MBB: Duke 90, Bucknell 60 Post Game Thread

[ClemmonsDevil]

Sporkz. Jim Sumner mentions some of what you posted in one of his latest articles. Coach Scheyer discusses playing some of his bench guys getting them up to speed. A good read.

GoDuke!

Hey thanks!

[Mtn.Devil.91.92.01.10.15]

This doesn't include two games they will play in Hawaii this week. Those will be high major opponents and there is a chance they will play three top ten KenPom teams - Purdue, Tennessee, and Kansas. Gonzaga will play 8 top 75 teams, as opposed to 4 for Duke.

To be clear, I don't think Duke's schedule is horrible or anything. But the last two years, the lack of Q1 opportunities hurt Duke when it came to seeding, and I think it will be an issue again this year.

Oh, I think it will be an issue again this year too. But that's because of an increasingly watered down and poor ACC schedule.

Out of conference games are scheduled way in advance. It's difficult to try and predict what years Duke will need to bolster with non-conference games so far in advance.

And again, I know we have some cupcakes on the schedule for the next few weeks, but so does literally every P5 school. As ClemmonsDevil points out - those games serve a purpose.

Also, to just be contrarian, filling your schedule with top tier opponents only helps if you come out victorious. If Duke has say 5 in conference losses and lost 5 out of conference games, that certainly doesn't help our seeding.

I like the new Arizona home and home. I guess I wouldn't mind one more similar series against a Kansas/Kentucky/Gonzaga. But I tend to trust the powers that be in scheduling.

I generally find our seeding to be mostly fair. I see more fault in our team for not winning more games rather than the algorithms that help with tournament seeds.

[House P]

I thought at one point Coach K tried to schedule games against solid mid-majors.

Back when the RPI was the primary sorting tool for NCAA tournament selection purposes, a team's overall rating was heavily weighted by their opponent's overall winning percentage. In those days, it made a difference whether Duke played a top team from the Patriot League like American (Torvik rank 251, 17 predicted overall wins) instead of a projected bottom of the pack team like Bucknell (Torvik rank 347, 9 predicted overall wins).

Now that the NCAA uses a more sophisticated method, both of these matchups would seen as Q4 games which a top team should win 99% of the time.

In my opinion, the biggest problems are that 1) the ACC is no longer an elite conference the way it was for so many years and 2) we now play 20 conference games. It is both more imperative to schedule tough OOC games and more of a challenge with fewer available slots.

Here is an unofficial* list of the top 20 AP teams sorted by the number of Q3/4 games projected by Torvik. Duke will likely have 2-5 more Q3/4 games (and, therefore, 2-5 fewer Q1/2 games) than most of their NCAA seeding competition.

AP Rank	Team	Conf	Q1/2 Games	Q3/4 Games	MTE
11	Gonzaga	WCC	12	17	Maui
10	FAU	AAC	14	17	ESPN Inv
9	Duke	ACC	16	15	Blue Devil Challenge
12	Miami	ACC	16	15	Baha Mar
7	Tennessee	SEC	18	13	Maui
14	Arkansas	SEC	18	13	Atlantis
17	Kentucky	SEC	18	13	Wildcat Challenge
19	Texas	B12	18	13	Saatva Empire
5	Uconn	BE	19	12	Saatva Empire
6	Houston	B12	19	12	Charleston
13	Texas A&M	SEC	19	12	ESPN Inv
18	Michigan St	B10	19	12	Acrisure Classic
3	Arizona	P12	20	11	Acrisure Classic
4	Marquette	BE	20	11	Maui
16	USC	P12	20	11	Rady Childrens
20	UNC	ACC	20	11	Atlantis
8	Creighton	BE	21	10	Hall of Fame Classic
2	Purdue	B10	23	8	Maui
15	Baylor	B12	23	7	NIT
1	Kansas	B12	24	6	Maui

*meaning I likely made a couple errors in compiling it.

[robed deity]

Back when the RPI was the primary sorting tool for NCAA tournament selection purposes, a team's overall rating was heavily weighted by their opponent's overall winning percentage. In those days, it made a difference whether Duke played a top team from the Patriot League like American (Torvik rank 251, 17 predicted overall wins) instead of a projected bottom of the pack team like Bucknell (Torvik rank 347, 9 predicted overall wins).

Now that the NCAA uses a more sophisticated method, both of these matchups would seen as Q4 games which a top team should win 99% of the time.



Here is an unofficial* list of the top 20 AP teams sorted by the number of Q3/4 games projected by Torvik. Duke will likely have 2-5 more Q3/4 games (and, therefore, 2-5 fewer Q1/2 games) than most of their NCAA seeding competition.

AP Rank	Team	Conf	Q1/2 Games	Q3/4 Games	MTE
11	Gonzaga	WCC	12	17	Maui
10	FAU	AAC	14	17	ESPN Inv
9	Duke	ACC	16	15	Blue Devil Challenge
12	Miami	ACC	16	15	Baha Mar
7	Tennessee	SEC	18	13	Maui
14	Arkansas	SEC	18	13	Atlantis
17	Kentucky	SEC	18	13	Wildcat Challenge
19	Texas	B12	18	13	Saatva Empire
5	Uconn	BE	19	12	Saatva Empire
6	Houston	B12	19	12	Charleston
13	Texas A&M	SEC	19	12	ESPN Inv
18	Michigan St	B10	19	12	Acrisure Classic
3	Arizona	P12	20	11	Acrisure Classic
4	Marquette	BE	20	11	Maui
16	USC	P12	20	11	Rady Childrens
20	UNC	ACC	20	11	Atlantis
8	Creighton	BE	21	10	Hall of Fame Classic
2	Purdue	B10	23	8	Maui
15	Baylor	B12	23	7	NIT
1	Kansas	B12	24	6	Maui

*meaning I likely made a couple errors in compiling it.

Yeah, I have no problem with some tune up games, but Bucknell and Southern Indiana? These teams are a couple of the worst teams in division 1. I'm sure it's somewhat hard to predict when scheduling though.

For me, this year, there's probably 1 or 2 too many "cupcakes." I like K's method of playing a champion from these mid majors. They are usually pretty good, and also hover in the 100-150 range.

[ClemmonsDevil]

Yeah, I have no problem with some tune up games, but Bucknell and Southern Indiana? These teams are a couple of the worst teams in division 1. I'm sure it's somewhat hard to predict when scheduling though.

For me, this year, there's probably 1 or 2 too many "cupcakes." I like K's method of playing a champion from these mid majors. They are usually pretty good, and also hover in the 100-150 range.

If you play a tougher team, your starters are going to be pulling 25 to 30 minutes. And the 8-11 guys are going to get fewer minutes. This is also the website where people talk about how tired Duke seems every time there is a loss. The schedule is set this way for a reason.

[lotusland]

Sporkz. Jim Sumner mentions some of what you posted in one of his latest articles. Coach Scheyer discusses playing some of his bench guys getting them up to speed. A good read.

GoDuke!

I just meant to play Davidson, Charleston, Charlotte or even Winthrop instead of Bucknell. The starters can work up a lather and the bench should still play a lot. Idk how good any of those teams I mentioned are this year but, generally speaking, they’re not cupcakes. Or we could play Richmond and VCU.

[ClemmonsDevil]

I just meant to play Davidson, Charleston, Charlotte or even Winthrop instead of Bucknell. The starters can work up a lather and the bench should still play a lot. Idk how good any of those teams I mentioned are this year but, generally speaking, they’re not cupcakes. Or we could play Richmond and VCU.

See previous posts.

[House P]

See previous posts.

Mostly serious question. Kansas, Kentucky, Arizona, Michigan St, UNC, and Villanova only play 5 true cupcakes (Quad 4 games), is there a reason why Duke benefits from playing 8?

EDIT: Upon closer review, 3 of Duke's Quad 4 games come in the ACC season, so there really isn't much of a disparity in Nov/Dec games.

[CDu]

For me, this year, there's probably 1 or 2 too many "cupcakes." I like K's method of playing a champion from these mid majors. They are usually pretty good, and also hover in the 100-150 range.

Coach K regularly played cupcakes too. In Coach K's last two normal nonconference seasons (2019-20 and 2021-22), here were his OOC matchups (Torvik ranks in parentheses):
2019: Colorado St (#113), Central Arkansas (#285), Georgia St (#131), SF Austin (#104), Winthrop (#169), Wofford (#151), Brown (#209), along with Kansas (#1), Michigan State (#5), Georgetown (#76), and Cal (#143)
2021: Army (#292), Campbell (#199), Gardner Webb (#150), Lafayette (#314), Citadel (#247), South Carolina St (#317), App St (#153), Elon (#255), along with 3 tough games in UK (#11), Gonzaga (#1), and Ohio St (#31)

If anything, this year's pre-conference schedule is perhaps slightly tougher than usual. We play Dartmouth (#242), Bucknell (#346), La Salle (#203), S Indiana (#352), Charlotte (#131), Hofstra (#116), and Queens (#235), along with Arizona (#9), Michigan St (#10), Baylor (#23), and Arkansas (#35). We have the ~7 Q3/Q4 games, but 4 legitimately tough games.

[House P]

Coach K regularly played cupcakes too. In Coach K's last two normal nonconference seasons (2019-20 and 2021-22), here were his OOC matchups (Torvik ranks in parentheses):
2019: Colorado St (#113), Central Arkansas (#285), Georgia St (#131), SF Austin (#104), Winthrop (#169), Wofford (#151), Brown (#209), along with Kansas (#1), Michigan State (#5), Georgetown (#76), and Cal (#143)
2021: Army (#292), Campbell (#199), Gardner Webb (#150), Lafayette (#314), Citadel (#247), South Carolina St (#317), App St (#153), Elon (#255), along with 3 tough games in UK (#11), Gonzaga (#1), and Ohio St (#31)

If anything, this year's pre-conference schedule is perhaps slightly tougher than usual. We play Dartmouth (#242), Bucknell (#346), La Salle (#203), S Indiana (#352), Charlotte (#131), Hofstra (#116), and Queens (#235), along with Arizona (#9), Michigan St (#10), Baylor (#23), and Arkansas (#35). We have the ~7 Q3/Q4 games, but 4 legitimately tough games.

2017-18 was the last season the RPI was used as the primary sorting tool by the NCAA tournament selection committee. Duke's schedule that season was a masterclass in gaming the RPI with a slate jammed packed with relatively weak teams who ended up with solid overall win/loss records.

Here were Duke's non-P6 games that season.

Team	KenPom	W/L	Result
South Dakota	81	26-9	+16
Furman	95	23-10	+29
Utah Valley	92	23-11	+30
Portland St	164	20-14	+18
St. Francis PA	214	18-13	+57
Evansville	134	17-15	+64
Southern	326	15-18	+17
Elon	248	14-18	+29
Average	169	20-14	+33

This year's non-P6 schedule includes just two teams projected by KenPom to have winning records, Hofstra (18-12) and Charlotte (16-14).

If you wanna know how messed up the RPI was, Duke's strength of schedule per the RPI was likely helped more by their home game vs 18-13 St Francis (#214 KenPom) than their road game vs 16-17 St. Johns (#74 KenPom). The NET has some issues, but it is light years ahead of the RPI.

[ClemmonsDevil]

Is there any way we can break this out into a schedule Construction thread? This is pretty fascinating and it is two pronged. I am discussing schedules construction and why you schedule how you do. There is a concurrent discussion of how a schedule can be judged that is more of a retrospective look. I think both of these are interesting. My experience is more in why a coach schedules the way they do but I like the other stuff too.