The short bench meme—does it hold up to scrutiny?

[rocketeli]

The short bench meme—does it hold up to scrutiny?
Sometimes an idea becomes entrenched in people’s minds and they will repeat it without thought or citation, believing it must be true (historically this has been a problem for example in my profession, medicine.)
Similarly we have discussion and comments on DBR about Duke having a “short bench” as if it’s an established fact and/or an outlier or a bad thing. But are any of these things true? Does Duke have a short bench? And what is the optimal “length” of a bench anyway?
I think we can all agree that winning the NCAA basketball tournament is a respectable measure of success for a college basketball team. Using the somewhat arbitrary cutoffs of how many players saw double digit minutes, how many played at least half the game and how many went over thirty minutes, let’s look at the usage rates for players for the winning teams in the NCAA Division I men’s basketball tournament-for 2000-2017.
Year team double digits at least ½ of game over thirty minutes
2017 UNC 8 5 4
2016 Villanova 8 6 3
2015 Duke 7 7 3
2014 UConn 8 5 3
2013 Louisville 7 4 3
2012 Kentucky 6 6 4
2011 UConn 7 6 2
2010 Duke 5 5 5
2009 UNC 7 6 3
2008 Kansas (OT) 7 7 4
2007 Florida 7 6 4
2006 Florida 8 5 4
2005 UNC 8 6 3
2004 UConn 6 5 4
2003 Syracuse 8 6 3
2002 Maryland 7 7 3
2001 Duke 7 6 4
2000 Mich St 7 5 4
Note: In 2010 and 2015 Duke had a player log 9 minutes-just below our arbitrary cut off.
Interesting tidbits: 9 of 18 games in this span were won by teams that at some point have been in the ACC (although not necessarily when they won the tournament) and 9 of 18 games were won by three teams; UNC, Duke and UConn. Kentucky for all their visibility, hasn’t won since 2012 and Kansas hasn’t won since 2008. Kansas is the only team located west of the Mississippi River to win the championship during this span. In fact 14 of the 18 teams listed are located approximately 150 miles or less from the Atlantic Ocean.
On to the numbers;
(visualize caveats here—one game, yada, yada)
For players over double digits the distribution is 5-1, 6-2, 7-9 and 8-5. The modal number is 7.
For teams at the highest level of college basketball 9 or 10 man rotations are just not a thing.
For players playing at least 20 minutes the distribution is 4-1, 5-6, 6-8, and 7-3. The modal number is 6 only four teams did not play either 5 or 6 players for at least a half.
For 30+ minute players the distribution is 2-1,3-8, 4-8, 5-1. 3 and 4 tie for modal.
Generally speaking then 3 or 4 players will have a high usage rate on most of these teams.
If you look at individual games you can see that a variety of strategies can work—the year after Duke beat Butler getting only 17 minutes from its bench, UConn spread its minutes around such that only 2 players got 30+, and was also successful.
How does Duke bench use stack up, for example compared to UNC—since UNC is often held up as an exemplar of substituting?
UNC 2017 8, 5, 4 2009 7, 6, 3, 2005 8, 6, 3 average 7.67, 5.67, 3.34
Duke 2015 8, 8, 3 2010 5, 5, 5 2001 7, 6, 4 average 6.67, 6.34, 4
In two of Duke’s games players just missed the 10 minute cutoff-so let’s lower the cutoff to 9 minutes for each team.
UNC 2017 8, 5, 4 2009 7,6, 3 2005 8,6, 3 average 7.67, 5.67, 3.34
Duke 2015 8, 8, 3 2010 6, 6, 5 2001 7, 6, 4 average 7, 6.67, 4
You could speculate, a least looking at this small sample, that UNC does spread minutes around a small bit more than Duke does. Interestingly, Duke is a bit more variable as well—the 2015 Duke team spread minutes around more “significantly” i.e. 7 players over 20 minutes (22,32,35,23,37,21,21,9) than the 2017 UNC team (22,30,30,37,37,18,10,13), but the starters in the 2010 Butler game basically played 90%+ of available minutes.
So maybe the eye test is correct in two common assumptions at DBR—that the Duke coaching staff is more flexible than the Carolina one, and that Duke’s bench is shorter. However, it’s not 2 or 3 players shorter—more like 2/3 s of a player.
However—I think the most important point is that Duke is not really an outlier. Nobody’s playing more than 8 players and the mode is 7. For almost all teams 5 or 6 players get really significant minutes. People who wish for 9 or 10 man (or even 8.5) rotations are wishing for a thing that doesn’t exist for any elite team. In addition, many different teams (including UNC) have had success in a game playing only seven players significant minutes. Kentucky (Calipari of course is another coach is doesn’t go deep) in 2012, Kansas in 2008 (in OT no less) UConn in 2004 and Florida in 2007 didn’t use their bench beyond 5 or 6 players, and they all won.

[jv001]

I think Duke's bench is about 10 foot long. Where as Uncheat's bench is 15 foot long. Reason being, the cheats needed the extra length for Sean May and Kennedy Meeks. Rumor has it, the cheats bench will now return to 10 feet since those two players have graduated and passed all the paperless classes. They will however keep the longer bench in storage in case they have more May and Meeks type players. :cool:GoDuke!

[Mtn.Devil.91.92.01.10.15]

I think Duke's bench is about 10 foot long. Where as Uncheat's bench is 15 foot long. Reason being, the cheats needed the extra length for Sean May and Kennedy Meeks. Rumor has it, the cheats bench will now return to 10 feet since those two players have graduated and passed all the paperless classes. They will however keep the longer bench in storage in case they have more May and Meeks type players. :cool:GoDuke!

They put "McDonald's" back in the McDonald's All-Americans.

[grad_devil]

The short bench meme—does it hold up to scrutiny?
...
Division I men’s basketball tournament-for 2000-2017.
Year team double digits at least ½ of game over thirty minutes
2017 UNC 8 5 4
2016 Villanova 8 6 3
2015 Duke 7 7 3
2014 UConn 8 5 3
2013 Louisville 7 4 3
2012 Kentucky 6 6 4
2011 UConn 7 6 2
2010 Duke 5 5 5
2009 UNC 7 6 3
2008 Kansas (OT) 7 7 4
2007 Florida 7 6 4
2006 Florida 8 5 4
2005 UNC 8 6 3
2004 UConn 6 5 4
2003 Syracuse 8 6 3
2002 Maryland 7 7 3
2001 Duke 7 6 4
2000 Mich St 7 5 4
...

I've attempted to better format your first chart:

+------+-------------+---------------+-------------+-----------------+
| Year | team | double_digits | 1/2_of_game | over_30_minutes |
+------+-------------+---------------+-------------+-----------------+
| 2017 | UNC | 8 | 5 | 4 |
| 2016 | Villanova | 8 | 6 | 3 |
| 2015 | Duke | 7 | 7 | 3 |
| 2014 | UConn | 8 | 5 | 3 |
| 2013 | Louisville | 7 | 4 | 3 |
| 2012 | Kentucky | 6 | 6 | 4 |
| 2011 | UConn | 7 | 6 | 2 |
| 2010 | Duke | 5 | 5 | 5 |
| 2009 | UNC | 7 | 6 | 3 |
| 2008 | Kansas_(OT) | 7 | 7 | 4 |
| 2007 | Florida | 7 | 6 | 4 |
| 2006 | Florida | 8 | 5 | 4 |
| 2005 | UNC | 8 | 6 | 3 |
| 2004 | UConn | 6 | 5 | 4 |
| 2003 | Syracuse | 8 | 6 | 3 |
| 2002 | Maryland | 7 | 7 | 3 |
| 2001 | Duke | 7 | 6 | 4 |
| 2000 | Mich_St | 7 | 5 | 4 |
+------+-------------+---------------+-------------+-----------------+

[FadedTackyShirt]

Good analysis, thanks.

8 seems to be an optimal rotation. K previously rarely used all 13 rides on recruited players in order to reward 1 or 2 senior walk-ons.

Now needs to hedge against OADs, early entrants, or transfers with longer term plays (9 though 13).

[lotusland]

OP's post length far exceeds my attention span so I'll just assume it contains a lot of interesting stats and factoids about bench depth at Duke and elsewhere. Count me among those who would like K to "target" guys like Jack, Vrank and Javin to develop and remain engaged. Make it clear they have a role on the court whether it be 5, 7 or 10 mpg and stress what they should focus on to help the team succeed when they play. Obviously PT isn't guaranteed and expectations have to be met, yada yada. The recruiting landscape has changed and it seems clear that recruiting 4 year players does not = having 4 year players. Those 3 guys are good players who are D1 starters elsewhere. It's not realistic to expect them to watch from the bench for 3-years and hope they're ready to be role player as a senior. It also means some recruiting restraint. You just don't bring Bolden when you already have OAD Giles and 5th yr Amile coming back if you want a junior and senior Jeter around especially when you have OAD Carter coming in and your still recruiting OAD Bamba for next year.

If you follow IC meltdown threads like I do you'll notice how cheat fans bemoan the fact that Roy can't get OADs because he won't let talented freshmen have the reigns like K does. They also rip him for playing Seventh Woods, Robinson, Stillman White and Britt in close games instead of riding his studs to the finish line like K. I think there's a happy median where some OADs can
star while other players get some minutes to develop and are trusted to be the man when they are juniors and seniors. Just realized my post is almost as long or longer than OP's - oops.

[OldPhiKap]

Excellent work by the OP.

If you assume that 2010 is sort of an outlier (which it sure seems to be), and that 2015 is closer to what K typically does (I would argue it was tighter than normal after Sulu was off the team), K seems to be in the mainstream generally. Rotation of 7 (sometimes 8), and 3 over 30 (which is on the low side of the chart). We probably are a bit higher on the "play half a game" than average.

So, my take -- we generally play as many folks as other championship teams do as part of the typical rotation. Those on the bottom half of the rotation probably play a bit less than other schools do (or, put another way, our 4th - 6th players play more than the average school). Which fits K's philosophy -- have your best players on the court for the most minutes possible.

I would also bet that our 4th through 6th players are generally seniors who know the system (and thus get burn), or players who return the next year as one of the first three in the rotation. A good example would be Kennard, who was 6th in minutes in 2015-16 but first last year.

[FadedTackyShirt]

I would also bet that our 4th through 6th players are generally seniors who know the system (and thus get burn), or players who return the next year as one of the first three in the rotation.

Your theory also would apply to the limited utility of grad transfers specific to Duke. Thin yield/ROI for one year players who may not crack the rotation.

[Native]

I think there's a happy median where some OADs can
star while other players get some minutes to develop and are trusted to be the man when they are juniors and seniors.

Not to derail the thread, but this was always the plan: have one OAD a season to inject some fresh blood and firepower into a team of two-, three-, and four-year players.

This is what was pursued in 2011 (Kyrie), 2012 (Austin), and 2014 (Jabari). This was also the plan for 2015 — Okafor was the only surefire lock to go OAD — but we were a victim of our own success in Tyus and Justise also leaving early. We went back to a single OAD (Ingram) in 2016.

Interestingly, the single-OAD-per-year model hasn't resulted in the kind of tournament success one might expect, with no team with a single OAD going beyond the S16 (doing this from memory, so apologies if I'm off):

2010: Champions (0)
2011: S16 (1)
2012: R64 (1)
2013: E8 (0)
2014: R64 (1) (Not counting Rodney Hood)
2015: Champions (3)
2016: S16 (1)
2017: R32 (?)

I'll go a step further and say I'm not convinced the happy-median model does work, as nice as it sounds. You either go all-in with multiple projected OADs and hit the jackpot (2012 UK) or it backfires on you (2013 UK, 2017 Duke). Between us and UK — the two schools most perceived as OAD hotspots — we only have one championship apiece to show for it.

The single-OAD model has less variance. As another recent example, look at this year's Kansas squad with Josh Jackson. Same principle — no NCAAT banner to bring home, F4 or title.

The formula for success, then, is simple: good high school talent outperforming their projections and going OAD because they improved over the course of the season and the team was successful. We should start looking more at the Tyus Joneses and Justise Winslows of the world: players who aren't locks but are juuuuust on the border of going OAD at the start of the season and can make the argument after a successful season. If we win a bunch, they win, and they go early. If not, they stay and develop.

Perfect example: I think if you had asked most Cheat fans if Bradley was going to even consider the draft at the start of this season, they too would have laughed in your face.

[FerryFor50]

I think part of the narrative that got left out is *who* isn't playing.

With Duke, the issue isn't the short bench as much as it's the short bench with X number of McD AAs, 4-5 star recruits sitting. I think that's what rankles people the most.

Do teams like L'ville, UConn and even UNC bench their highly touted recruits? Or are the guys who aren't playing truly not worthy of playing in most cases?

Side note: I don't necessarily feel this way, and I also understand that a variety of factors play into a top recruit sitting. But this is the narrative I've gleaned from those who make the argument.

[OldPhiKap]

I think part of the narrative that got left out is *who* isn't playing.

With Duke, the issue isn't the short bench as much as it's the short bench with X number of McD AAs, 4-5 star recruits sitting. I think that's what rankles people the most.

Do teams like L'ville, UConn and even UNC bench their highly touted recruits? Or are the guys who aren't playing truly not worthy of playing in most cases?

Side note: I don't necessarily feel this way, and I also understand that a variety of factors play into a top recruit sitting. But this is the narrative I've gleaned from those who make the argument.

Yeah, but.

Grayson averaged 9.2 MPG, and 4.4 PPG, in 2014-15. An extreme example I guess, but still.

In any event, I'm not sure who it rankles but the players all know that they have nothing promised to them -- other than the chance to compete for minutes.

As far as benching highly-regarded recruits, the only one I can really think of in the last few years is Bolden -- who was injured pre-season and fouled whenever he came into the game. It's not that Bolden will not become an incredible player -- I think he will -- but he never had time to adjust to the speed of the game and tried to jump on a fast-moving train when he was finally cleared to play. I am glad he is coming back, because I think he will be a stud. Last year, though, I would agree with the decision to keep Amile on the floor in general and Harry was ahead of Bolden in terms of game readiness.

Who am I missing that just sat the pines as a "top recruit" that did not play? (seriously asking, not being snarky)

[FerryFor50]

Yeah, but.

Grayson averaged 9.2 MPG, and 4.4 PPG, in 2014-15. An extreme example I guess, but still.

In any event, I'm not sure who it rankles but the players all know that they have nothing promised to them -- other than the chance to compete for minutes.

As far as benching highly-regarded recruits, the only one I can really think of in the last few years is Bolden -- who was injured pre-season and fouled whenever he came into the game. It's not that Bolden will not become an incredible player -- I think he will -- but he never had time to adjust to the speed of the game and tried to jump on a fast-moving train when he was finally cleared to play. I am glad he is coming back, because I think he will be a stud. Last year, though, I would agree with the decision to keep Amile on the floor in general and Harry was ahead of Bolden in terms of game readiness.

Who am I missing that just sat the pines as a "top recruit" that did not play? (seriously asking, not being snarky)

I didn't say the perception made any sense. :D

[OldPhiKap]

I didn't say the perception made any sense. :D

Oh, agreed, not picking at you obviously.

[lotusland]

Not to derail the thread, but this was always the plan: have one OAD a season to inject some fresh blood and firepower into a team of two-, three-, and four-year players.

This is what was pursued in 2011 (Kyrie), 2012 (Austin), and 2014 (Jabari). This was also the plan for 2015 — Okafor was the only surefire lock to go OAD — but we were a victim of our own success in Tyus and Justise also leaving early. We went back to a single OAD (Ingram) in 2016.

Interestingly, the single-OAD-per-year model hasn't resulted in the kind of tournament success one might expect, with no team with a single OAD going beyond the S16 (doing this from memory, so apologies if I'm off):

2010: Champions (0)
2011: S16 (1)
2012: R64 (1)
2013: E8 (0)
2014: R64 (1) (Not counting Rodney Hood)
2015: Champions (3)
2016: S16 (1)
2017: R32 (?)

I'll go a step further and say I'm not convinced the happy-median model does work, as nice as it sounds. You either go all-in with multiple projected OADs and hit the jackpot (2012 UK) or it backfires on you (2013 UK, 2017 Duke). Between us and UK — the two schools most perceived as OAD hotspots — we only have one championship apiece to show for it.

The single-OAD model has less variance. As another recent example, look at this year's Kansas squad with Josh Jackson. Same principle — no NCAAT banner to bring home, F4 or title.

The formula for success, then, is simple: good high school talent outperforming their projections and going OAD because they improved over the course of the season and the team was successful. We should start looking more at the Tyus Joneses and Justise Winslows of the world: players who aren't locks but are juuuuust on the border of going OAD at the start of the season and can make the argument after a successful season. If we win a bunch, they win, and they go early. If not, they stay and develop.

Perfect example: I think if you had asked most Cheat fans if Bradley was going to even consider the draft at the start of this season, they too would have laughed in your face.

If you are consistently looking for 3 or more freshman to start each year your are all in on the OAD plan and you just have maintain the cycle year after year. That means you've got 3 bench players who got almost no PT and haven't advanced enough to start the following year. So now you need to go get 3 more OADs for next year which means your 3 bench players are now going to transfer. I won't argue that UK's model doesn't win but it's not a good fit for Duke university imo and selfishly, I'm just not interested in a minor league pro team from Durham. It's definitely a personal preference. A lot of folks dream about hitting the court with with Carter, Bamba, Duvall, Knox and Trent next year but I see a team I don't know or care about. Likewise the next round of non-student athletes that hope to replace them. Neither method either guarantees or precludes success but one is more in tune with Duke imo.

[FadedTackyShirt]

I think part of the narrative that got left out is *who* isn't playing.

With Duke, the issue isn't the short bench as much as it's the short bench with X number of McD AAs, 4-5 star recruits sitting. I think that's what rankles people the most.

Do teams like L'ville, UConn and even UNC bench their highly touted recruits? Or are the guys who aren't playing truly not worthy of playing in most cases?

Not just OADs, but the overall talent on K's Duke teams has risen too. Used to be that no K transfer from Duke ever went on to play in the NBA. Eliot Williams and Silent G broke that curse and Semi may too.

Never thought Seth Curry would be a future NBA player, but he's doing fine. Matt Cassel never started at 'SC, but he's had a solid NFL career (including a Pro Bowl). Programs that amass talent also engender resentment.

[lotusland]

Yeah, but.

Grayson averaged 9.2 MPG, and 4.4 PPG, in 2014-15. An extreme example I guess, but still.

In any event, I'm not sure who it rankles but the players all know that they have nothing promised to them -- other than the chance to compete for minutes.

As far as benching highly-regarded recruits, the only one I can really think of in the last few years is Bolden -- who was injured pre-season and fouled whenever he came into the game. It's not that Bolden will not become an incredible player -- I think he will -- but he never had time to adjust to the speed of the game and tried to jump on a fast-moving train when he was finally cleared to play. I am glad he is coming back, because I think he will be a stud. Last year, though, I would agree with the decision to keep Amile on the floor in general and Harry was ahead of Bolden in terms of game readiness.

Who am I missing that just sat the pines as a "top recruit" that did not play? (seriously asking, not being snarky)

Jeter comes to mind. I think he may have been ranked higher than Grayson. But my issue is more with how the next tier recruits are used - Vrank, White, DeLaurier, Semi. Duke really doesn't have much history with multiple OADs. We just haven't recruited at our current level long in this OAD era. The intersection of players who contribute as a freshman AND play 4 years is pretty small today. It includes Amile, Tyler and Matt but most likely it doesn't include Luke and Grayson now or going forward. The common thread seems to be above average defense from the jump. Past results don't support Nolan, Kyle, Mason, Grayson and Luke staying once they're projected in the first round anymore. So how do we recruit at a high level AND manage to keep some upper class men around these days?

[CDu]

Yeah, but.

Grayson averaged 9.2 MPG, and 4.4 PPG, in 2014-15. An extreme example I guess, but still.

In any event, I'm not sure who it rankles but the players all know that they have nothing promised to them -- other than the chance to compete for minutes.

As far as benching highly-regarded recruits, the only one I can really think of in the last few years is Bolden -- who was injured pre-season and fouled whenever he came into the game. It's not that Bolden will not become an incredible player -- I think he will -- but he never had time to adjust to the speed of the game and tried to jump on a fast-moving train when he was finally cleared to play. I am glad he is coming back, because I think he will be a stud. Last year, though, I would agree with the decision to keep Amile on the floor in general and Harry was ahead of Bolden in terms of game readiness.

Who am I missing that just sat the pines as a "top recruit" that did not play? (seriously asking, not being snarky)

Depends on your definition of "highly-recruited". The list of top-50 guys who didn't crack the rotation as freshmen since 2010 include:
Josh Hairston (#32) - also failed to reach the rotation as a sophomore
Michael Gbinije (#28)
Alex Murphy (#49) - also failed to reach the rotation as a sophomore
Semi Ojeleye (#33) - also failed to reach the rotation as a sophomore
Marques Bolden (#11)
Javin DeLaurier (#35)

Players who barely cracked the rotation as freshmen include:
Quinn Cook (#31)
Grayson Allen (#24, and only because of the dismissal of Sulaimon)
Chase Jeter (#16, and only because of the injury to Jefferson)

[FerryFor50]

Depends on your definition of "highly-recruited". The list of top-50 guys who didn't crack the rotation as freshmen since 2010 include:
Josh Hairston (#32) - also failed to reach the rotation as a sophomore
Michael Gbinije (#28)
Alex Murphy (#49) - also failed to reach the rotation as a sophomore
Semi Ojeleye (#33) - also failed to reach the rotation as a sophomore
Marques Bolden (#11)
Javin DeLaurier (#35)

Players who barely cracked the rotation as freshmen include:
Quinn Cook (#31)
Grayson Allen (#24, and only because of the dismissal of Sulaimon)
Chase Jeter (#16, and only because of the injury to Jefferson)

Thanks for doing the research CDu.

Of those 9 players, 4 transferred out of Duke and Bolden *nearly* transferred.

Also, if you like, you could possibly toss Derryck Thornton in there. He did play decent minutes, but he also didn't play as much as he or his handlers wanted, so he left.

[chriso]

Jeter comes to mind. I think he may have been ranked higher than Grayson. But my issue is more with how the next tier recruits are used - Vrank, White, DeLaurier, Semi. Duke really doesn't have much history with multiple OADs. We just haven't recruited at our current level long in this OAD era. The intersection of players who contribute as a freshman AND play 4 years is pretty small today. It includes Amile, Tyler and Matt but most likely it doesn't include Luke and Grayson now or going forward. The common thread seems to be above average defense from the jump. Past results don't support Nolan, Kyle, Mason, Grayson and Luke staying once they're projected in the first round anymore. So how do we recruit at a high level AND manage to keep some upper class men around these days?
Nothing to add, just wanted to say I enjoyed reading your posts this morning. Some interesting things to think about, and good points made. :)

[OldPhiKap]

Depends on your definition of "highly-recruited". The list of top-50 guys who didn't crack the rotation as freshmen since 2010 include:
Josh Hairston (#32) - also failed to reach the rotation as a sophomore
Michael Gbinije (#28)
Alex Murphy (#49) - also failed to reach the rotation as a sophomore
Semi Ojeleye (#33) - also failed to reach the rotation as a sophomore
Marques Bolden (#11)
Javin DeLaurier (#35)

Players who barely cracked the rotation as freshmen include:
Quinn Cook (#31)
Grayson Allen (#24, and only because of the dismissal of Sulaimon)
Chase Jeter (#16, and only because of the injury to Jefferson)

I think injuries played a key in a lot of those. Either being injured, or being behind Amile in an unexpected sixth season in Durham (for which I am very grateful).

If we look at a top-20 recruit as the bar to "top-rated" recruit (no slight to anyone else of course), we have Marques and Chase. Both had injury issues, and in regard to Chase he played well but was still behind a sixth-year Amile. I like Chase and wish he was back next year, but understand it's his decision and not mine.

Again, to be clear, I love all of these guys. Everybody runs their own race. As Quinn and Grayson show, patience can be a virtue. Josh, for example, got a heck of a lot more time than many on the board wished he did his senior year. I recognize that patience has become an old-fashioned virtue in today's world though.

In looking at that list, and taking injuries into account, it's still hard for me to see which of those players should have bumped someone above them in the rotation.

[sagegrouse]

Rocketeli,

Thanks for doing the work. I wonder if all schools shorten their rotations for the Final Four and the NC game. Do you think that season-long data would show something different? Or, maybe you have done that.

As I stated in another thread, there are two strategies to deal with the one-and-done occurrences at programs like Duke: (a) Recruit more players every year -- which we have done. (b) Expand the rotation with the view to developing players for the next season in the face of higher turnover rates -- which we have NOT done -- YET.

I also have a trailer for some research on roster turnover that I am doing and haven't posted yet. If we look at "rotation players" --- using ten minutes per game as line of demarcation -- our annual loss rate has been 50 percent or more each of the past three seasons. Only three times ever -- EVAH!-- have we reached 50 percent during K's tenure: not surprisingly, 1986, when the Dawkins class graduated; 1999, when we unexpectedly lost the Tres Amigos, Langdon and Burgess; and 2002, when JWill, Boozer and Dunleavy departed for the NBA after three years at Duke. Now it's happened three years in a row!

[CDu]

I think injuries played a key in a lot of those. Either being injured, or being behind Amile in an unexpected sixth season in Durham (for which I am very grateful).

If we look at a top-20 recruit as the bar to "top-rated" recruit (no slight to anyone else of course), we have Marques and Chase. Both had injury issues, and in regard to Chase he played well but was still behind a sixth-year Amile. I like Chase and wish he was back next year, but understand it's his decision and not mine.

Again, to be clear, I love all of these guys. Everybody runs their own race. As Quinn and Grayson show, patience can be a virtue. Josh, for example, got a heck of a lot more time than many on the board wished he did his senior year. I recognize that patience has become an old-fashioned virtue in today's world though.

In looking at that list, and taking injuries into account, it's still hard for me to see which of those players should have bumped someone above them in the rotation.

I don't think injuries were the issue for Gbinije, Murphy, Ojeleye, Allen, Hairston, DeLaurier, Cook, or Jeter. Injuries MIGHT have been the issue for Bolden this year. But I just think that Coach K plays a tight rotation, and if you aren't in the top 7 you risk being out of the rotation.

Had we been a team that was willing to consistently play 8-9 guys, I think that Gbinije, Cook, Allen, Jeter, and Bolden would have been in the rotation. I think Murphy, Hairston, Ojeleye, and DeLaurier would have still been the odd man out.

And I think looking only at top-20 is taking a myopic view of the situation. Top-50 guys play as freshmen for almost every program in the country. Including the top-tier ones. We have just gotten too comfortable with the idea that top-50 guys are afterthoughts because Coach K has an extreme preference for short rotations. Not all top-50 guys are ready, but they are ready MUCH more often than is evidenced at Duke.

[lotusland]

Right. These are guys who are in the rotation most of the time as freshmen even at non-Duke ACC schools. My point is not strictly to criticize K as much as emphasize that you cannot expect those guys NOT to transfer when they are nailed to the bench their sophomore year AND you already have another OAD signed at his spot for his junior year and you are still recruiting another. That's where Chase was this year. That transfer is going to happen 9 times out of 10 these days. Why even sign him to start with if you're going to do that? I know some people think freshmen and sophomores don't continue to improve but Jete showed me potential to be a VERY good junior or senior. But now his 2 years in duke's system are a waste. All because freshman Bamba might be better the junior Jeter AND we might be able to land another OAD better than senior Jeter the next year?

[Kedsy]

Depends on your definition of "highly-recruited". The list of top-50 guys who didn't crack the rotation as freshmen since 2010 include:

Josh Hairston (#32) - also failed to reach the rotation as a sophomore
Michael Gbinije (#28)
Alex Murphy (#49) - also failed to reach the rotation as a sophomore
Semi Ojeleye (#33) - also failed to reach the rotation as a sophomore
Marques Bolden (#11)
Javin DeLaurier (#35)

Players who barely cracked the rotation as freshmen include:
Quinn Cook (#31)
Grayson Allen (#24, and only because of the dismissal of Sulaimon)
Chase Jeter (#16, and only because of the injury to Jefferson)

You missed Matt Jones (5.7 mpg in ACC games his freshman year, 3.3 mpg if you don't count the three "line change" games).

But also, this is not a new phenomenon at Duke. Here's the list of top 50 guys who didn't crack the rotation as freshmen from 2003 to 2010:

Ryan Kelly (#14) (3.5 mpg in ACC games)
Taylor King (#27) (7.3 mpg in ACC games)
Brian Zoubek (#20) (5.3 mpg in ACC games)
Eric Boateng (#39) (0.4 mpg in ACC games)
Michael Thompson (#30) (0.8 mpg in ACC games)

Players who barely cracked the rotation as freshmen include:
Sean Dockery (#21) (9.2 mpg in ACC games)
Elliott Williams (#15) (inserted into rotation on 2/19, but before that 8.7 mpg in ACC games)

Also, Marty Pocius (#53) and Jamal Boykin (#60), if you want to stretch top 50 to top 60. And if you're willing to do that, it's pretty close to the same number in the older timeframe as in the more recent one.

That said, King, Boateng, Thompson, Williams, and Boykin all transferred too.

[jws]

The short bench meme—does it hold up to scrutiny?
Sometimes an idea becomes entrenched in people’s minds and they will repeat it without thought or citation, believing it must be true (historically this has been a problem for example in my profession, medicine.)
Similarly we have discussion and comments on DBR about Duke having a “short bench” as if it’s an established fact and/or an outlier or a bad thing. But are any of these things true? Does Duke have a short bench? And what is the optimal “length” of a bench anyway?
I think we can all agree that winning the NCAA basketball tournament is a respectable measure of success for a college basketball team. Using the somewhat arbitrary cutoffs of how many players saw double digit minutes, how many played at least half the game and how many went over thirty minutes, let’s look at the usage rates for players for the winning teams in the NCAA Division I men’s basketball tournament-for 2000-2017.
Year team double digits at least ½ of game over thirty minutes
2017 UNC 8 5 4
2016 Villanova 8 6 3
2015 Duke 7 7 3
2014 UConn 8 5 3
2013 Louisville 7 4 3
2012 Kentucky 6 6 4
2011 UConn 7 6 2
2010 Duke 5 5 5
2009 UNC 7 6 3
2008 Kansas (OT) 7 7 4
2007 Florida 7 6 4
2006 Florida 8 5 4
2005 UNC 8 6 3
2004 UConn 6 5 4
2003 Syracuse 8 6 3
2002 Maryland 7 7 3
2001 Duke 7 6 4
2000 Mich St 7 5 4
Note: In 2010 and 2015 Duke had a player log 9 minutes-just below our arbitrary cut off.
Interesting tidbits: 9 of 18 games in this span were won by teams that at some point have been in the ACC (although not necessarily when they won the tournament) and 9 of 18 games were won by three teams; UNC, Duke and UConn. Kentucky for all their visibility, hasn’t won since 2012 and Kansas hasn’t won since 2008. Kansas is the only team located west of the Mississippi River to win the championship during this span. In fact 14 of the 18 teams listed are located approximately 150 miles or less from the Atlantic Ocean.
On to the numbers;
(visualize caveats here—one game, yada, yada)
For players over double digits the distribution is 5-1, 6-2, 7-9 and 8-5. The modal number is 7.
For teams at the highest level of college basketball 9 or 10 man rotations are just not a thing.
For players playing at least 20 minutes the distribution is 4-1, 5-6, 6-8, and 7-3. The modal number is 6 only four teams did not play either 5 or 6 players for at least a half.
For 30+ minute players the distribution is 2-1,3-8, 4-8, 5-1. 3 and 4 tie for modal.
Generally speaking then 3 or 4 players will have a high usage rate on most of these teams.
If you look at individual games you can see that a variety of strategies can work—the year after Duke beat Butler getting only 17 minutes from its bench, UConn spread its minutes around such that only 2 players got 30+, and was also successful.
How does Duke bench use stack up, for example compared to UNC—since UNC is often held up as an exemplar of substituting?
UNC 2017 8, 5, 4 2009 7, 6, 3, 2005 8, 6, 3 average 7.67, 5.67, 3.34
Duke 2015 8, 8, 3 2010 5, 5, 5 2001 7, 6, 4 average 6.67, 6.34, 4
In two of Duke’s games players just missed the 10 minute cutoff-so let’s lower the cutoff to 9 minutes for each team.
UNC 2017 8, 5, 4 2009 7,6, 3 2005 8,6, 3 average 7.67, 5.67, 3.34
Duke 2015 8, 8, 3 2010 6, 6, 5 2001 7, 6, 4 average 7, 6.67, 4
You could speculate, a least looking at this small sample, that UNC does spread minutes around a small bit more than Duke does. Interestingly, Duke is a bit more variable as well—the 2015 Duke team spread minutes around more “significantly” i.e. 7 players over 20 minutes (22,32,35,23,37,21,21,9) than the 2017 UNC team (22,30,30,37,37,18,10,13), but the starters in the 2010 Butler game basically played 90%+ of available minutes.
So maybe the eye test is correct in two common assumptions at DBR—that the Duke coaching staff is more flexible than the Carolina one, and that Duke’s bench is shorter. However, it’s not 2 or 3 players shorter—more like 2/3 s of a player.
However—I think the most important point is that Duke is not really an outlier. Nobody’s playing more than 8 players and the mode is 7. For almost all teams 5 or 6 players get really significant minutes. People who wish for 9 or 10 man (or even 8.5) rotations are wishing for a thing that doesn’t exist for any elite team. In addition, many different teams (including UNC) have had success in a game playing only seven players significant minutes. Kentucky (Calipari of course is another coach is doesn’t go deep) in 2012, Kansas in 2008 (in OT no less) UConn in 2004 and Florida in 2007 didn’t use their bench beyond 5 or 6 players, and they all won.

I think excluding all players under double digits minutes misses much of the argument for a deep bench.
Roy Williams plays his depth, and IMO that depth pays off over the long run; not just in that season, but over multiple seasons.
A great example of this is Stilman White, who as a frosh averaged low single digits playing time over the regular season, but who was nevertheless given regular minutes with the rotation players in non-garbage situations. When Kendall Marshall went down with a broken hand in the NCAA Tournament, Stilman was ready and stepped up. He didn't do much in the way of scoring, but he took care of the ball and distributed it, ending up with 13 assists and only 1 turnover in the Sweet 16 and Elite 8.

Similarly, Luke Maye played very low minutes last season, but he was given fairly regular non-garbage minutes with rotation players and I think that's paid off this season, and Kenny Williams also got low but meaningful minutes during the regular season, and later on came in and hit a big shot and did a great job defending Malcom Brogdon in the ACC Tourney. Roy did pretty much the same this season with Seventh Woods.

There are a lot of ways to skin a cat. It's impossible to rationally argue with K's success the way he does things and the benefits/drawbacks of a deep bench are certainly up for debate, but I don't think arbitrarily ignoring anyone who doesn't average double digit minutes accurately frames the issue.

[Kedsy]

A great example of this is Stilman White, who as a frosh averaged low single digits playing time over the regular season, but who was nevertheless given regular minutes with the rotation players in non-garbage situations. When Kendall Marshall went down with a broken hand in the NCAA Tournament, Stilman was ready and stepped up.

Disagree. Amile Jefferson (2013), Elliott Williams (2009), Grayson Allen (2015), and Casey Sanders (2001), are all examples of Duke guys who hardly played who were "ready and stepped up" when the situation called for it. Non-garbage game minutes might be helpful but they simply aren't necessary for game readiness.

[sagegrouse]

Disagree. Amile Jefferson (2013), Elliott Williams (2009), Grayson Allen (2015), and Casey Sanders (2001), are all examples of Duke guys who hardly played who were "ready and stepped up" when the situation called for it. Non-garbage game minutes might be helpful but they simply aren't necessary for game readiness.

Pretty strong words, Keds. Court experience doesn't help players get used to the big-time spotlight at Duke? I wonder. It seems to me that our big guys, early in their careers, all suffer from "buck fever." When they are sent into the game, they all seem to commit a foul within 30 seconds. Somehow, they all get beyond that over time, which I attribute to game experience, and you may not.

[flyingdutchdevil]

Pretty strong words, Keds. Court experience doesn't help players get used to the big-time spotlight at Duke? I wonder. It seems to me that our big guys, early in their careers, all suffer from "buck fever." When they are sent into the game, they all seem to commit a foul within 30 seconds. Somehow, they all get beyond that over time, which I attribute to game experience, and you may not.

Same. I don't think players develop their game during games, but I think experience absolutely helps for the average recruit.

Also, "buck fever" sounds like those stupid "duck fuke" or "cuck farolina" shirts. I laughed when I read that. Because "buck fever" has a pretty funny connotation...

[CDu]

You missed Matt Jones (5.7 mpg in ACC games his freshman year, 3.3 mpg if you don't count the three "line change" games).

But also, this is not a new phenomenon at Duke. Here's the list of top 50 guys who didn't crack the rotation as freshmen from 2003 to 2010:

Ryan Kelly (#14) (3.5 mpg in ACC games)
Taylor King (#27) (7.3 mpg in ACC games)
Brian Zoubek (#20) (5.3 mpg in ACC games)
Eric Boateng (#39) (0.4 mpg in ACC games)
Michael Thompson (#30) (0.8 mpg in ACC games)

Players who barely cracked the rotation as freshmen include:
Sean Dockery (#21) (9.2 mpg in ACC games)
Elliott Williams (#15) (inserted into rotation on 2/19, but before that 8.7 mpg in ACC games)

Also, Marty Pocius (#53) and Jamal Boykin (#60), if you want to stretch top 50 to top 60. And if you're willing to do that, it's pretty close to the same number in the older timeframe as in the more recent one.

That said, King, Boateng, Thompson, Williams, and Boykin all transferred too.

Oh I totally agree that this is not a new phenomenon at Duke. It has been going on ever since Coach K started landing "super classes" on the regular (i.e., 1997). What has changed is not the approach, but rather the landscape. The early entry era as well as the explosion of TV coverage has done a few things:
1. Made more teams palatable to recruits, as almost every decent BCS school gets plenty of air time nationally
2. Reshaped players' views on "what is normal" for a collegiate career
3. Recruiting services and hype have increased as a result, because we "need" to know the big names quickly

Back in the 90s, there wasn't as much TV coverage nationally, which meant that fewer teams got air time. Thus, the national brand mattered more. Additionally, the "norm" was for players to stay in college for 3-4 years, and coming off the bench as freshmen was normal. So guys in the 30-50 range were more willing to sit and wait.

But that's not true anymore. Guys know the pecking order on the recruiting trail much moreso now than before. They have more peers, and know those peers better. And they are much more cognizant of how their peers are performing at other places. And they know that there are good opportunities to be had at other places. So whereas a guy like Alaa Abdelnaby would stay at Duke for 4 years and develop into a future NBA guy as a senior back in the late 80s, that same guy (maybe Jeter?) will be more likely to transfer today. A Nate James might stick around in the late 90s, but now (maybe a Gbinije?) is more likely to find playing time elsewhere.

And the proof is in the pudding. We rarely had transfers at Duke as late as entering the 1999 season. But after the 1999 season, Burgess (who would have likely started as a junior in 2000) transferred. And we've had a fairly steady stream ever since.

Guys in the 30+ recruiting range have not typically cracked the rotation as freshmen at Duke. The ones that have have typically stayed long-term. But in this decade, the ones that have not gotten PT right away have - more often than not - gone elsewhere.

It's just a part of the changing landscape of college basketball, and something that I'm sure the staff is trying to figure out as we speak. We need those solid-but-not-top-tier recruits to stick around for program continuity.

[Kedsy]

Pretty strong words, Keds. Court experience doesn't help players get used to the big-time spotlight at Duke? I wonder. It seems to me that our big guys, early in their careers, all suffer from "buck fever." When they are sent into the game, they all seem to commit a foul within 30 seconds. Somehow, they all get beyond that over time, which I attribute to game experience, and you may not.

I realize there is disagreement on this topic. I will say that during the 2013 NCAA Tournament, I conducted some locker room interviews for Mark Watson, and I happened to speak to both Amile Jefferson and Jeff Capel, and I spoke to both of them about this. They both believed that readiness to step up and play in "the big-time spotlight" was related to practice and development and not so much to game minutes. I'm sure Coach K would say the same.

This is not to say that all players who practice achieve the requisite development, nor that high-leverage game minutes don't help player development. I'm simply saying that players can get and be ready without previous substantial game minutes.

I think it's funny the OP brought Stilman White up as an example of the other side. I was happy every single time I saw UNC play Stilman White -- it meant they weren't playing a better player for those minutes. If he hadn't played earlier in 2012 (or in 2011), would he still have been ready to play when Marshall got hurt? I don't know. But I do know that the players I mentioned (Jefferson, Williams, Allen, Sanders) were all ready to play under similar circumstances at Duke, so maybe White would have, too.

[FerryFor50]

I think it's funny the OP brought Stilman White up as an example of the other side. I was happy every single time I saw UNC play Stilman White -- it meant they weren't playing a better player for those minutes. If he hadn't played earlier in 2012 (or in 2011), would he still have been ready to play when Marshall got hurt? I don't know. But I do know that the players I mentioned (Jefferson, Williams, Allen, Sanders) were all ready to play under similar circumstances at Duke, so maybe White would have, too.

Plus, what became of Stillman White?

He basically turned into Nick Pagliuca.

[OldPhiKap]

And I think looking only at top-20 is taking a myopic view of the situation. Top-50 guys play as freshmen for almost every program in the country. Including the top-tier ones. We have just gotten too comfortable with the idea that top-50 guys are afterthoughts because Coach K has an extreme preference for short rotations. Not all top-50 guys are ready, but they are ready MUCH more often than is evidenced at Duke.

Who said that? Certainly not me.

And in terms of short rotation, as the OP showed, only 6 of the last 18 teams to win the NC had an 8-player rotation (defined as double digit mpg).

I hearken back to the days before the number of scholarships were as limited as they are now, when UNC would stockpile talent on the bench. As freshman, they mainly sat for the most part unless they were super studs. K is able to do now what Dean did then -- recruit multiple deep classes in multiple years. (Or go back further, when freshmen were wholly ineligible). As the old joke went -- the only person who could hold Michael Jordan to ten points a game was Dean Smith.

The "problem" is not new. What is new, if anything, is the increased unwillingness to wait one's turn to move through a system. Whether that is good or bad, it's endemic to millennials and the increased view of players participating in a free market/free agent environment.

Regardless, it comes back to the same issue -- meritocracy. Should Chase have played more minutes at the expense of Amile? I'd say no.

[FerryFor50]

Who said that? Certainly not me.

And in terms of short rotation, as the OP showed, only 6 of the last 18 teams to win the NC had an 8-player rotation (defined as double digit mpg).

I hearken back to the days before the number of scholarships were as limited as they are now, when UNC would stockpile talent on the bench. As freshman, they mainly sat for the most part unless they were super studs. K is able to do now what Dean did then -- recruit multiple deep classes in multiple years. (Or go back further, when freshmen were wholly ineligible).

The "problem" is not new. What is new, if anything, is the increased unwillingness to wait one's turn to move through a system. Whether that is good or bad, it's endemic to millennials.

Regardless, it comes back to the same issue -- meritocracy. Should Chase have played more minutes at the expense of Amile? I'd say no.

The difference between Dean and K is that kids have realized they don't HAVE to stick around. They can transfer or leave early for the draft. Why rot on the bench if you can play elsewhere? Parity is so evenly distributed now that you don't need to be at a top flight program like UNC to shine. Semi is showing that now.

I don't think that's endemic to millenials; I think student athletes are realizing the system isn't set up to benefit them, so they will use the system as much as they can. And they're right to do it, IMO.

[CDu]

Plus, what became of Stillman White?

He basically turned into Nick Pagliuca.

Well, he started off as a Nick Pagliuca too. He was a 2-star player in high school, more a SG than a PG. He was never expected to be anything more than a backup. He got some notoriety for stepping in tolerably when Marshall got hurt. But he was in WAY over his head.

He went on his Mormon mission, and when he returned UNC had better players (Paige, Berry, Britt, now Woods) in front of him. So he just wasn't needed.

[OldPhiKap]

The difference between Dean and K is that kids have realized they don't HAVE to stick around. They can transfer or leave early for the draft. Why rot on the bench if you can play elsewhere? Parity is so evenly distributed now that you don't need to be at a top flight program like UNC to shine. Semi is showing that now.

I don't think that's endemic to millenials; I think student athletes are realizing the system isn't set up to benefit them, so they will use the system as much as they can. And they're right to do it, IMO.

I guess I don't see sitting on the bench as a freshman while an upperclassman plays as "rotting on the bench."

If after two seasons it's pretty clear that you're not going to break into the rotation for whatever reason, that's a whole different matter of course.

[jws]

Disagree. Amile Jefferson (2013), Elliott Williams (2009), Grayson Allen (2015), and Casey Sanders (2001), are all examples of Duke guys who hardly played who were "ready and stepped up" when the situation called for it. Non-garbage game minutes might be helpful but they simply aren't necessary for game readiness.

As I said, it's certainly arguable, but since Roy got to Carolina, he's got a significantly better NCAA Tourney record and his teams tend to do significantly better than K's late in the season, on average. I think that has a lot to do with Roy's bench depth, which sometimes costs him early on in the season, but pays off big time when injuries and/or foul trouble occur, and which also gives Roy more versatility to answer matchup problems and tends to have opponents more gassed at the end of games than Roy's players due to the pace Roy's teams play at.

[Kedsy]

And the proof is in the pudding. We rarely had transfers at Duke as late as entering the 1999 season.

Well, off the top of my head:

Bill Jackman ('83)
Greg Wendt ('83)
Crawford Palmer ('91)
Bill McCaffrey ('91)
Christian Ast ('92)
Joey Beard ('94)
Mike Chappell ('98)

And IIRC, Phil Henderson announced a transfer before changing his mind.

Maybe it's a lower rate than in recent years, but especially starting in the early '90s, it's not that much lower. (If you don't count grad-transfers Sean Obi and Rasheed Sulaimon, then we had 6 transfers in the eight year period from 2010 to 2017, while we had 5 transfers in the eight year period from 1991 to 1998; not that much of a difference.)

[FerryFor50]

I guess I don't see sitting on the bench as a freshman while an upperclassman plays as "rotting on the bench."

If after two seasons it's pretty clear that you're not going to break into the rotation for whatever reason, that's a whole different matter of course.

I don't disagree. I think it's admirable when a guy sticks around after not playing much in the first couple seasons.

But I also don't begrudge someone who wants an easier path to show what he can do on the court. Sometimes, the coaches aren't right.

[NM Duke Fan]

As I said, it's certainly arguable, but since Roy got to Carolina, he's got a significantly better NCAA Tourney record and his teams tend to do significantly better than K's late in the season, on average. I think that has a lot to do with Roy's bench depth, which sometimes costs him early on in the season, but pays off big time when injuries and/or foul trouble occur, and which also gives Roy more versatility to answer matchup problems and tends to have opponents more gassed at the end of games than Roy's players due to the pace Roy's teams play at.

Unlike some on this board, I think you raise some valid points worthy of futher intelligent, measured discusssion. With the mood some are in lately, you may instead get some vitriol. Different coaches have some strengths to their programs as well, Ol Roy does have certain successful qualities, Dad gum him!

[OldPhiKap]

I don't disagree. I think it's admirable when a guy sticks around after not playing much in the first couple seasons.

But I also don't begrudge someone who wants an easier path to show what he can do on the court. Sometimes, the coaches aren't right.

Agreed, I don't begrudge any player for making the decision that is best for them.

[CDu]

Regardless, it comes back to the same issue -- meritocracy. Should Chase have played more minutes at the expense of Amile? I'd say no.

I don't think it is as simple as that. We clearly have a continuity problem now. If Allen and Kennard both leave, then the returning juniors and seniors have a combined 125 minutes of basketball played at the collegiate level. And neither is likely to play more than 50 minutes this coming year either.

That can become a problem. Without that continuity, it makes coaching each season that more more difficult. It REALLY helps to have guys who have been there in the trenches to help the younger guys figure it out.

So while I totally get the meritocracy-based belief that we should play our best players as much, I think that it overlooks some key issues. It could very well be that, for the health of the program, the rotation needs to be a bit looser. The guys that we have historically relied on as 3-4 year program guys are becoming more and more of a flight risk. And if we keep losing all of our "3-4 year guys" to transfer, we aren't going to have the experience that is preferred come tournament time.

On top of that, there is the question of whether or not game reps matter in development. Coach K doesn't believe so, some do (I'm agnostic).

[CDu]

Well, off the top of my head:

Bill Jackman ('83)
Greg Wendt ('83)
Crawford Palmer ('91)
Bill McCaffrey ('91)
Christian Ast ('92)
Joey Beard ('94)
Mike Chappell ('98)

And IIRC, Phil Henderson announced a transfer before changing his mind.

Maybe it's a lower rate than in recent years, but especially starting in the early '90s, it's not that much lower. (If you don't count grad-transfers Sean Obi and Rasheed Sulaimon, then we had 6 transfers in the eight year period from 2010 to 2017, while we had 5 transfers in the eight year period from 1991 to 1998; not that much of a difference.)

I respectfully disagree. Over an 18-year period, we had 7 transfers. We've had 7 in the past 8 years, not including Obi and Sulaimon. And I'm not sure why you'd exclude Obi as playing time was obviously the reason he transferred.

If we want to compare the 1980-1998 transfer period to the 1999-2017 transfer period, it's decidedly more common in the more recent period.

[elvis14]

So how do we recruit at a high level AND manage to keep some upper class men around these days?

This is the question, the balancing act. There seems to be a tradeoff: experience or talent. Teams seem to do really well with guys that make a big jump their sophomore our junior year. Grayson Allen or Brice Johnson come to mind even MP3 his 5th year. Jason Tatum and Brandon Ingram both made a big jump midway through their freshman year and left. Heck Giles never even made the jump and left. I don't know what the answer is, it almost seems like you have to get lucky have a talented player not live up to expectations for a year or two (Hicks, J.Jackson). Plus, I think it helps to have some lower level recruits that expect to play as a Jr. and Sr. but guys like that can be hard to find or keep (think Javin or Semi).


Disagree. Amile Jefferson (2013), Elliott Williams (2009), Grayson Allen (2015), and Casey Sanders (2001), are all examples of Duke guys who hardly played who were "ready and stepped up" when the situation called for it. Non-garbage game minutes might be helpful but they simply aren't necessary for game readiness.

Every player is built differently. For some players game minutes aren't necessary for game readiness. For some guys they really are. I've been around sports for a long time playing and coaching and different players just react differently in a real game than they do in practice. So to say absolutely that they aren't necessary just isn't something I can agree with. Having Amile and Capel support Coach K's way of doing things is nice but you have to consider the bias of the source.

[johnb]

K's principle goal is NC's. To win an NC, a team probably needs a few NBA players, 4 or 5 excellent college players, chemistry, and luck. Chemistry and luck are tough to control. Excellent college players are easy for Duke to find. What's difficult and controllable? Recruitment of future NBA players. Giles and Tatum are NBA players. Allen and Kennard are great college players who may also be NBA players. With this roster, we were in the hunt for a NC. Didn't work out, but it might have. Does anyone seriously think that our OAD's, Giles and Tatum, weren't welcome at every D 1 school? Every one of so-called OAD's would have been welcome everywhere. No one turns them down.

We aren't going to win an NC because of Vranc, Jack, or Javin, though any one of those might play a prominent role on a future championship team. Their development would be a nice story, and would be nice to them and to those who know them, but we don't need all of them to thrive (or not transfer) for the team to succeed. To win an NC in 2018, we need to hope a couple of our guys develop into NBA players (Bolden and Frank, assuming Kennard and Allen are gone) and find a couple more NBA players to add to a complement of role players. We have already recruited a couple who are likely NBA players. Another one or two would be terrific for our chances of winning an NC.

Having said that, I agree that the recent teams have mattered less to me personally--the stars are clearly biding their time for the NBA, and some excellent college players are getting minimal PT. Without a multi-year human interest story, I simply care less for the team and the players. At this stage in his career, K is not going to try to convert our team into a midmajor and hope that lightning strikes before he retires. As is true for any strong leader, he will disregard the whining (including mine but also from any 4 or 5 star player who finds himself playing behind a lottery pick) and focus on his objective.

[CDu]

This is the question, the balancing act. There seems to be a tradeoff: experience or talent. Teams seem to do really well with guys that make a big jump their sophomore our junior year. Grayson Allen or Brice Johnson come to mind even MP3 his 5th year. Jason Tatum and Brandon Ingram both made a big jump midway through their freshman year and left. Heck Giles never even made the jump and left. I don't know what the answer is, it almost seems like you have to get lucky have a talented player not live up to expectations for a year or two (Hicks, J.Jackson). Plus, I think it helps to have some lower level recruits that expect to play as a Jr. and Sr. but guys like that can be hard to find or keep (think Javin or Semi).

It certainly helps to have a combination of talent and experience. Generally speaking, the teams that go deepest in the tournament tend to be experienced teams. There have been exceptions (2012, 2015), but for the most part the teams that get to the Final Four are pretty darn experienced teams. Obviously, it can be done with youth. It just tends to be less likely. And if you're gonna have youth, it had better be REALLY good youth


Every player is built differently. For some players game minutes aren't necessary for game readiness. For some guys they really are. I've been around sports for a long time playing and coaching and different players just react differently in a real game than they do in practice. So to say absolutely that they aren't necessary just isn't something I can agree with. Having Amile and Capel support Coach K's way of doing things is nice but you have to consider the bias of the source.

On this I agree. I think the quotes from Jefferson and Capel saying "you get ready through practice" are exactly what I'd expect them to say. Both are company men, trained by Coach K, who obviously believes that readiness is achieved through practice and not games. And it may (or may not) be true more often than not. But to say that it is uniformly true is almost certainly incorrect. Humans are - by definition - complex and individual. A cookie-cutter policy may work on average, but is not likely to work for everyone.

The trick is figuring out for which people it works and only bringing in those guys. But there's the rub: all of the guys we recruit have been full-time players and stars everywhere they've been before. So it's hard to know how they'll handle being stuck on the bench.

[Kedsy]

As I said, it's certainly arguable, but since Roy got to Carolina, he's got a significantly better NCAA Tourney record and his teams tend to do significantly better than K's late in the season, on average. I think that has a lot to do with Roy's bench depth, which sometimes costs him early on in the season, but pays off big time when injuries and/or foul trouble occur, and which also gives Roy more versatility to answer matchup problems and tends to have opponents more gassed at the end of games than Roy's players due to the pace Roy's teams play at.

Roy only has a "significantly better NCAA Tourney record" because of the last two seasons. His first 12 seasons at UNC, Roy's NCAAT record was 31-9 (.775) while K's record was 28-10 (.737), so Roy's was a little better but not much (also, he missed the NCAAT entirely in 2010, saving himself a potentially poor NCAAT record that season). If you believe UNC's NCAA performance the past two seasons was because Roy played a long bench, then fine, but I disagree.

As for doing "signficantly better than K's late in the season, on average," I think this is untrue. I'm not going to go back and count individual games at the end of the regular season, but the ACC tournament is pretty late in the season, right? And Coach K's ACCT record in the 14 seasons since Roy got to UNC (27-8, .771) is better than Roy's (22-11, .667).

[OldPhiKap]

Roy's players also were not burdened with, you know, going to classes and all.

[oakvillebluedevil]

If a short bench can't hold up scrutiny, it should have been built better *ducks*

[OldPhiKap]

Do vacated NCAA wins count as losses, or just as a "the games never happened"? Those tracking Roy's win percentages want to know.

Serious question.

[lotusland]

I realize there is disagreement on this topic. I will say that during the 2013 NCAA Tournament, I conducted some locker room interviews for Mark Watson, and I happened to speak to both Amile Jefferson and Jeff Capel, and I spoke to both of them about this. They both believed that readiness to step up and play in "the big-time spotlight" was related to practice and development and not so much to game minutes. I'm sure Coach K would say the same.

This is not to say that all players who practice achieve the requisite development, nor that high-leverage game minutes don't help player development. I'm simply saying that players can get and be ready without previous substantial game minutes.

I think it's funny the OP brought Stilman White up as an example of the other side. I was happy every single time I saw UNC play Stilman White -- it meant they weren't playing a better player for those minutes. If he hadn't played earlier in 2012 (or in 2011), would he still have been ready to play when Marshall got hurt? I don't know. But I do know that the players I mentioned (Jefferson, Williams, Allen, Sanders) were all ready to play under similar circumstances at Duke, so maybe White would have, too.

You interviewed 2 players who played as freshmen. Being "ready" to play is less than half of the challenge anyway imo. Chase Jeter might be supremely prepared to play next year but it won't improve Duke's depth at all. If Delaurier and White don't play significantly more next year and Duke signs another OAD 3/4 player for the following year look for one or both to transfer. So is there another path where these guys see a light at the end of the tunnel or does the revolving door continue?

[CDu]

Roy only has a "significantly better NCAA Tourney record" because of the last two seasons. His first 12 seasons at UNC, Roy's NCAAT record was 31-9 (.775) while K's record was 28-10 (.737), so Roy's was a little better but not much (also, he missed the NCAAT entirely in 2010, saving himself a potentially poor NCAAT record that season).

Even if you apply a first-round loss to the 2010 team, that's 31-10. And that's excluding the last two years. That's the definition of cherry-picking, right?


If you believe UNC's NCAA performance the past two seasons was because Roy played a long bench, then fine, but I disagree.

I think there is more nuance to this discussion than you are realizing. Directly speaking? I probably agree with you. I don't think that UNC was better this season because of playing a deeper bench THIS season. But there is an indirect, long-term effect that you appear to be ignoring. I would argue that the deeper bench keeps more kids engaged, feeling more invested in its success in each game. Now, not every kid will feel disengaged by being out of the rotation, but it's way more likely to feel disengaged when you aren't playing at all than when you are.

So by playing a deeper rotation and keeping the guys like Seventh Woods in the rotation, it is more likely that guys like Seventh Woods will stick around for 3-4 years and potentially become key contributors as upperclassmen.

Take Isiah Hicks for example. The dude was obviously not on the same level as Brice Johnson, James Michael McAdoo, and Kennedy Meeks as a freshman. At Duke, that would have meant he'd have been out of the rotation altogether. And he'd have been a transfer candidate. At UNC, he played every game and got 5-10 minutes per game. Not a ton, but enough to keep him engaged. So he stuck around. And he was a key contributor on this championship team.

Same thing for Theo Pinson. He was an extremely raw freshman wing on a team with Marcus Paige, Justin Jackson, JP Tokoto, and Joel Berry. At Duke, he sits. At UNC, he played 12 mpg. And he was a key contributor as a junior.

UNC was successful this year in large part because they were a veteran team. And they were a veteran team probably in large part because they play a lot of guys even as freshmen.

[kako]

Using the somewhat arbitrary cutoffs of how many players saw double digit minutes, how many played at least half the game and how many went over thirty minutes, let’s look at the usage rates for players for the winning teams in the NCAA Division I men’s basketball tournament-for 2000-2017.

Great work by the OP. I'm not throwing rocks, but just a question - where did this data come from? Wondering why the word "arbitrary" is included...

[Saratoga2]

Right. These are guys who are in the rotation most of the time as freshmen even at non-Duke ACC schools. My point is not strictly to criticize K as much as emphasize that you cannot expect those guys NOT to transfer when they are nailed to the bench their sophomore year AND you already have another OAD signed at his spot for his junior year and you are still recruiting another. That's where Chase was this year. That transfer is going to happen 9 times out of 10 these days. Why even sign him to start with if you're going to do that? I know some people think freshmen and sophomores don't continue to improve but Jete showed me potential to be a VERY good junior or senior. But now his 2 years in duke's system are a waste. All because freshman Bamba might be better the junior Jeter AND we might be able to land another OAD better than senior Jeter the next year?


No one mentioned that coach K himself said that the rotation needs to be deeper after the 65 point second half by South Carolina. I assume he was referring to what appeared to be fatigue of our team. I hope he acts on his statement this coming year and adjusts his playing rotation upwards. It can mean further player development, a reduced tendency to transfer and who knows, we may play fresher and better in the second half.

[jws]

Roy only has a "significantly better NCAA Tourney record" because of the last two seasons. His first 12 seasons at UNC, Roy's NCAAT record was 31-9 (.775) while K's record was 28-10 (.737), so Roy's was a little better but not much (also, he missed the NCAAT entirely in 2010, saving himself a potentially poor NCAAT record that season). If you believe UNC's NCAA performance the past two seasons was because Roy played a long bench, then fine, but I disagree.

As for doing "signficantly better than K's late in the season, on average," I think this is untrue. I'm not going to go back and count individual games at the end of the regular season, but the ACC tournament is pretty late in the season, right? And Coach K's ACCT record in the 14 seasons since Roy got to UNC (27-8, .771) is better than Roy's (22-11, .667).

Why would you cherry-pick the last two seasons out of the equation? If we're going to do that, why not exclude the past 5 seasons; then we get Roy with a 27-7 record compared to 19-8 for K?

But even with your cherry-picking the past two seasons out of the equation, Roy had 6 runs past the Sweet 16 in those 12 seasons, compared to 4 for K; and K had 3 years in which he didn't win a single NCAA Tourney game, compared to one for Roy. That goes to late season performance, and if one includes all seasons, as one should, the disparity is much greater.

Also, where late season is concerned, K's won significantly more games head to head against Roy overall, but Roy is 8-6 against K on the last day of the regular season, which includes 7-0 against K on the last day of the regular season with the ACC Regular Season title on the line for Roy's team(It would be 8-0 if Notre Dame had beaten Louisville on the last day this season).

[CDu]

No one mentioned that coach K himself said that the rotation needs to be deeper after the 65 point second half by South Carolina. I assume he was referring to what appeared to be fatigue of our team. I hope he acts on his statement this coming year and adjusts his playing rotation upwards. It can mean further player development, a reduced tendency to transfer and who knows, we may play fresher and better in the second half.

I don't know if he meant fatigue or foul trouble, but both are completely reasonable arguments for a single-game situation.

Getting a bit of rest throughout the game helps in two ways:
1. Reduces in-game fatigue, which is absolutely a real thing. These guys are typically gassed by the end of the game. If they aren't, that means they aren't playing hard enough or long enough.
2. Reduces risk of foul trouble causing your stars to play tentatively.

I don't know if #1 was an issue in the USC game, but #2 definitely was. We were forced to play zone for the end of the first half and a lot of the second half due to excessive foul trouble. Maybe if we have one or two more guys play for a few minutes in each half, our guys are better rested AND better prepared for foul trouble.

It's a difficult balance of course. Play lesser players TOO much and you hurt the team earlier in the game. Especially if foul trouble doesn't arise. But when foul trouble does arise, you probably need to be more willing to go to your bench.

Look at the title game. Think we stand a chance if they call whistles as quickly on us as they did both teams in that game? Notice what happened to Few's team when he tried to have his guys play through it. They either started playing tentative (Williams) or they plowed ahead with aggressive play (Collins) and picked up more fouls.

For much of the season, we were able to get away with a short rotation because our players generally avoided foul trouble. It did probably cost us in some of our losses this year. But in the USC game, we absolutely suffered from riding our guys through foul trouble, and USC took advantage.

[lotusland]

Who said that? Certainly not me.

And in terms of short rotation, as the OP showed, only 6 of the last 18 teams to win the NC had an 8-player rotation (defined as double digit mpg).

I hearken back to the days before the number of scholarships were as limited as they are now, when UNC would stockpile talent on the bench. As freshman, they mainly sat for the most part unless they were super studs. K is able to do now what Dean did then -- recruit multiple deep classes in multiple years. (Or go back further, when freshmen were wholly ineligible). As the old joke went -- the only person who could hold Michael Jordan to ten points a game was Dean Smith.

The "problem" is not new. What is new, if anything, is the increased unwillingness to wait one's turn to move through a system. Whether that is good or bad, it's endemic to millennials and the increased view of players participating in a free market/free agent environment.

Regardless, it comes back to the same issue -- meritocracy. Should Chase have played more minutes at the expense of Amile? I'd say no.

I agree with the question but I would answer that "yes" Amile can sit for a few more minutes a game on average while Chase plays. I think Amile was asked to play tentatively too often anyway due to foul trouble. I'd gladly trade a first half foul or two from Amile and a few minutes of Jeter for Amile not having to play "smart" or watch from the bench during crunch time. Playing smart = tentative or else you would just play smart all the time.

[jws]

And Coach K's ACCT record in the 14 seasons since Roy got to UNC (27-8, .771) is better than Roy's (22-11, .667).

Sorry; forgot to address this in the previous post.

K does indeed appear to put much more emphasis on winning the ACCT than Roy does(Roy has, in fact, made it clear he doesn't like the ACC Tourney and wishes he didn't have to play in it.), and his success there shows it.

Fact is, though, that winning the ACCT is really not very important for the elite teams, where the ultimate goal of winning a National Championship is concerned(especially compared to the ACC Regular Season Title); and hasn't been important since the NCAA Tourney went to at-large bids and seeding. That's why elite teams often sit starters in the ACC Tournament games, especially if they have nagging injuries. I don't think any of Roy's National Championships were won in years when they won the ACC Tourney, but all of them were in years when they won the ACC Regular Season title.

Personally, I think they should defenestrate the ACC Tourney and just add two or three games to the regular conference season, to make the schedules less unbalanced.

Impossible to say for certain, but I also think Duke playing 4 games in 4 days in the ACCT with their relatively short bench may well have cost them in the NCAA Tourney, fatigue-wise.

[Kedsy]

I respectfully disagree. Over an 18-year period, we had 7 transfers. We've had 7 in the past 8 years, not including Obi and Sulaimon. And I'm not sure why you'd exclude Obi as playing time was obviously the reason he transferred.

If we want to compare the 1980-1998 transfer period to the 1999-2017 transfer period, it's decidedly more common in the more recent period.

Well, if you want to compare Duke in the '80s, then I'll agree with you, but for a different reason. What I'm saying is the Duke transfer epidemic didn't start, as you said, in '99 (along with the Duke early entry epidemic). It started well before that. I would argue that it's been an issue at Duke every time we've had a loaded roster (meaning, a roster loaded with a large number of highly-regarded high school players) fed by large recruiting classes:

- We had a six-man recruiting class in '82, and two transfers in '82-'83.
- We had loaded rosters in the early '90s, and we lost four transfers from '91 to '94.
- We had a loaded roster from '97 to '99, and lost two transfers in that period.
- We had a five-man recruiting class in 2002 and lost one of them to transfer.
- In the three-year period from 2005 to 2007, we had 12 recruits, and lost three of them to transfer.
- We had a five-man recruiting class in 2011 and lost two of them to transfer.
- In the three-year period from 2014 to 2106, we had 15 recruits, and had three transfers out from 2015 to 2017 (four if you count Obi; five if you count Sulaimon).

Basically, the only transfers we had in Coach K's time that arguably weren't related to large recruiting classes/loaded rosters were Andre Sweet in 2001 (who reportedly failed out) and Olek Czyz in 2010. And in the mid-80s and early-00s, when our rosters weren't particularly deep, we didn't have many (or any) transfers. So I'm sticking with the idea that it's not a new phenomenon, except to the extent that Coach K is loading up his roster more in this one-and-done age than he usually did in the past.

All that said, I do believe the transfer rate nationally is up the past few years (though I don't know how much of that is fed by the recent trend of grad-transfers). At Duke, maybe we're losing a slightly increased number when we do have big recruiting classes. And I agree players seem less willing to bid their time and wait to start as juniors or seniors.

But even so, our transfers out in the past ten years all seemed to be players who were never likely to have big roles at Duke, except for E Williams, D Thornton, and maybe Gbinije. Maybe Jeter as a senior, but maybe not, sort of like Burgess. Even E Williams and D Thornton have reasonably close 20th century analogs in McCaffrey and Chappell.

So it's possible transfers are trending up a little, but it's in no way a new phenomenon. That's all I'm saying.

[niveklaen]

Great work by the OP. I'm not throwing rocks, but just a question - where did this data come from? Wondering why the word "arbitrary" is included...

http://www.duke-forum.com/forum/viewtopic.php?f=4&t=637&view=unread#unread

I don't know if the OP read this, but I also looked at the impact of depth a couple of months ago using a 30/20/10 cut-off and noted that that choice was arbitrary - better to use a %pt played by starters, but that math was beyond my limited ability...

...IMHO, its more likely that the OP thinks similarly to me than that they borrowed without attribution. This is unfortunate for them as thinking like me is not necessarily a good thing.

[Kedsy]

Why would you cherry-pick the last two seasons out of the equation?

I cherry-picked them out because you were making an observation about a 14-year period which isn't really true. Roy's teams have significantly outperformed K's in the NCAAT for the past two seasons, not 14. To me, this whole discussion is classic recency bias.


Also, where late season is concerned, K's won significantly more games head to head against Roy overall, but Roy is 8-6 against K on the last day of the regular season, which includes 7-0 against K on the last day of the regular season with the ACC Regular Season title on the line for Roy's team(It would be 8-0 if Notre Dame had beaten Louisville on the last day this season).

I'm not sure how the word "significantly" can be applied to an 8-6 record, nor how the fact that the regular season title may have been on the line tells me much about late season performance. I don't even see how looking at the last regular season rivalry game can be extrapolated to your assertion that Roy's teams "tend to do significantly better than K's late in the season, on average." If we're talking about late-season performance, why would that one game per season count more than ACC tournament games?


K does indeed appear to put much more emphasis on winning the ACCT than Roy does(Roy has, in fact, made it clear he doesn't like the ACC Tourney and wishes he didn't have to play in it.), and his success there shows it.

You think the players aren't trying to win the ACC tourney games? For that matter, are you really buying the lame rationalization that Roy isn't trying to win them? Late-season performance is late-season performance. If the debate is which of Roy's or K's substitution-pattern strategies help you late in the season, the ACC tournament is certainly relevant. Much more relevant than the data you chose to emphasize regarding the last regular season game.



Fact is, though, that winning the ACCT is really not very important for the elite teams, where the ultimate goal of winning a National Championship is concerned(especially compared to the ACC Regular Season Title)...

You're saying the regular season title is more important than the ACC tournament championship? Because it matters more to seeding, or what? Because, in general, while it may be debatable that the ACC regular season title from an imbalanced schedule is a more important achievement than winning the ACC championship, I'm not sure you'd win that debate.


I don't think any of Roy's National Championships were won in years when they won the ACC Tourney.
You are correct about this, but a sample size of 3 doesn't carry a lot of evidentiary weight with me.

[Kedsy]

I think there is more nuance to this discussion than you are realizing. Directly speaking? I probably agree with you. I don't think that UNC was better this season because of playing a deeper bench THIS season. But there is an indirect, long-term effect that you appear to be ignoring. I would argue that the deeper bench keeps more kids engaged, feeling more invested in its success in each game. Now, not every kid will feel disengaged by being out of the rotation, but it's way more likely to feel disengaged when you aren't playing at all than when you are.

Your point is a pretty good one. It's very possible guys like Hicks and Pinson would have transferred out of Duke if they hadn't played at all as freshmen, and those two guys certainly contributed to UNC's past two NCAAT journeys. On the other hand, Seventh Woods and Luke Maye aren't really good examples because (a) they weren't highly regarded enough to expect to play as freshmen; and (b) if they were the transferring type, they probably didn't play enough as freshmen to keep them there.

But overall, I'll concede your point, though I still don't think it's enough to win the debate about which substitution style is more likely to succeed.

[CDu]

Your point is a pretty good one. It's very possible guys like Hicks and Pinson would have transferred out of Duke if they hadn't played at all as freshmen, and those two guys certainly contributed to UNC's past two NCAAT journeys. On the other hand, Seventh Woods and Luke Maye aren't really good examples because (a) they weren't highly regarded enough to expect to play as freshmen; and (b) if they were the transferring type, they probably didn't play enough as freshmen to keep them there.

But overall, I'll concede your point, though I still don't think it's enough to win the debate about which substitution style is more likely to succeed.

Well, I didn't bring up Luke Maye, so not sure where that came from. As for Woods, I disagree a bit, although I acknowledge he is on the edge of the argument for sure. But first, he was a top-50 recruit. Those guys are expected to play at pretty much every school but Duke and Kentucky. While he only averaged 8 mpg, he played in every game, including the championship run. Even the 3-4 minutes per game he logged in the tournament is better than what our guys at that level would get.

The key to this is that Williams has had enough good recruits to make it work over the years. Next year (if Jackson goes pro, and especially if Jackson and Bradley go pro), they'll still have two seniors in Berry and Pinson (unless Berry ALSO bolts after the title). But because Williams suffered a rough recruiting class two years ago, he'll have only Luke Maye and Kenny Williams as juniors. Now, Williams is a very tough, athletic defender, and Maye can shoot. But both are off the periphery of the type of talent you need. That's why guys like Bradley and Woods are so critical. They're the guys that Williams has made his living on: the 20-60 recruits have been his gold. They are good enough to contribute right away but not usually good enough to bolt. So you get guys like Paige, Meeks, Johnson, Berry, Tokoto, etc providing production as juniors and seniors. And in years where you catch a break in slow development of guys like Jackson, Hicks, and Pinson, you can have what happened this year: elite high school talent finally tapping into their potential as juniors/seniors.

I don't necessarily believe one approach is "right" and one approach is "wrong." And certainly an approach can work for the wrong reasons (arguably like this year's UNC team) and can fail despite the right reasons (like this year's Duke team). I just think it is important to consider more than just the in-game consequences of the minutes rotation, and I don't think folks (including myself, prior to this year) have given that its due consideration.

[OldPhiKap]

I agree with the question but I would answer that "yes" Amile can sit for a few more minutes a game on average while Chase plays. I think Amile was asked to play tentatively too often anyway due to foul trouble. I'd gladly trade a first half foul or two from Amile and a few minutes of Jeter for Amile not having to play "smart" or watch from the bench during crunch time. Playing smart = tentative or else you would just play smart all the time.

Of course, when Amile sat, it was to give time to Harry (and occasionally Bolden until he picked up a few fouls). So maybe the better question is -- how many minutes should Chase get a game when our bigs are Amile, Harry, Marques and Jayson too?

We can't play everyone as many minutes as any of us want (myself included). And to be absolutely clear -- I liked Jeter's game and wish he was staying. I understand and respect his decision though, based on his assessment. I still don't see how he eeks out any more meaningful minutes though, and it's not like we blew anyone out after the GT game where we could put folks in to get mop-up experience. Especially so since Jayson, Marques and Harry were coming off injuries as freshman and thus EVERY minute they could get was paramount.

[English]

Of course, when Amile sat, it was to give time to Harry (and occasionally Bolden until he picked up a few fouls). So maybe the better question is -- how many minutes should Chase get a game when our bigs are Amile, Harry, Marques and Jayson too?

We can't play everyone as many minutes as any of us want (myself included). And to be absolutely clear -- I liked Jeter's game and wish he was staying. I understand and respect his decision though, based on his assessment. I still don't see how he eeks out any more meaningful minutes though, and it's not like we blew anyone out after the GT game where we could put folks in to get mop-up experience.

And this also ignores all the DBR armchair coaches who wanted to distribute handfuls of minutes upon Vrank and Javin to keep them happy and developing in-game. Your point stands, and I agree with it, but thought it merited mention that when you have a lot of players who seemingly can contribute (now and/or after developing in the future), there simply aren't enough minutes to go around.

Perhaps if Duke misses on all its recruiting targets for the next two-three seasons, then all the current guys can finally develop as they have down the road*.


*Sarcasm emoji

[jws]

I cherry-picked them out because you were making an observation about a 14-year period which isn't really true. Roy's teams have significantly outperformed K's in the NCAAT for the past two seasons, not 14. To me, this whole discussion is classic recency bias.



I'm not sure how the word "significantly" can be applied to an 8-6 record, nor how the fact that the regular season title may have been on the line tells me much about late season performance. I don't even see how looking at the last regular season rivalry game can be extrapolated to your assertion that Roy's teams "tend to do significantly better than K's late in the season, on average." If we're talking about late-season performance, why would that one game per season count more than ACC tournament games?



You think the players aren't trying to win the ACC tourney games? For that matter, are you really buying the lame rationalization that Roy isn't trying to win them? Late-season performance is late-season performance. If the debate is which of Roy's or K's substitution-pattern strategies help you late in the season, the ACC tournament is certainly relevant. Much more relevant than the data you chose to emphasize regarding the last regular season game.



You're saying the regular season title is more important than the ACC tournament championship? Because it matters more to seeding, or what? Because, in general, while it may be debatable that the ACC regular season title from an imbalanced schedule is a more important achievement than winning the ACC championship, I'm not sure you'd win that debate.


You are correct about this, but a sample size of 3 doesn't carry a lot of evidentiary weight with me.

As I clearly demonstrated, my observation was entirely true, even with your cherry-picked selection of seasons.

Where late season performance is concerned, Roy's 8-6 last day of the regular season record vs K; including his 7-0 record in those games with the ACC Title on the line for Roy; plus Roy's 42-31 lead in total NCAA Tourney wins; plus Roy's 8 Elite 8 appearances to K's 4; plus Roy's 5 Final Four appearances compared to K's 3; plus Roy's 4 National Championship Game appearances compared to K's 2; plus Roy's 3 National Championships compared to K's 2; plus Roy's single season without an NCAA Tourney win compared to K's 3, absolutely can, and do, have "significantly" attached to them.

As to the ACCT, of course players are trying to win their ACCT games. It's just that on the elite teams who already have a high seed sewn up, they're not going to put the same level of effort and concentration into it that they would in an important ACC Regular season game or, especially, the NCAA Tourney, and the coaches of those teams will often rest players for the NCAA Tourney; clearly demonstrating that winning the ACC Tourney, while cool, is not a huge deal.

There's simply no rational question that the ACC Regular Season title is far more important than the Tourney in every important way(for elite teams); from rankings to seeding for the NCAA Tourney to being a far more accurate measure of which team is the best. The only way in which the ACC Tourney title is superior to the Regular Season title is that it arbitrarily comes with an automatic NCAA Tourney bid, which is only important to teams in danger of not getting one. That Duke won the ACC Tourney and got a worse seed than UNC, who lost to Duke in the 2nd round; is due to the much greater importance of the ACC Regular Season standings, in which UNC finished 1st while Duke finished tied for 5th.

In the 14 years since Roy took over at Carolina, there have been 7 years in which the winner of the ACC Regular season did not also win the ACCT; and in 6 of those 7 seasons, the team that won the ACC Regular season went farther in the NCAA Tourney than the ACCT winner(The exception was 2015, when Notre Dame won the ACCT and went farther than UVA in the NCAA Tourney.).

Regarding Roy winning all three of his titles without winning the ACCT in that same season, neither of K's Championships since Roy got to Carolina were won in the same season as an ACCT Title either, so that's a sample size of 5 in 14 seasons, and K's title teams finished 1st and 2nd in the ACC Regular Season those years.

[Kedsy]

...clearly demonstrating that winning the ACC Tourney, while cool, is not a huge deal.

I don't think it's clear at all. In fact I think most people around here would disagree with your statement above.


There's simply no rational question that the ACC Regular Season title is far more important than the Tourney in every important way(for elite teams); from rankings to seeding for the NCAA Tourney to being a far more accurate measure of which team is the best.

I think there's plenty of rational dissent to your assertion and, again, I believe a LOT of people around here don't agree with you.



Regarding Roy winning all three of his titles without winning the ACCT in that same season, neither of K's Championships since Roy got to Carolina were won in the same season as an ACCT Title either, so that's a sample size of 5 in 14 seasons, and K's title teams finished 1st and 2nd in the ACC Regular Season those years.

Well, neither of them except the 2010 championship, which was won the same season as an ACCT title.

[sagegrouse]

Same. I don't think players develop their game during games, but I think experience absolutely helps for the average recruit.

Also, "buck fever" sounds like those stupid "duck fuke" or "cuck farolina" shirts. I laughed when I read that. Because "buck fever" has a pretty funny connotation...

Maybe, but the term has been around my whole life -- novice hunters getting super nervous when they see a deer and either freezing or shooting up the landscape or mistaking a doe for a buck. Or, a seven-foot freshman forgetting all elements of defensive positioning when they hit the court for the first few times in a Duke game.

[sagegrouse]

Roy's players also were not burdened with, you know, going to classes and all.

Actually, that's one positive indication of Tony Bradley's possible early exit. The schoolwork may have become more burdensome.

[JNort]

Maybe, but the term has been around my whole life -- novice hunters getting super nervous when they see a deer and either freezing or shooting up the landscape or mistaking a doe for a buck. Or, a seven-foot freshman forgetting all elements of defensive positioning when they hit the court for the first few times in a Duke game.

I have to disagree, at least from personal experience. Game time is equally important.

[kako]

http://www.duke-forum.com/forum/viewtopic.php?f=4&t=637&view=unread#unread

I don't know if the OP read this, but I also looked at the impact of depth a couple of months ago using a 30/20/10 cut-off and noted that that choice was arbitrary - better to use a %pt played by starters, but that math was beyond my limited ability...

...IMHO, its more likely that the OP thinks similarly to me than that they borrowed without attribution. This is unfortunate for them as thinking like me is not necessarily a good thing.

In any case, there are lies, damn lies and statistics. My eyeball test says that an 8-deep team is important. One guard, one forward and one big man for the rotation. At least one of them (probably the guard) should be able to score reasonably. They average 15 mpg.

That being said, I'm not questioning K with his 5 titles. I'd just feel better if Duke had an 8-deep team.

[lotusland]

Of course, when Amile sat, it was to give time to Harry (and occasionally Bolden until he picked up a few fouls). So maybe the better question is -- how many minutes should Chase get a game when our bigs are Amile, Harry, Marques and Jayson too?

We can't play everyone as many minutes as any of us want (myself included). And to be absolutely clear -- I liked Jeter's game and wish he was staying. I understand and respect his decision though, based on his assessment. I still don't see how he eeks out any more meaningful minutes though, and it's not like we blew anyone out after the GT game where we could put folks in to get mop-up experience. Especially so since Jayson, Marques and Harry were coming off injuries as freshman and thus EVERY minute they could get was paramount.

I argued at the time and still believe we didn't need Bolden once Amile came back. I don't see how you could look at the lineup including Giles, Bolden, Amile and Tatum imagine all of those guys would get significant PT much less Jeter. Why bring in a OAD player to maybe play 5-10 minutes at best for only a year? Just because you can sign a guy doesn't mean you should. I'm glad Bolden is back fo another year but I'd still trade his one more year for 2 more From Jete.

[CDu]

I argued at the time and still believe we didn't need Bolden once Amile came back. I don't see how you could look at the lineup including Giles, Bolden, Amile and Tatum imagine all of those guys would get significant PT much less Jeter. Why bring in a OAD player to maybe play 5-10 minutes at best for only a year? Just because you can sign a guy doesn't mean you should. I'm glad Bolden is back fo another year but I'd still trade his one more year for 2 more From Jete.

There is no guarantee that Jeter stays even if Bolden wasn't here. In fact, given that Jeter apparently decided to transfer during the season (when Bolden was obviously still undecided on his return), it suggests Jeter would have been gone anyway.

[Indoor66]

The schoolwork may have become more burdensome.

Obvious hyperbole! It IS UnCheat we are talking about.

[RPS]

As I clearly demonstrated, my observation was entirely true, even with your cherry-picked selection of seasons.
Small sample size. Insanely small sample size. The sample size isn't remotely large enough to draw meaningful conclusions. Small. Sample. Size.