 62 mop-up minutes in the first 25 games, averaged 8.5 mpg in the next eight (he did not play in the Final Four after Boozer returned). Matt Christensen averaged 9 mpg in the first five games after Boozer was hurt (he did not play in the regionals or the Final Four).

The point is that K had some options. Those didn't seem to be great options on the night when Boozer went down, but he made them work.

He has no options with this team. Really, if Duke is going to succeed, his core six have to play major minutes and they have to play well. Plumlee and Allen give him some flexibility, but that's it.

The sprained ankle that sidelined Jahlil Okafor Saturday is evidence of both how thin the line is with just eight useable players on the roster … and how deep the talent is on this team that it could rout a solid Clemson team with just seven players.

Now, if Okafor was to be out long term, that would be disastrous. But it would be disastrous more because he's the team's best player than because of Duke's limited depth. Seriously, does anybody think Duke would win the ACC or make the Final Four without Okafor - even if K had Ojeleye, Sulaimon and even earlier transfers Michael Gbinije and Alex Murphy still on the bench?

Duke faces a lot of challenges before the Blue Devils can achieve what they want to achieve this season.

But the eight-man rotation is not one of the team's major tests.