Duke MBB 2022-23: Phase 1

[scottdude8]

New stat from the social media sphere: we’ve allowed the second lowest points through 4 games since 1949.

https://twitter.com/dukembb/status/1...KbEv7XDfCGXGPA

When you consider that offense has tended to increase over time with the shortening shot clock and style of play, that’s pretty remarkable even considering less than stellar competition. This team’s defensive ceiling is sky high.

[DukieInBrasil]

New stat from the social media sphere: we’ve allowed the second lowest points through 4 games since 1949.

https://twitter.com/dukembb/status/1...KbEv7XDfCGXGPA

When you consider that offense has tended to increase over time with the shortening shot clock and style of play, that’s pretty remarkable even considering less than stellar competition. This team’s defensive ceiling is sky high.

Nice stat, and thanks for bringing it up! There are times when the defense looks mediocre, like a bunch of guys who haven't played together before working to figure it out. And then, there are times when the defense is truly remarkable, making it impossible for the opponent to do anything competently. And we still haven't fully integrated 2 guys who are supposed to be the defensive savants on the team. It could get really special!

[SkyBrickey]

Nice stat, and thanks for bringing it up! There are times when the defense looks mediocre, like a bunch of guys who haven't played together before working to figure it out. And then, there are times when the defense is truly remarkable, making it impossible for the opponent to do anything competently. And we still haven't fully integrated 2 guys who are supposed to be the defensive savants on the team. It could get really special!

Once fully healthy, I hope to see Blakes/Proctor/Whitehead/Mitchell/Lively all on the floor together for some extended minutes. Seems like this could be one of the most potent defensive 5s we've seen at Duke in many years based on reputations of Lively/Whitehead and what we've seen early in the year from the others. And there's not a big fall-off with Roach or Grandison or even Flip subbing back in...

[scottdude8]

Once fully healthy, I hope to see Blakes/Proctor/Whitehead/Mitchell/Lively all on the floor together for some extended minutes. Seems like this could be one of the most potent defensive 5s we've seen at Duke in many years based on reputations of Lively/Whitehead and what we've seen early in the year from the others. And there's not a big fall-off with Roach or Grandison or even Flip subbing back in...

I'm not so sure about extended minutes (just because it's going to be hard to keep our junior captain off the floor), but I'd love to see disruptive spurts of that lineup for sure. Now that Whitehead is back healthy, I would imagine the player who will lose the most minutes is Blakes. If I were Coach Scheyer, I'd use that to tell him to go all out in every minute he's on the court. If he plays for one 4 minute spurt each half, but in that spurt just becomes the most active and irritating defensive pest imaginable, he can be a major boost off the bench. We talk a lot about wanting a "microwave" offensive threat off the bench, but I think Blakes could have a similar effect with his defensive presence.

We'll see, though. I imagine PK80 will give us a solid look at what the complete rotation looks like.

[jv001]

I'm not so sure about extended minutes (just because it's going to be hard to keep our junior captain off the floor), but I'd love to see disruptive spurts of that lineup for sure. Now that Whitehead is back healthy, I would imagine the player who will lose the most minutes is Blakes. If I were Coach Scheyer, I'd use that to tell him to go all out in every minute he's on the court. If he plays for one 4 minute spurt each half, but in that spurt just becomes the most active and irritating defensive pest imaginable, he can be a major boost off the bench. We talk a lot about wanting a "microwave" offensive threat off the bench, but I think Blakes could have a similar effect with his defensive presence.

We'll see, though. I imagine PK80 will give us a solid look at what the complete rotation looks like.

Blakes looks like he's going all out now on every play. Like I said in another thread, he's a tough cookie. If he keeps hitting the three-pointers, he'll get his minutes, because the defense will always be there.

GoDuke!

[DukieTiger]

I had a realization about Phase 1 that I thought my be worth sharing.

While there are/were a handful of cupcakes in this phase and no road games, this still could end up being one of Duke’s toughest phases.

Caveat that the KenPom rankings are only marginally meaningful and always fluid but…

In Phase 1, Duke will end up playing 5 games against top 31 KenPom teams by December 6.

Duke only has three such games on their schedule for the rest of the season! If you include VT who just barely missed the cut, then it bumps up to 5 - but still… not many more top 30ish teams left on the schedule.

All that to say, there’s a pretty tough stretch starting tomorrow, with 3 top 20ish opponents in a 2 week span. The team will be uniquely battle tested at the end of this phase, and even though I trust the ACC to figure some things out and provide its share of tests, it’s still a nice thing to have a strong SOS during a down conference year.

[Dr. Maturin]

Taking a peek at some stats before Duke starts ACC play.

Offense: Duke is #18 in KenPom adjusted offensive efficiency. This number would probably be lower, but is influenced (I think) by the preseason projection of Duke having a top 10 offense. Offense has been a struggle at times, but I think there is reason to hope for improvement.

- Offensive rebounding- Duke is #1 in the country. This has really kept our offense afloat while we figure things out. I imagine we'll regress to the mean a bit, but this should be an elite OR team.

- 2-point shooting - For all the hand-wringing about shooting 3s, this is actually more interesting in some ways. Right now, Duke is making just 48.9% of its twos, which is 205th in the country. I say it's more interesting than 3pt% because I think there is more hope that this number will improve. Duke has tended to be a very good team at converting 2-point shots, even moreso in the one and done era. With Duke's size and rebounding ability, we figure to get a lot of shots around the basket. However, Duke's percentage right now is its worst since 2010. I don't know that this team will be elite at 2-point shooting, but I'd take the over on the current number. The ball movement we saw against Ohio State created a lot more easy buckets than we saw against Kansas and Purdue.

- Free throw shooting - Duke is shooting an excellent 76.7% from the line. Perhaps this gives some reason to hope that Duke's shooting will improve in other areas, but we'll see. (also, Duke has had the misfortune of seeing its opponents make a ridiculous 78.4% of their FTs, a number that is likely to go down)


Defense: Duke is #31 in KenPom adjusted defensive efficiency, and is in the top 100 in all of the 4 factors. I imagine the preseason projections are holding this number back a bit.

- Defensive rebounding - after a great start, Duke wasn't great on the glass against Purdue and Ohio State. Still, Duke is #77 in the country. This may not sound great until you recall that Duke has been a top-100 DR team just once in the KenPom era (2007). With Duke's size, hopefully this can be a strength relative to Duke teams of the past. A top 100 finish would be really good, in my opinion.

- Turnovers - Duke is #98 in the country, forcing TOs on 20.9% of possessions. That's pretty good. However, last year's team had some pretty good TO games early and then completely collapsed in that area. Duke had a forced TO rate of 20% or better 8 times, but only one of those came after January 12 (incidentally, AJ Griffin's first game in the starting lineup). Obviously the personnel is totally different this year. However, I wonder if some of our success in forcing TOs has to do with the nature of early season play in CBB. As teams improve over the course of the season, will we still be able to force turnovers? I'm skeptical, but who knows. If we can be average at both turnovers and defensive rebounding, that would be nice.

- Defensive free throw rate - Duke is 34th in the country, allowing 22.6 free throw attempts for every 100 field goal attempts. That's good, but not as good as last year's team. Duke's young bigs have struggled with foul trouble against good competition, so this is something to keep an eye on. This is typically a big strength for Duke defenses.


Other:

- Pace - Duke is averaging 65.9 possessions per game, which would be the slowest since 2010. This is surprising, both because Scheyer indicated that he wanted to run, but also because pace tends to slow over the course of the season in CBB. Last year, we played 11 games that went over 70 possessions by January 18, then had just 3 over the rest of the season.

3-point defense - Duke's opponents are taking 35.2% of their shots from 3. Only once in a season did K's teams allow that many 3s - the zone year of 2017-2018. It's still early, but Jon may be tweaking the defense to stay at home more. That may mean a few more 3s allowed but better defensive rebounding. Or it could just be statistical noise after 9 games. We'll see.

[azzefkram]

2-point shooting - For all the hand-wringing about shooting 3s, this is actually more interesting in some ways. Right now, Duke is making just 48.9% of its twos, which is 205th in the country. I say it's more interesting than 3pt% because I think there is more hope that this number will improve. Duke has tended to be a very good team at converting 2-point shots, even moreso in the one and done era. With Duke's size and rebounding ability, we figure to get a lot of shots around the basket. However, Duke's percentage right now is its worst since 2010. I don't know that this team will be elite at 2-point shooting, but I'd take the over on the current number. The ball movement we saw against Ohio State created a lot more easy buckets than we saw against Kansas and Purdue.

Yeah, Duke needs to limit its far 2s (Flip, Jeremy, and Tyrese are the worst offenders). It is currently making 31.3% (Torvik) or 32.8% (Hoop-Math). If you are going to shoot that poorly, step back and take a 3. At least you would be better compensated for the odd make.

As for close 2s, I'd expect Flip, Mark, and Dariq to get better which should help with those shots.

[Southgate0809]

Yeah, Duke needs to limit its far 2s (Flip, Jeremy, and Tyrese are the worst offenders). It is currently making 31.3% (Torvik) or 32.8% (Hoop-Math). If you are going to shoot that poorly, step back and take a 3. At least you would be better compensated for the odd make.

As for close 2s, I'd expect Flip, Mark, and Dariq to get better which should help with those shots.

I hate long 2's. There's nothing worse than shooting a shot with your foot on the 3-point line. When the player with the ball makes a move, and steps in half a foot on purpose for a 2-pointer with his foot on the line, it drives me crazy. Whitehead hit one of those a few games ago. When I was in 7th grade (which was about as far as my basketball career got), I hit a 2-pointer at the end of the game with my foot accidentally on the line, and we lost by 1-point. My coach told me after the game to always be aware of where my feet are. The more important part of the lesson being that having your feet back another 3 inches isn't going to decrease your chance of making the shot, so why not get the extra point. Kind of like Duke football throwing the ball 8 yards on 3rd and 10.

I think we have a chance to be an efficient 2-point scoring team. Lively finally gave us a glimpse of how efficiently he can score. We know Young can be efficient. If the offense is based around pick and rolls with Lively, Young posting up, or driving and dishing to either of them, that is very efficient offense and will lead to less 3 -point attempts, which will then lead to a higher 3-point percentage (based on the dichotomy we've seen so far of good 3-point shooting in games with fewer attempts vs. abysmal shooting in games with 20+ attempts).

[Kedsy]

Yeah, Duke needs to limit its far 2s (Flip, Jeremy, and Tyrese are the worst offenders). It is currently making 31.3% (Torvik) or 32.8% (Hoop-Math). If you are going to shoot that poorly, step back and take a 3. At least you would be better compensated for the odd make.

As for close 2s, I'd expect Flip, Mark, and Dariq to get better which should help with those shots.

Here's a table (numbers from Hoop-Math) showing our FG% at the rim, our FG% on two-point jumpers, and our % of shots which are two-point jumpers:

Year	FG% at rim	FG% 2pt J	% of 2pt J
2023	57.1%		32.8%		21.9%
2022	65.9%		38.9%		23.5%
2021	63.8%		40.2%		24.5%
2020	60.2%		38.3%		24.2%
2019	69.3%		35.9%		21.1%
2018	68.5%		38.3%		26.3%
2017	63.9%		40.9%		27.9%
2016	60.4%		38.6%		25.9%
2015	68.5%		38.2%		27.6%
2014	61.0%		39.9%		30.7%
2013	65.3%		37.0%		34.0%
2012	65.4%		35.8%		29.9%

As you can see, the number/percentage of our shots which are two-point jumpers are already low (2nd-lowest over 12 year period and significantly lower than average).

But you're right that our success rate on two-point jumpers is shockingly low, the lowest in at least the last 12 years, by a lot. Perhaps of more concern, our FG% at the rim is also the lowest in at least the last 12 years, also by a lot. The fact that our Elite Eight+ teams in the period (2022, 2019, 2018, 2015, 2013) were all over 65% at the rim makes our lowly 57% this year look even worse.

So we've not just been shooting poorly at threes this season, we've been shooting extremely poorly from all three levels. And while I don't have stats on this, you'd think shooting percentages would be highest in the pre-season part of our schedule (especially at the rim), which may not bode well (so far, we've faced four non-P6 teams plus one of the worst P6 teams in the country among our nine games; moving forward, we only have one game left against a non-P6 team among our 22 remaining regular season games).

Still, I have hope that all three levels will improve as the season wears on. Unfortunately, that hope isn't really based on anything.

[DavidBenAkiva]

Still, I have hope that all three levels will improve as the season wears on. Unfortunately, that hope isn't really based on anything.

Great stats and provides good perspectives on this team. One thing I've been tracking is how often Duke gets blocked on FG attempts. A LOT of Duke FG attempts at the rim are getting rejected. Kansas had 13 blocks against Duke. They don't have a huge team, either. Ernest Udeh is a good rim protector, but that's about it. That number has been trending down, thank goodness. Xavier had 0 blocks, Purdue only had 3 and tOSU had 4.

As for reasons for hope, I think that the emergence of Dereck Lively, if he really does emerge, will help with finishing at the rim. He had at least one impressive finish around the rim through contact against tOSU and is shooting better than 70% on close 2's this year according to T-Rank. More Lively dunks and pick-and-rolls and less Jeremy Roach and Jaylen Blakes layup attempts could help Duke get its close 2 FG% up. I also think Dariq Whitehead could help here, too, as he is strong enough to finish through contact at the rim.

My main takeaway from this team so far is that it is a work in progress. What we have seen is only a preview of what they could be. Some of it looks good (Filipowski overall, Young's ability to rebound and score inside, Roach's overall improvement, etc.). Some of it is concerning (Roach's inconsistency from 3 and team-wide issues with scoring for stretches).

[Kedsy]

If the offense is based around pick and rolls with Lively, Young posting up, or driving and dishing to either of them, that is very efficient offense and will lead to less 3 -point attempts, which will then lead to a higher 3-point percentage (based on the dichotomy we've seen so far of good 3-point shooting in games with fewer attempts vs. abysmal shooting in games with 20+ attempts).

I'm not convinced any of the above is true. If you are exploiting a defensive weakness that's one thing, but you can only do so many PnRs before the opponent makes adjustments, so a strong PnR game won't necessarily lead to fewer threes.

Regarding the idea that fewer threes leads to better 3pt shooting for this team, it may be true or it may not. Here's a table, sorted by % of our shots taken that were threes:

Opponent		eFG%	3p%	2p%	%threes
Xavier			54.6	45.5	51.2	20.4
Delaware		54.4	42.9	51.9	20.6
Ohio State		50.0	38.5	47.6	23.6
USC Upstate		49.3	33.3	48.9	30.9
Kansas			38.1	14.3	45.7	31.3
Purdue			37.9	10.5	48.7	32.8
Oregon St		30.8	17.2	35.5	48.3
Jacksonville		54.2	34.5	56.7	49.2
Bellarmine		58.2	40.0	55.0	63.6

Sure, three of our top four distance performances were our three games with the fewest threes, but our third and fifth highest 3pt performances were our two games with the most threes. (Those two games also happened to be our two best two-point shooting performances (and not because those two teams were the worst defenses, either; Jacksonville is actually the 30th-best team in the country at opposing eFG%)). So you might be right about the low attempts/high percentage correlation, or it might simply be low-sample noise.

[Southgate0809]

I'm not convinced any of the above is true. If you are exploiting a defensive weakness that's one thing, but you can only do so many PnRs before the opponent makes adjustments, so a strong PnR game won't necessarily lead to fewer threes.

Regarding the idea that fewer threes leads to better 3pt shooting for this team, it may be true or it may not. Here's a table, sorted by % of our shots taken that were threes:

Opponent		eFG%	3p%	2p%	%threes
Xavier			54.6	45.5	51.2	20.4
Delaware		54.4	42.9	51.9	20.6
Ohio State		50.0	38.5	47.6	23.6
USC Upstate		49.3	33.3	48.9	30.9
Kansas			38.1	14.3	45.7	31.3
Purdue			37.9	10.5	48.7	32.8
Oregon St		30.8	17.2	35.5	48.3
Jacksonville		54.2	34.5	56.7	49.2
Bellarmine		58.2	40.0	55.0	63.6

Sure, three of our top four distance performances were our three games with the fewest threes, but our third and fifth highest 3pt performances were our two games with the most threes. (Those two games also happened to be our two best two-point shooting performances (and not because those two teams were the worst defenses, either; Jacksonville is actually the 30th-best team in the country at opposing eFG%)). So you might be right about the low attempts/high percentage correlation, or it might simply be low-sample noise.

Definitely low sample size. I'm not sure that I really on the early games this year to tell us much anyway. We broke those teams down with our defense and were able to play a different game than most opponents will allow the rest of the year. Agreed that pick and rolls won't be an option to score if you run it every single time. That's why I also mentioned penetrating and dishing down low. We've shown good passing inside. If pick and rolls don't lead to a dunk by Lively, they might lead to a high percentage short jumper by Proctor. Lots of different ways we can score inside. I think Scheyer showed us a lot with his gameplan against Ohio State, even if the Buckeyes were vulnerable to that type of offense. To me, it seems like we have the personnel to allow for inside scoring in different ways. If the team can get better at executing, it doesn't seem crazy to me to think that would lead to more 2-point attempts and less 3-point attempts, and increase the % of makes on both.

[Kedsy]

If the team can get better at executing, it doesn't seem crazy to me to think that would lead to more 2-point attempts and less 3-point attempts, and increase the % of makes on both.

I agree it's not crazy. Though if the team can get better at executing, the percentages might go up regardless of how many attempts of each type we take.

[Southgate0809]

I agree it's not crazy. Though if the team can get better at executing, the percentages might go up regardless of how many attempts of each type we take.

Of course, although I haven't really felt like we took that many "bad" 3's in those games where we shot so poorly. Roach and Flip shoot them a little too early in the shot clock sometimes, but most of the time, we're shooting open 3's off good passes. They just weren't making them.

[Kedsy]

While I was doing that other research, I noticed something (a variation on an old chestnut we used to brag about back in the 90s and early 2000s):

Since Coach K's great recruiting class of 1982 (freshmen in the 1982-83 season), every recruited Duke player who stayed four years has been on an Elite Eight (or better) team except four guys (Dave McClure; DeMarcus Nelson; Greg Paulus; Marty Pocius). I feel bad for the four guys, but to me at least, that's a pretty amazing stat.

[AGDukesky]

As usual Kedsy great stats! I’m curious if there is a stat or easy way to measure 2pt efficiency integrating offensive rebounds. Those numbers are scary but I agree it feels like the opportunity to improve exists given two of our best players have been less than 100% and we are dealing with so many players who never played together. This team is a great rebounding team so I’m hoping that offsets some of the shooting woes- although even bad rebounding Duke teams typically suffer more on the defensive end.

[azzefkram]

Here's a table (numbers from Hoop-Math) showing our FG% at the rim, our FG% on two-point jumpers, and our % of shots which are two-point jumpers:

Year	FG% at rim	FG% 2pt J	% of 2pt J
2023	57.1%		32.8%		21.9%
2022	65.9%		38.9%		23.5%
2021	63.8%		40.2%		24.5%
2020	60.2%		38.3%		24.2%
2019	69.3%		35.9%		21.1%
2018	68.5%		38.3%		26.3%
2017	63.9%		40.9%		27.9%
2016	60.4%		38.6%		25.9%
2015	68.5%		38.2%		27.6%
2014	61.0%		39.9%		30.7%
2013	65.3%		37.0%		34.0%
2012	65.4%		35.8%		29.9%

As you can see, the number/percentage of our shots which are two-point jumpers are already low (2nd-lowest over 12 year period and significantly lower than average).

But you're right that our success rate on two-point jumpers is shockingly low, the lowest in at least the last 12 years, by a lot. Perhaps of more concern, our FG% at the rim is also the lowest in at least the last 12 years, also by a lot. The fact that our Elite Eight+ teams in the period (2022, 2019, 2018, 2015, 2013) were all over 65% at the rim makes our lowly 57% this year look even worse.

So we've not just been shooting poorly at threes this season, we've been shooting extremely poorly from all three levels. And while I don't have stats on this, you'd think shooting percentages would be highest in the pre-season part of our schedule (especially at the rim), which may not bode well (so far, we've faced four non-P6 teams plus one of the worst P6 teams in the country among our nine games; moving forward, we only have one game left against a non-P6 team among our 22 remaining regular season games).

Still, I have hope that all three levels will improve as the season wears on. Unfortunately, that hope isn't really based on anything.

Thank you for the good info. Jeremy is respectable on 2pt Js at 40%. Flip doesn't shoot a ton of them (23.3%) but what he does isn't fabulous (29.2%). Tyrese probably shoots too many (30%) for the amount he makes (33.3%). I agree that this probably isn't the biggest issue Duke has shooting the ball, but it does seem to be one with a fairly easy fix.

WRT FG% at the rim. I think that Flip can get better (56.5%) as he gains more experience. More shots for Dereck (72.2%) would help. I am not sure what to make of Mark (54.1%). I am hoping more experience will help, but he does seem to get his shot blocked an awful lot (the dreaded eye test) for a 6'8" guy.

[Kedsy]

As usual Kedsy great stats! I’m curious if there is a stat or easy way to measure 2pt efficiency integrating offensive rebounds. Those numbers are scary but I agree it feels like the opportunity to improve exists given two of our best players have been less than 100% and we are dealing with so many players who never played together. This team is a great rebounding team so I’m hoping that offsets some of the shooting woes- although even bad rebounding Duke teams typically suffer more on the defensive end.

I don't know how to what you're asking without going through the play by play (for years), something I'm not willing to do.

But I can say this about offensive rebounding-- under Coach K's system, teams that got a lot of offensive rebounds did very well in the NCAA tournament. Below is a table showing Duke's top 12 (including this season so far) offensive rebounding teams (since they started tracking offensive rebounds in 1987), along with their eventual NCAA tournament performance:

Year	OR%	NCAAT	Nat'l OR% rank
1999	44.3%	2	6
2023	42.1%	??	2
1990	40.9%	2	n/a
1988	40.5%	4	n/a
2010	40.3%	1	8
1998	39.7%	8	40
1992	39.5%	1	n/a
2004	39.2%	4	n/a
2018	38.6%	8	1
1996	38.3%	64	n/a
1991	38.0%	1	n/a
1994	38.0%	4	n/a

National rank is not available for a lot of those years, but I believe there has been a national trend toward fewer offensive rebounds over time (e.g., 1998 was #40 with a 39.7% OR rate, while 1998 was #1 with a 38.6%). Note also that national rank is among all Division I teams, rather than only P6 teams, which might be a better measure (but I didn't feel like parsing that out).

The above 11 represent all but six of Duke's Elite Eight (or better) teams (since 1987). The remaining teams were 2001 (37.0%; #63); 1989 (37.0%); 2015 (35.8%; #35); 2019 (35.3%; #15); 2022 (31.6%; #70); and 2013 (28.8%; #273).

So, as you can see, not including this season, 10 of the 11 teams in the table above got at least to the Elite Eight (and 8 of 11 got to the Final Four). And looking at it from another angle, of Duke's 17 Elite Eight and better teams since 1987, all but two (2022 and 2013) were either in the national top 40 in OR% or at least collected 37%+ of available offensive rebounds.

None of which means a whole lot, except to say for Duke teams, it appears to be better to grab a lot of offensive rebounds than not. Hopefully this year's Duke team continues gobbling up those offensive boards.

[MChambers]

- Pace - Duke is averaging 65.9 possessions per game, which would be the slowest since 2010. This is surprising, both because Scheyer indicated that he wanted to run, but also because pace tends to slow over the course of the season in CBB. Last year, we played 11 games that went over 70 possessions by January 18, then had just 3 over the rest of the season.

I think Lively and Whitehead getting more playing time should help with pace. Lively is a fantastic finisher on the fast break and supposedly can run all day long. Whitehead, from what I've seen, is very quick for his size and also should be a fantastic finisher, once he gets all the way back from his injury. On defense, both of them should help increase Duke's pressure, which also should help the pace, by causing turnovers and blocked shots.