Hot Take: Duke's best offensive lineup is without Marvin Bagley

[dragoneye776]

Before I get completely flamed: I don't necessarily think this is true, but it would be interesting to consider this point of view

Duke has reached amazing offensive efficiency in recent history due to a four guard + one big lineup (at times with significant defensive cost). This is the first year where they consistently play a two big lineup. The advantages are in post play, rebounding, and paint defense, but a significant disadvantage this year is in floor spacing for our guards to drive.

A big part of the reason this forum thinks Grayson is taking such a huge step back this year is because there are always two bigs in the paint and he is driving a lot less. Similarly, Duval's greatest strength is also his driving ability. Thus my hot take is that going back to a four guard + one big lineup (ex. Duval, Grayson, O'Connell, Trent, Carter) may elevate the offense of our guards more than we miss the post presence of Bagley. Four Duke quality guards who can shoot, drive, and create would be a nightmare offensively.

I've also noticed that decent defenses do a great job to preventing the entry pass to Bagley, and we often just waste over half the shot clock attempting to get the ball to him. And when he gets doubled, he’s not that good at passing out of it, so he just dribbles to the perimeter and our offense resets. During the time we are trying to make the entry pass, most of our other guards just stand around on the perimeter. Perhaps a more motion-oriented offense would better suit them?

Perhaps the best offensive lineup is with Bagley instead of Carter as the one big, but the point of the big in this case is to just set screens and create for the guards, which Carter is better than Bagley at. And Carter can perhaps try an occasional pick n pop too.

tl;dr: A four guard + one big lineup would be better offensively than the current three guard + two big lineup, and Carter would be a better choice at the big than Bagley

[FerryFor50]

Before I get completely flamed: I don't necessarily think this is true, but it would be interesting to consider this point of view

Duke has reached amazing offensive efficiency in recent history due to a four guard + one big lineup (at times with significant defensive cost). This is the first year where they consistently play a two big lineup. The advantages are in post play, rebounding, and paint defense, but a significant disadvantage this year is in floor spacing for our guards to drive.

A big part of the reason this forum thinks Grayson is taking such a huge step back this year is because there are always two bigs in the paint and he is driving a lot less. Similarly, Duval's greatest strength is also his driving ability. Thus my hot take is that going back to a four guard + one big lineup (ex. Duval, Grayson, O'Connell, Trent, Carter) may elevate the offense of our guards more than we miss the post presence of Bagley. Four Duke quality guards who can shoot, drive, and create would be a nightmare offensively.

I've also noticed that decent defenses do a great job to preventing the entry pass to Bagley, and we often just waste over half the shot clock attempting to get the ball to him. And when he gets doubled, he’s not that good at passing out of it, so he just dribbles to the perimeter and our offense resets. During the time we are trying to make the entry pass, most of our other guards just stand around on the perimeter. Perhaps a more motion-oriented offense would better suit them?

Perhaps the best offensive lineup is with Bagley instead of Carter as the one big, but the point of the big in this case is to just set screens and create for the guards, which Carter is better than Bagley at. And Carter can perhaps try an occasional pick n pop too.

tl;dr: A four guard + one big lineup would be better offensively than the current three guard + two big lineup, and Carter would be a better choice at the big instead of Bagley

It's not a crazy theory, but a few things:

- Bagley is essentially able to play 3-5 on the court, which makes him invaluable
- Carter can step out and hit threes if necessary, as well as the mid range
- Duke's offensive rating is 127.2, good for #2 in KenPom. So they're already as good as they're going to get.
- Defense is this team's biggest issue, and Carter/Bagley are better defensive options than most, if not all of the guards

[CDu]

The other problem with this theory is that we can't really play a 4-guard, 1-big lineup. At least not without a huge dropoff in ability. We only have 4 top-100 talents at guard on the roster.

The best backup for Bagley at the moment is Marques Bolden, who is decidedly less of a floor spacer than Bagley. After Bolden, it's O'Connell. But after O'Connell, it's a pair of PFs in DeLaurier and White.

If we had at least one more guard on the roster, I might be willing to discuss it. But as is, I think it's fair to say that this team is better offensively with Bagley than without.

[dragoneye776]

Yeah I agree that defense is our problem but say that we are playing Virginia just down a few points and we need some injection of quick offense. What would theoretically be the best offensive lineup Duke could put out there to close a half or something?

And you don't think O'Connell is good enough to be the 4th guard in the 4 guard lineup?

[English]

Yeah I agree that defense is our problem but say that we are playing Virginia just down a few points and we need some injection of quick offense. What would theoretically be the best offensive lineup Duke could put out there to close a half or something?

And you don't think O'Connell is good enough to be the 4th guard in the 4 guard lineup?

The point isn't that AOC isn't good enough to be the fourth guard, it's that our entire roster only has four guards good enough to play rotation minutes, so if we roll Duval, Allen, AOC, and Trent out there...then what? This isn't a video game. These guys get tired, or pick up fouls, and there's nobody behind them to spell them.

[Kedsy]

Duke has reached amazing offensive efficiency in recent history due to a four guard + one big lineup (at times with significant defensive cost). This is the first year where they consistently play a two big lineup. * * *

Thus my hot take is that going back to a four guard + one big lineup (ex. Duval, Grayson, O'Connell, Trent, Carter) may elevate the offense of our guards more than we miss the post presence of Bagley. Four Duke quality guards who can shoot, drive, and create would be a nightmare offensively.

DUKE ADJUSTED OFFENSIVE EFFICIENCY IN KENPOM ERA
------------------------------------------------------
2002: 121.0
2003: 115.0
2004: 120.8
2005: 115.7
2006: 119.4
2007: 113.2
2008: 116.3
2009: 117.5
2010: 121.1
2011: 119.7
2012: 117.7
2013: 119.1
2014: 124.7
2015: 124.5
2016: 121.9
2017: 121.0
2018: 127.2

So I question the assumptions underlying your theory.

Earlier this season we were moving the ball well and assisting in a high percentage of our made shots. That’s has totally gone away lately, in favor of attempting to just dump it in the post (which also causes a ton of turnovers).

%assisted on made shots (Duke vs.
-----------------------------------
Elon: 51.2%
Utah Valley: 66.7%
Michigan State: 70.0%
Southern: 56.0%
Furman: 46.2%
Portland State: 59.4%
Texas: 70.0%
Florida: 69.0%
Indiana: 45.5%
South Dakota: 48.6%
St. Francis: 70.8%
BC: 40.6%
Evansville: 82.1%
FSU: 48.7%
NCSU: 32.2%
Pitt 1: 61.8%
Wake 1: 72.0%
Miami: 58.0%
Pitt 2: 55.2%
Wake 2: 50.0%
Virginia: 55.5%
Notre Dame: 46.9%
St. John's: 41.7%
UNC: 63.3%
Ga Tech: 61.5%

On the one hand, it may not be coincidence that our lowest %assisted totals were (in order): NC State, BC, and St. John's. It also may not be coincidence that our highest %assisted totals were (in order) Evansville, Wake 1, and St. Francis. And while it's true that four of our highest ten %assisted numbers came in November, it's also true that four of the highest ten came in the last month. So you may be right that a higher percentage of assisted baskets is a sign of a more efficient offense, but I think the pattern is more complex than an "earlier in the season"/"lately in the season" split.

[Wander]

So I question the assumptions underlying your theory.

I don't think kenpom numbers are really directly comparable across different seasons like that.

That said, I agree with you that our offense is very good and doesn't really need an overhaul. Our defense on the other hand... would playing a smaller lineup allow us to play a different defense that would result in more forced turnovers? I think that is our biggest weakness.

[Kedsy]

Our defense on the other hand... would playing a smaller lineup allow us to play a different defense that would result in more forced turnovers? I think that is our biggest weakness.

Interesting idea. But considering that the only small lineup we can reasonably play would insert Alex (1.1 steal%) -- who has the lowest steal percentage on the team among players with 100 or more minutes -- for Marvin (1.7 stl%) or Javin (3.1%), or even Marques (1.8%), it's hard to see more forced turnovers from the smaller lineup. I know that forcing turnovers is more than just getting steals, but at least to me Alex doesn't appear to be a defensive dynamo in other ways, either.

[kAzE]

I don't think kenpom numbers are really directly comparable across different seasons like that.

That said, I agree with you that our offense is very good and doesn't really need an overhaul. Our defense on the other hand... would playing a smaller lineup allow us to play a different defense that would result in more forced turnovers? I think that is our biggest weakness.

I keep seeing different versions of this comment in various threads. While it's true that our offense is better than our defense, we might actually have more room to improve on offense than defense. We still commit a ton of unforced turnovers, Grayson still hasn't really been himself since Michigan State, Bagley still only goes left, and Duval still hasn't really found his place on the team yet. If Grayson had been his usual all-american self all year, or we make a couple fewer dumb turnovers, we could be 24-1 right now, even with the atrocious defense.

These guys are all offensive players. ALL of them are better on offense than D. There is no Matt Jones or Amile Jefferson on this team. My point is that we might actually be closer to our defensive ceiling than our offensive ceiling. So the "our offense is fine, we need to improve on D" comments don't make much sense to me. We need to improve on both ends. Just because we're better than almost everyone else at offense doesn't mean we can't still improve a lot on that end.

[flyingdutchdevil]

I keep seeing different versions of this comment in various threads. While it's true that our offense is better than our defense, we might actually have more room to improve on offense than defense. We still commit a ton of unforced turnovers, Grayson still hasn't really been himself since Michigan State, Bagley still only goes left, and Duval still hasn't really found his place on the team yet. If Grayson had been his usual all-american self all year, or we make a couple fewer dumb turnovers, we could be 24-1 right now, even with the atrocious defense.

These guys are all offensive players. ALL of them are better on offense than D. There is no Matt Jones or Amile Jefferson on this team. My point is that we might actually be closer to our defensive ceiling than our offensive ceiling. So the "our offense is fine, we need to improve on D" comments don't make much sense to me. We need to improve on both ends. Just because we're better than almost everyone else at offense doesn't mean we can't still improve a lot on that end.

I sadly agree. I think this team can get so much better on offense. I unfortunately don't see how to fix some of the following issues: forcing turnovers (Allen has never been good at forcing TOs. Duval's reaction speed is shockingly slow for a PG. The bigs block shots but don't poke balls), preventing penetration (it's a modern Duke team. This ain't happenin'), on-court awareness (Duval and Bagley seem to be the biggest culprits of this), PnR defense (Carter is getting better but it's still not great), and communication (does this get fixed in a year? Probably not).

[CDu]

I keep seeing different versions of this comment in various threads. While it's true that our offense is better than our defense, we might actually have more room to improve on offense than defense. We still commit a ton of unforced turnovers, Grayson still hasn't really been himself since Michigan State, Bagley still only goes left, and Duval still hasn't really found his place on the team yet. If Grayson had been his usual all-american self all year, or we make a couple fewer dumb turnovers, we could be 24-1 right now, even with the atrocious defense.

These guys are all offensive players. ALL of them are better on offense than D. There is no Matt Jones or Amile Jefferson on this team. My point is that we might actually be closer to our defensive ceiling than our offensive ceiling. So the "our offense is fine, we need to improve on D" comments don't make much sense to me. We need to improve on both ends. Just because we're better than almost everyone else at offense doesn't mean we can't still improve a lot on that end.

This is a fair point, for the most part. I think that there is defensive potential in Bagley and Duval at the very least. And Allen is physically capable of solid defense, in the same way that Cook became solid defensively in the second half of his senior year. But I agree that the offense does have room to improve.

I will note, though, that improvements in offense may not dramatically shift the offensive efficiency numbers. At least not terribly dramatically. Part of our insanely high offensive efficiency this season is driven by having the #1 offensive rebounding team in the country (though that status may change if things keep going as they are going; UNC has outrebounded us comfortably in conference, and is currently #3 overall). Those offensive rebounds generally result in opportunities for our bigs, who are our most efficient scorers. So if we get more efficient on our first shots, some of that will come at the cost of second-chance opportunities.

Obviously still better to be more efficient on first-chance opportunities of course.

[kAzE]

This is a fair point, for the most part. I think that there is defensive potential in Bagley and Duval at the very least. And Allen is physically capable of solid defense, in the same way that Cook became solid defensively in the second half of his senior year. But I agree that the offense does have room to improve.

I will note, though, that improvements in offense may not dramatically shift the offensive efficiency numbers. At least not terribly dramatically. Part of our insanely high offensive efficiency this season is driven by having the #1 offensive rebounding team in the country (though that status may change if things keep going as they are going; UNC has outrebounded us comfortably in conference, and is currently #3 overall). Those offensive rebounds generally result in opportunities for our bigs, who are our most efficient scorers. So if we get more efficient on our first shots, some of that will come at the cost of second-chance opportunities.

Obviously still better to be more efficient on first-chance opportunities of course.

This is true, but a lot of times, we don't even get our first shot. Looking at our last 4 losses, two things stand out: Our opponent always commits fewer turnovers than we do, and also attempts more shots (in some cases WAY more shots) than we do. This is actually even worse than it sounds, since we are the best offensive rebounding team in the nation. We should really be getting more shot attempts than anyone we play against. Part of this is also not rebounding on the defensive glass and giving up extra possessions. That's one area on defense I know we can improve on.

But still, as good as our offense has been, it could be much, much better. I think this team needs to outscore opponents to win, because like FDD said, a lot of this team's defensive flaws aren't easily fixed. In order to outscore teams, we've got to take better care of the ball and make sure we get something out of every possession.

[Kedsy]

This is true, but a lot of times, we don't even get our first shot. Looking at our last 4 losses, two things stand out: Our opponent always commits fewer turnovers than we do, and also attempts more shots (in some cases WAY more shots) than we do. This is actually even worse than it sounds, since we are the best offensive rebounding team in the nation. We should really be getting more shot attempts than anyone we play against. Part of this is also not rebounding on the defensive glass and giving up extra possessions. That's one area on defense I know we can improve on.

Yeah, in our losses to State, UNC, and St. John's, our defensive rebounding was atrocious. That's clearly an area we can improve. And while it's true UNC committed more turnovers than us, we only had 8. Hard to reasonably expect fewer than that in our offense (I recognize we had too many TOs against St. John's and Virginia and State; but those were somewhat anomalous performances). On the other hand, we only forced two UNC turnovers. FDD may be right that we'll never be a great turnover-causing team, but we should have a much higher chance of forcing more than two (or the five we forced against UVa) than we do of cutting down on the 8 miscues we made in the UNC game.

But your stats don't differentiate between turnovers forced by the opponent and turnovers like throwing a post entry pass out of bounds or dribbling into traffic with no plan and coughing it up. These are turnovers that other good offensive teams don't make. And again, just because this is one of the best offensive teams ever on KenPom doesn't mean it can't be even better than it's been. Anybody who's watched our 4 losses should agree with that.

You appear to be looking for perfection, but I don't think that's reasonable. Overall, we don't make too many turnovers on offense. The fact that we did in a few games (including three of our five losses) doesn't change that, it just means we had a few bad games.

However, I do think the 1 major thing we have to do better is defensive rebounding. With our size, there's no excuse for not being better at that.

I agree with this, and this alone will help our defensive efficiency considerably.

[HereBeforeCoachK]

And you don't think O'Connell is good enough to be the 4th guard in the 4 guard lineup?

I believe AOC is good enough to be the fourth guard, I just don't think he's good enough to justify playing only one of our bigs....

[Ian]

The solution is not to just play one big, the solution is to feature the post up less in our offense, and space the floor with AOC instead of Duval.

[dragoneye776]

The solution is not to just play one big, the solution is to feature the post up less in our offense, and space the floor with AOC instead of Duval.

Love this. I've watched a few games of the New Orleans Pelicans this year who played Demarcus Cousins and Anthony Davis in a two big lineup before Cousins got injured. While they do occasionally post up, they still run most of their offense through ball movement pick n roll and cutting principles.

And great discussion so far. Keep in mind my point was not at all that we should value improvement in offense over defense, I was just wondering if four guards would theoretically be a better offense than the two big offense.

Those offensive rebounds generally result in opportunities for our bigs, who are our most efficient scorers. So if we get more efficient on our first shots, some of that will come at the cost of second-chance opportunities.

Interesting if there is a relationship between how well our guards can finish vs. how will our bigs can rebound offensively. If we have four guards to space the floor and the last big may be rolling after setting a screen, I expect our guards to finish very well but no bigs around for the offensive rebound. On the other hand, if a guard tries to drive into a paint where Bagley and Carter are hanging out at the rim, that guard won't finish as well but we will have a very high rate of offensive rebounding. The problem is for example during the UNC game where we had two bigs in the paint but were still getting out rebounded.

[ipatent]

Duke's offense is better when Grayson Allen takes and hits shots. Too small a sample size right now to prove he has more of those big games when Bagley is out.

[kAzE]

Grayson's game is more suited for the 4 out 1 in lineup, but that doesn't mean he can't also be good with Bagley and another big man in the lineup. I just think Bagley needs to pass out of the post a bit more often. He's outrageously efficient in the paint, but sometimes, he still attempts some extremely difficult shots. I wish he would kick it back out or look for a cutter in those situations. He is the #1 option on offense, but it needs to be a team effort to get everyone involved, particularly Grayson. Guys need to look for him more and try to get him the ball where he's most effective. Bagley's actually better crashing the glass than as a post up player, where his over-reliance on his left hand has caused him to struggle against better defensive teams lately.

[flyingdutchdevil]

It's not a crazy theory, but a few things:

- Bagley is essentially able to play 3-5 on the court, which makes him invaluable
- Carter can step out and hit threes if necessary, as well as the mid range
- Duke's offensive rating is 127.2, good for #2 in KenPom. So they're already as good as they're going to get.
- Defense is this team's biggest issue, and Carter/Bagley are better defensive options than most, if not all of the guards

Agreed on your points except the last one. 4 guards + 1 big isn't the answer (because AOC and Duval aren't good defensively). 3 guards + 2 bigs is clearly better. But Carter/Bagley isn't a great defensive tandem. I really like the Carter/Bolden tandem. Bolden moves his feet better, is significantly longer, reads the O better, and an all-round smarter defender. Bagley is a significantly better D rebounder, but that's about all he does well on the defensive end.

The issue is on O, where Bolden can only play down low. That means Carter needs to hit more of those 3s that we're accustomed to him taking.

I would love to see Bolden play ~15-20 a game. I'm not saying his D will cure our D, but at least it's moving in the right direction.

[Truth&Justise]

Agreed on your points except the last one. 4 guards + 1 big isn't the answer (because AOC and Duval aren't good defensively). 3 guards + 2 bigs is clearly better. But Carter/Bagley isn't a great defensive tandem. I really like the Carter/Bolden tandem. Bolden moves his feet better, is significantly longer, reads the O better, and an all-round smarter defender. Bagley is a significantly better D rebounder, but that's about all he does well on the defensive end.

The issue is on O, where Bolden can only play down low. That means Carter needs to hit more of those 3s that we're accustomed to him taking.

I would love to see Bolden play ~15-20 a game. I'm not saying his D will cure our D, but at least it's moving in the right direction.

I'm also against moving to a primarily 4-guard lineup, but I disagree with you about the Carter-Bolden pairing. Most of the good teams we will face do not play two traditional bigs (ironic when considering how Duke was historically at the vanguard of this change). Bagley is more adept than Bolden or Carter at staying with a smaller player on the perimeter. This is also pronounced in the zone, where one of the bigs has to rotate to the corner. Yes, we've had some struggles on D when Bagley and Carter play, but errors like over-helping or bad rotations are things that continue slowly get better as the season goes on. And as you've noted, there are problems pairing Carter and Bolden on offense. So unless the other team plays with two traditional bigs, I'm not in favor of the Carter-Bolden pairing.