collapse

* Resources


UDM 4

* SOTG


2017-18 Season SoG Tally
Rowsey9
Hauser5
Howard5
Anim2

'16-17 * '15-16
'14-15 * '13-14
'12-13 * '11-12 * '10-11

* Big East Standings

* Recent Posts

Recruiting as of 8/15/18 by We R Final Four
[Today at 11:43:07 AM]


[Cracked Sidewalks] Bethune-Cookman Preview by Herman Cain
[Today at 11:24:58 AM]


NM by Herman Cain
[Today at 09:44:24 AM]


Major NCAA Changes: Agents, Returns If Not Drafted by MU82
[Today at 08:14:58 AM]


New favorite Scoop GIF? by SaveOD238
[Today at 08:11:40 AM]


Top 100 players 2019 (TMW) by MuMark
[August 18, 2018, 05:59:28 PM]


Dwyane Wade to retire after 2018-19 season by Newsdreams Likes Chardonnay
[August 18, 2018, 12:45:58 PM]

Please Register - It's FREE!

The absolute only thing required for this FREE registration is a valid e-mail address.  We keep all your information confidential and will NEVER give or sell it to anyone else.
Login to get rid of this box (and ads) , or register NOW!

* Next up: A long offseason

Marquette
Marquette

Warriors



Wiki entry for 2017-18
21-14

Author Topic: Nice MUBB Preview  (Read 3356 times)

#bansultan

  • All American
  • *****
  • Posts: 24408
  • 18-19 Top 4 BE 5 Seed NCAA
Re: Nice MUBB Preview
« Reply #50 on: August 09, 2018, 05:07:51 PM »
Thanks TAMU.

Pretty incredible that last season we had 3 of the Top 5 Value Add seasons in MU ball since 2002.  Was a real shame to miss the tournament with that kind of Offensive talent. 

Sure is going to be an interesting season ahead.  Major upgrade defensively with Chardonnay, but we also have a major downgrade offensively with the loss of Andrew. 

I don’t think it’s that major. When you have so much offensive talent already, his impact isn’t as much as you would suspect.
My expectations for Marquette basketball in 2018-19: Top four in conference.  Five seed in NCAA.

Its DJOver

  • All American
  • *****
  • Posts: 697
Re: Nice MUBB Preview
« Reply #51 on: August 09, 2018, 05:32:04 PM »
I don’t think it’s that major. When you have so much offensive talent already, his impact isn’t as much as you would suspect.

I agree that it's not that major,  but do we really have that much offensive talent. 

Sam and Markus are both studs which means that we realistically only need one other major contributor, but the now sophomores arnt there yet. JC will make a bigger impact on the defensive end, and Ed hasn't played in over a year. Between all of them getting one or two to be real contributors should be a safe bet, especially throwing in the freshman, but it's still a concern. I'm still much more concerned about how big a jump the D can make, but a top 10 offensive shouldn't be a foregone conclusion. All IMO of course.

Floorslapper

  • All American
  • *****
  • Posts: 604
Re: Nice MUBB Preview
« Reply #52 on: August 09, 2018, 05:57:00 PM »
I don’t think it’s that major. When you have so much offensive talent already, his impact isn’t as much as you would suspect.

Guess we will see.  I feel Rowdy had a very positive impact on both Sam and Markus's offensive games.  The crazy range, and being so good with the ball as primary ball handler, distributor, dribble drive - all of that created better opportunities for Markus/Sam, than I feel they will see this year with Chardonnay. 

Not sure I see a very effective penetrator on the roster this year.  Markus generally was a shoot first, second and third if he did penetrate.  Drive/kick game will suffer without Rowdy and overall floor spacing.


Jay Bee

  • All American
  • *****
  • Posts: 6025
    • Late Night Hoops
Re: Nice MUBB Preview
« Reply #53 on: August 09, 2018, 06:17:30 PM »
I don’t think it’s that major. When you have so much offensive talent already, his impact isn’t as much as you would suspect.

Agreed (although I think our offense could be a little weaker, just because it was so good last year... some %'s are just tough comps).

We have a different kind of PG in JC who is experienced.. we have Morrow. We have a (hopefully) healthier Sam. Joey and Brendan are both capable -- Brendan at least of being similar to a Cain last year offensively, Joey of being a stud... and then you've got a trio of freshmen who are making the frosh to soph jump, which means good things when projecting offense.

Dr. Blackheart

  • All American
  • *****
  • Posts: 6351
Re: Nice MUBB Preview
« Reply #54 on: August 09, 2018, 08:21:05 PM »
Also, you mention a lot about how new players will help. How about how removing a terrible defensive player will help?

I tried to structure my commentary with the sensitivities where you could ratchet down the adjDE:  Height, rebounding steals in particular. I would be hard pressed for me to believe that MU could get to Nova's 94, for example. Or a Top 50 at around 97.5.  I can see a 101-102 which would put MU around a 100-110 rank, give or take. 

Defense is a team game.  It is easy to blame one guy but MU is not loaded with defensive specialists. I am also not a fan of Wojo's scheme as you well know is it spreads out the defense too far causing break downs, dumb fouls, easy points in the paint, poor rebounding, lower steal rates. Joe and Ed can help on some of these.  But, if Ed is out chasing on hedges north of the arc, that means he isn't grabbing a rebound that could shut down an opponents' possession versus extending it.

Freeway

  • Walk-On
  • *
  • Posts: 30
Re: Nice MUBB Preview
« Reply #55 on: August 09, 2018, 11:27:32 PM »
Guess we will see.  I feel Rowdy had a very positive impact on both Sam and Markus's offensive games.  The crazy range, and being so good with the ball as primary ball handler, distributor, dribble drive - all of that created better opportunities for Markus/Sam, than I feel they will see this year with Chardonnay. 

Not sure I see a very effective penetrator on the roster this year.  Markus generally was a shoot first, second and third if he did penetrate.  Drive/kick game will suffer without Rowdy and overall floor spacing.

I am hopeful that the improvement on offense this year will come from a pg who is more capable of finishing at the rim than Rowsey was, causing the defense to collapse more and leaving Markus and Sam with as good if not better shots than last year. I am also hopeful for better offensive rebounding out of this team (Morrow, Joey and Bailey) leading to easy 2s on putback’s and “jailbreak” 3s for our shooters.

brewcity77

  • All American
  • *****
  • Posts: 16990
  • Alan Bykowski
    • Brew City Ball
Re: Nice MUBB Preview
« Reply #56 on: August 10, 2018, 06:33:07 AM »
I’m not sure playing slower leads to better D. Sure, UVa was #1 last year, but Gonzaga was #1 two years ago and played at a top-100 pace. Most seasons there's a pretty fair spread of teams in the top-100, 100s, 200s, and 300s in terms of pace in the top-20 of defensive efficiency.

As far as Rowsey, I think Chartouny is a better fit for the rest of our team offensively. Andrew was a great offensive player & last season was fantastic, but he wasn't comparably speaking a distributor or a drive and kick guy.

Chartouny is far lower usage, which should push the ball more into the hands of Markus and Sam, who were both better eFG% guys than Rowsey. When Chartouny does shoot, he takes more twos than threes, which fits better with a drive & kick PG. Chartouny also had higher assist rates with less talent around him.

I don't think Chartouny will put up the kind of raw numbers or even efficiency numbers Rowsey did. But as a guy who does drive and prides himself on distribution, he's exactly what we need.

I also feel the roster consistency and maturation of the freshmen will help.

Jay Bee

  • All American
  • *****
  • Posts: 6025
    • Late Night Hoops
Re: Nice MUBB Preview
« Reply #57 on: August 10, 2018, 08:31:08 AM »
https://twitter.com/jbbauer612/status/1027268364031471616?s=21

Take out Rowsey’s weak 2FG% & MU’s 2FG% jumps up nicely

Dr. Blackheart

  • All American
  • *****
  • Posts: 6351
Re: Nice MUBB Preview
« Reply #58 on: August 10, 2018, 09:48:59 AM »
I’m not sure playing slower leads to better D. Sure, UVa was #1 last year, but Gonzaga was #1 two years ago and played at a top-100 pace. Most seasons there's a pretty fair spread of teams in the top-100, 100s, 200s, and 300s in terms of pace in the top-20 of defensive efficiency.

Are there exceptions to the tempo rule?  Yes.  That Zags team did it with height (two 7 footers) and was #1 in eFG%.  But, here is the list of the Top 10 defensive teams last year and their adj. tempo rank (where high is slow):

Virgina (351)
Cincinnati (330)
Michigan (324)
Texas Tech (247)
Syracuse (345)
Tennessee (282)
Clemson (274)
UCF (333)
Duke (93)
Michigan State (234)

So, one exception (Duke who was third in height).  All but UCF were NCAA teams. Eight were 5 seeds or better.  Seven were 3 seeds or better.

Can Wojo match his mentor?  Even his mentor is changing, employing the Cuse zone at times in the one and done era.  The recent rules changes have forced teams to adapt, as I said. This cannot be underestimated.

Tempo and certain positional height matter. MU is taller at PG this year. Is it enough to make MU elite? Or can the Duke formula work now that Wojo is in Year 2 of just his recruits for his system?  We'll know early.
« Last Edit: August 10, 2018, 10:15:25 AM by Dr. Blackheart »

Jay Bee

  • All American
  • *****
  • Posts: 6025
    • Late Night Hoops
Re: Nice MUBB Preview
« Reply #59 on: August 10, 2018, 10:42:04 AM »
Are there exceptions to the tempo rule?  Yes.  That Zags team did it with height (two 7 footers) and was #1 in eFG%.  But, here is the list of the Top 10 defensive teams last year and their adj. tempo rank (where high is slow):

Virgina (351)
Cincinnati (330)
Michigan (324)
Texas Tech (247)
Syracuse (345)
Tennessee (282)
Clemson (274)
UCF (333)
Duke (93)
Michigan State (234)

Not sure I'm on board with this. I don't know that there is a tempo rule or that great defenses without low tempos are exceptions (I see your 2017-18 above, but we can go back just one year and find 76, 84 and 26 in tempo all in the top ten of AdjD). I think going further is important -- something like average possession length... tempo is offense and defense.. now if your defense makes the other team's offense play much slower than they prefer, I think you're onto something. But, if you're a methodical offense that milks the clock, that impacts tempo but not sure it does much for your defense.

brewcity77

  • All American
  • *****
  • Posts: 16990
  • Alan Bykowski
    • Brew City Ball
Re: Nice MUBB Preview
« Reply #60 on: August 10, 2018, 12:31:56 PM »
Are there exceptions to the tempo rule?  Yes.  That Zags team did it with height (two 7 footers) and was #1 in eFG%.  But, here is the list of the Top 10 defensive teams last year and their adj. tempo rank (where high is slow):

Virgina (351)
Cincinnati (330)
Michigan (324)
Texas Tech (247)
Syracuse (345)
Tennessee (282)
Clemson (274)
UCF (333)
Duke (93)
Michigan State (234)

So, one exception (Duke who was third in height).  All but UCF were NCAA teams. Eight were 5 seeds or better.  Seven were 3 seeds or better.

Can Wojo match his mentor?  Even his mentor is changing, employing the Cuse zone at times in the one and done era.  The recent rules changes have forced teams to adapt, as I said. This cannot be underestimated.

Tempo and certain positional height matter. MU is taller at PG this year. Is it enough to make MU elite? Or can the Duke formula work now that Wojo is in Year 2 of just his recruits for his system?  We'll know early.

You're looking at one year. Look at 5 or 10. In 2011 there were 6 teams in the top-10 of defensive efficiency that were top-100 in tempo, and 8 in the top half. So by that season, better defense is played by teams that play fast.

There may be a slight correlation, but over time I don't think that's a defensible position (pun not intended).

Dr. Blackheart

  • All American
  • *****
  • Posts: 6351
Re: Nice MUBB Preview
« Reply #61 on: August 10, 2018, 01:07:22 PM »
Not sure I'm on board with this. I don't know that there is a tempo rule or that great defenses without low tempos are exceptions (I see your 2017-18 above, but we can go back just one year and find 76, 84 and 26 in tempo all in the top ten of AdjD). I think going further is important -- something like average possession length... tempo is offense and defense.. now if your defense makes the other team's offense play much slower than they prefer, I think you're onto something. But, if you're a methodical offense that milks the clock, that impacts tempo but not sure it does much for your defense.

I decided to use tempo for various reasons, some statistical, some practical.

First, was the rule changes. There are about 5-6% more possessions now in a game, on average. Possession length saw only about a 3% downtick. Thus, tempo provides more differentiation as a predictor variable.

Possession length and tempo are correlation. Possession length obviously impacts the number of possessions. Thus, I didn't want to throw in both in because of multicolinearity.  Tempo did just fine in the model.

Adj. Tempo, as you know is adjusted to the average team's possessions by KPom so it at least attempts to adjust to strength of opponents versus the raw possession numbers. 

Many here say offense affects defense. In fact, that seems to be the Duke philosophy.  That is a practical reason to try to consider and try to answer that question by including tempo.

I'd be careful about going back too many years for benchmarks or for inclusion in data sets. Teams are scoring more per 100 possessions. In 2013 the average adjusted offensive efficiency was 102.5. In 2018 it was 105.0. Again, the gradual rule changes and points of emphasis to enable more scoring, not just in the game but per possession.

Finally, there are always statistical exceptions to the rule.  You try to build in transformations to reduce error along the way. So, the value of tempo is not the same across teams, it's not a single estimate on impact to defense. This is like how you would measure price elasticity in business.  At a 2% price increase you may lose 2% of your volume, but at a 6% you may lose 10% as now more people cannot afford that 6% price.

I don't want to get too geeky here but I will mention that as folks are zeroing on exceptions or just one variable. We like to do this as it is simpler to understand, discuss, and communicate. I even did it in my post. However, while the model does need some work when/if I had time...but when you build multiplicative models, other variables effect each other versus singularly like in an additive linear regression model. These leads us to discussions like here.  i.e., Gonzaga played faster and yet was still great on defense.  Well, true, but they were also tall, controlled the boards, didn't foul as much to make up for it...or Marquette will be a great defensive team just because Rowsey is gone.

« Last Edit: August 10, 2018, 01:19:25 PM by Dr. Blackheart »

Dr. Blackheart

  • All American
  • *****
  • Posts: 6351
Re: Nice MUBB Preview
« Reply #62 on: August 10, 2018, 01:14:42 PM »
You're looking at one year. Look at 5 or 10. In 2011 there were 6 teams in the top-10 of defensive efficiency that were top-100 in tempo, and 8 in the top half. So by that season, better defense is played by teams that play fast.

There may be a slight correlation, but over time I don't think that's a defensible position (pun not intended).

I answered that question in my other post. The game has fundamentally changed due to gradual rule changes.  I didn't want to mix those up in comparisons, especially as we are talking Wojo and Rowsey.  Pomeroy will also caution users not to compare across years (although we all do it as his methodology is the same). Oh, and my time (I am not writing a thesis here).

If you have the time, go for it ;)

brewcity77

  • All American
  • *****
  • Posts: 16990
  • Alan Bykowski
    • Brew City Ball
Re: Nice MUBB Preview
« Reply #63 on: August 10, 2018, 01:20:43 PM »
I don't want to get too geeky here but I will mention that as folks are zeroing on exceptions or just one variable.

I'll admit I'm focusing on one variable because I think it's the most flawed of the variables and looking only at the top-10 of 2018 is zeroing in on an exception. 6 of the next 10 teams in 2018 (teams 11-20) were top half in tempo.

Dr. Blackheart

  • All American
  • *****
  • Posts: 6351
Re: Nice MUBB Preview
« Reply #64 on: August 10, 2018, 01:40:34 PM »
I'll admit I'm focusing on one variable because I think it's the most flawed of the variables and looking only at the top-10 of 2018 is zeroing in on an exception. 6 of the next 10 teams in 2018 (teams 11-20) were top half in tempo.

No worries...we all focus on the single issues that pop...and I did ask for feedback;)

If it helps put tempo in context, its impact is about 24% for the average team for the overall prediction.  So 76% is explained by other factors.  When I have time, I will try replacing tempo with defensive possession length (or finding the version I ran with it) to see if the model vastly improves.  As I said, I felt tempo provided more discrimination.

brewcity77

  • All American
  • *****
  • Posts: 16990
  • Alan Bykowski
    • Brew City Ball
Re: Nice MUBB Preview
« Reply #65 on: August 10, 2018, 02:04:19 PM »
No worries...we all focus on the single issues that pop...and I did ask for feedback;)

If it helps put tempo in context, its impact is about 24% for the average team for the overall prediction.  So 76% is explained by other factors.  When I have time, I will try replacing tempo with defensive possession length (or finding the version I ran with it) to see if the model vastly improves.  As I said, I felt tempo provided more discrimination.

I did a quick breakdown of the top-20 defenses over the past 5 years, so 100 teams. I broke them into three categories. The top-117 were considered "fast tempo", 118-234 was considered "medium tempo", and 235+ was considered "slow tempo". Here's the breakdown by category:

1-117: 23 teams
118-234: 23 teams
235-351: 54 teams

Still a limited sample, but I would say that tempo could bear out some of the predictor, but probably not very much of it. It does seem to be easier to have a high defensive efficiency with a slower pace, but playing medium or fast tempo doesn't preclude a team from having a good defense by any stretch. Obviously this is a very quick, generic breakdown with some awkward cutoffs (at least three teams fell exactly on cut-off numbers) that could be better evaluated by a deeper dive than the 10 minutes I spent looking at it and writing this post.

Also, I'm not sure if 20 is the ideal number, but the lowest defensive efficiency for a national champion of the past decade is #18 by North Carolina in 2009. However, even those numbers may be skewed against the season in general, because I recall Villanova in particular seeing their defensive efficiency numbers improve significantly by winning six straight games at the end of March. They may have been in the 30-40 range prior to the 2018 NCAA Tournament, but finished at #11.

One interesting correlation I did see, in 4 of the past 5 years, teams that played fast-tempo made up 50+% of the bottom-20 in the country. The only exception was 2014. In each of the past three seasons, more than 50% of the 20 worst teams in terms of defensive efficiency played at a pace ranked 117 or faster.

I didn't dive any deeper than that, but what I think might be the case is that teams that aren't particularly good sometimes try to speed the game up and play recklessly. They are often behind, so they have to play fast to chase, and when you aren't very good, that doesn't tend to end well. Teams that are good enough to play at a fast pace can still be defensively efficient, but that has to be the gameplan going in.

This year will definitely go a long way toward determining how well Wojo's scheme works. Tempo aside, I share many of your concerns regarding this team's defensive trends. We haven't been good and I'm not convinced that will change as simply as some like to think.

I will admit, the article that spawned this thread, particularly that stat about how our defense played with Howard on and Rowsey off the floor gives me some hope, as does the addition of guys that seem to add defensive value like Chartouny and Morrow and the maturation of guys that were active, willing defenders like Cain, Elliott, and John. The jury is definitely out, but those thoughts give me optimism when for the past 2-3 years, I've had steadily declining confidence in our defensive trend.

Okay...that's enough blathering. Getting well into TL;DR territory.

Jay Bee

  • All American
  • *****
  • Posts: 6025
    • Late Night Hoops
Re: Nice MUBB Preview
« Reply #66 on: August 10, 2018, 02:13:32 PM »
Still can’t get there, but what I think I heard is that last season there was strong correlation b/w AdjD & tempo, but three seasons ago that wasn’t the case. Don’t understand

Dr. Blackheart

  • All American
  • *****
  • Posts: 6351
Re: Nice MUBB Preview
« Reply #67 on: August 10, 2018, 03:13:59 PM »
Thanks Brew for your digging.  I did find the defensive possession output instead of tempo (again just last year's data set).  It is a very strong variable like tempo...and it doesn't affect the other variables too much.  However, as a remembered the residuals are higher when I compare predicted to actual (underpredicting).  I could work to lessen that but it really wouldn't change the findings much for the effort. 

As to going back in time to include more data, we'd have to adjust for the rule changes somehow.  The shot clock, we could use tempo but are more threes being launched at the end of the shot clock. Possession length?  Pre and post. The wider dashed circle, I guess investigate if the free throw rate went up...but that will be confused with the hand check and paint focus.  Time outs? Not sure on those changes.

In any regard, there are a lot of adjustments one has to make, and that usually introduces error as ultimately teams are changing their style of play to adapt.