MUScoop

MUScoop => Hangin' at the Al => Topic started by: CrackedSidewalksSays on August 06, 2008, 11:15:03 AM

Title: [Cracked Sidewalks] Big East Preview - Part Two
Post by: CrackedSidewalksSays on August 06, 2008, 11:15:03 AM
Big East Preview - Part Two

Written by: noreply@blogger.com (Henry Sugar)

Back now with Part Two of the Big East Preview.  In case you missed Part One, you can find it at this link (http://www.crackedsidewalks.com/2008/08/big-east-preview-part-one.html).  If you want more details on the factors I'm using, you can check it out here (http://www.crackedsidewalks.com/2008/08/news-and-notes-and-start-to-big-east.html).

The data I used is taken from Pomeroy's site (http://kenpom.com/), where I extracted conference data only and worked with that.  As a reminder, the factors under consideration were:
Before we move onto the the top eight teams (Connecticut, Georgetown, Louisville, Marquette, Notre Dame, Pittsburgh, Syracuse, West Virginia), I wanted to make sure that something was clarified.

Projections are forecasts based on limited data.  They are inherently inaccurate. Don't focus too much on any predictions regarding specific Wins/Losses or where the selected team is ranked in the preview.  The most important value (IMO) is in the areas that I regard as statistical outliers, or where a team may have a red flag or potential bonus.

Clear?  Onto the preview

Competing for a BET first round bye
The next six teams are all fairly close, and I'd expect each team to be somewhat in contention for a spot in the top four

#8 - Syracuse Projection - 10-8. Expect Syracuse to be better this year.  They're definitely talented, but are they too young to compete for a top spot?  Expectations for the 'Cuse may be understated.

#7 - Notre Dame Projection - 11-7.  Despite the big advantage of returning almost everyone except the productive Rob Kurz, I see some major red flags for ND.

#6 - Pittsburgh   Projection - 11-7. As mentioned at the beginning of this section, any of the teams in here could finish between 8th and 3rd.  So why is Pitt down at #6?  Because I don't like Pitt!  Refreshing candor from Cracked Sidewalks, and hey... Pitt is above ND.

#5 - Georgetown   Projection - 12-6. Conventional wisdom is that Georgetown will be significantly down this year.  I'm less bearish on the Hoyas' chances than many, despite the two big red flags listed above.  The biggest reason is that GU will still have a decent number of returning contributors, as well as a huge influx of talent.

#4 - West Virginia   Projection - 12-6. This is a team that could just as easily finish 8th as well.  I just think that Huggins will get West Virginia to play more consistently and they won't be as bad.

#3 - Marquette   Projection - 12-6. Before anyone gets on me for making this a homer pick, I'll say that this is a team that could just as easily finish 8th.  I'm picking Marquette to finish third for one single reason:  this is the year for Marquette and the Three Amigos to achieve great things.  There is no grace period for new coach Buzz Williams (http://wiki.muscoop.com/doku.php/men_s_basketball/buzz_williams).  Expectations are, and should be, very high.

Truth is, this team underachieved in the regular season last year,  yet it returns every meaningful player and introduces a few potentially useful spare parts.  Despite coaching turnover, this mix should equate to more wins despite the overall improvement of the Big East across the board. MU has the veteran backcourt that coaches crave, admirable depth, and the seasoning of three straight NCAA appearances.

If this team significantly underachieves with all of their experience and talent, I could care less how good of a recruiter Buzz Williams is.  No matter how good the incoming talent will be starting in 2009, Marquette will take a significant step back after losing McNeal, James, and Matthews.  Anyone else not that interested in two disappointing years?    That puts the pressure on performing well in 2008 - 2009.

The Top Two
I think these two teams will be the class of the Big East next year:

#2 - Connecticut
Projection - 13-5.  At the end of the day, I just think that they're not the best team in the league.  But they're close.

#1 - Louisville
Projection - 14-4.  Pitino hauls in the BE title he should have won last year.

Again, the focus is more on the potential bonus or red flag areas for each team, instead of the projected wins/losses or projected final ranking.  Any of the teams from #3 through #9 could realistically fit in any random order. Besides, if I could accurately model results, I'd be typing this from my own private desert island.

 Thanks to NYWarrior for his additional contributions.  Finally, want to review the data in depth?  Here is a link to the spreadsheet (http://www.muscoop.com/uploads/BE%20Season%20Analysis.xls).  (The file opens up automatically)

http://www.crackedsidewalks.com/2008/08/big-east-preview-part-two.html
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: rocky_warrior on August 06, 2008, 01:05:48 PM
Quote from: CrackedSidewalksSays on August 06, 2008, 11:15:03 AM
#7 - Notre Dame
  • 2007 Results - Significant Overperformer. Based on their stats in conference, Notre Dame (14-4), should have finished with a record of 11.7 - 6.3.  Red Flag
  • Expected Regression - Significant impact.  Based on their win % last year, Notre Dame should feel some pull back towards 0.500.  Red Flag
Projection - 11-7.  Despite the big advantage of returning almost everyone except the productive Rob Kurz, I see some major red flags for ND.

I know you're only basing this on last season - but what if you extended the "regression" theory back a few years?  As I recall, in 06-07 ND lost a number of nail biters that they could (should?) have won. 

I don't like ND, but I think any over-performance last year was negated by the under-performance the previous year.  IMO - they've got a clean slate this year.
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: bma725 on August 06, 2008, 01:17:35 PM
Quote from: CrackedSidewalksSays on August 06, 2008, 11:15:03 AM
[/li][/list]Projection - 10-8. Expect Syracuse to be better this year.  They're definitely talented, but are they too young to compete for a top spot?  Expectations for the 'Cuse may be understated.


Better?  They lost their best player from last year to the NBA, and their second best will be coming back from an ACL tear.  It's going to be tought for them to improve much at all IMO.
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: Henry Sugar on August 06, 2008, 01:28:18 PM
ND's 2006-2007 results

Expected - 10.7 - 5.3 (0.667)
Actual - 11-5 (0.688)

They were slightly above expectations in 2007, not significantly below.

ND's results really stand out when you look at win%

Expected 11.7 - 6.3 (0.648)
Actual 14-4 (0.778)

That's more than two standard deviations of difference.  They should have won between 11 and 12 games last year.
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: Henry Sugar on August 06, 2008, 01:30:25 PM
Quote from: bma725 on August 06, 2008, 01:17:35 PM
    Better?  They lost their best player from last year to the NBA, and their second best will be coming back from an ACL tear.  It's going to be tought for them to improve much at all IMO.
That's funny... NYWarrior was giving me flack for having them so low.  A lot of pundits have the Cuse much higher than I do. 

And note that I only put them at an improvement from 9-9 to 10-8.  [/list]
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: rocky_warrior on August 06, 2008, 02:08:48 PM
Quote from: Henry Sugar on August 06, 2008, 01:28:18 PM
ND's 2006-2007 results

Hmmm...looks my mind blanked out an entire year - I guess I was thinking our first year in the BE - 05-06

http://kenpom.com/sked.php?&y=2006&team=Notre%20Dame
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: Henry Sugar on August 06, 2008, 03:04:34 PM
Quote from: rocky_warrior on August 06, 2008, 02:08:48 PM
Hmmm...looks my mind blanked out an entire year - I guess I was thinking our first year in the BE - 05-06

http://kenpom.com/sked.php?&y=2006&team=Notre%20Dame

Data to do that analysis isn't readily available.  I pull it from Pomeroy's Game Plan, which he only started reporting in 07.

Still, it's not like you get credit for games you /should/ have won.  Consider that ND had almost the easiest schedule in the league last year.  Maybe that helped them win 2.5 games more than they should have?

http://kenpom.com/conf.php?y=2008&c=BE (http://kenpom.com/conf.php?y=2008&c=BE)
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: Pardner on August 06, 2008, 09:00:57 PM
As always, a great post Henry.  I appreciate all the work to compile this POV. 

One thing I noticed about TC's teams is that they outfouled the competition.  In fact, MU fouled more than any team in the Top 25 and had the most fouls/game of any team in the BE.  Look at it this way, UNC fouled six times less per game than MU...UCONN fouled 8 less times per game.   Bo is the master at this.  ND with Hagrid fouled 6 times less per game.  That differential killed MU in the close games due to extra FT's and is why perhaps we were so inconsistent.  Commonality--a big man in the middle.

While I love TC's pressure defense as a fan, I like that Buzz is going taller and longer.  Hoping for more zone traps, pressing with our depth and better match-up advantages vs. our small, quick line-ups of years past. 
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: Henry Sugar on August 06, 2008, 09:34:09 PM
Pardner,

Glad you brought this up.  I've actually been working on a little something that tries to get into the four factors for Marquette some more.  Basically, I don't think that the benefits of our aggressive defense were enough to overcome the disadvantage we received from committing those extra fouls.

It's worth saving for CS, though...  ;)

And by the way, anyone that is hoping we unveil a full court press this year better take another look at the inconsistency numbers.  All a press will do is increase the volatility further and make us more inconsistent.

Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: denverMU on August 06, 2008, 10:02:54 PM
First, thanks for all the great analysis.  Second, I'm not sure how you come up with the statement about a full court press making us inconsistent.  Your other comments are backed up by data, that seems to be opinion.  I recall Louisville pressing us and being very consistent at kicking our asses!
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: Pardner on August 06, 2008, 10:15:55 PM
Quote from: denverMU on August 06, 2008, 10:02:54 PM
Second, I'm not sure how you come up with the statement about a full court press making us inconsistent.  Your other comments are backed up by data, that seems to be opinion.  I recall Louisville pressing us and being very consistent at kicking our asses!

I will eagerly await Henry's analysis to come, but my guess is he is looking at a team like Tenn who goes end to end yet fouls almost as much as MU vs. a Bo coached team that fouls 7 less times per game.  I too like the Louisville trap as it controls the tempo, slows down a quick team...as you basically have to throw over it often times to break it--or have the quickness to beat to along the sidelines on the dribble.  UL fouls 3 less time per game than MU, btw.
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: Henry Sugar on August 06, 2008, 10:22:53 PM
well, since you asked...

From "Basketball on Paper.  "All of these strategies fundamentally increase the variability of the difference between points scored and points allowed"

A team's predicted win% is generally derived on the difference between the PPG they score (ORtg) and the PPG they allow (DRtg).  This difference is called the efficiency margin.  

The actual formula is Win% = NORM[ (ORtg - DRtg) / (var(ORtg) + var(DRtg) - 2cov(ORtg, DRtg))^0.5]

A press increases the variability of points allowed.  A good team scores easily against a press.  A bad team suffers greatly against the press.  The variability (range) of DRtg is increased.  If you increase var(DRtg), that increases the denominator.

So what happens if you increase the denominator?  Your win% goes down.   

Examples of other risky strategies include shooting a lot of threes (increases the variability of your ORtg) and slowing the tempo down (which increases the variability of both ORtg and DRtg)
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: Henry Sugar on August 06, 2008, 10:46:35 PM
More on this, because I actually think the topic is fairly enlightening.  Of course, I also do basketball stats for fun.

Consistency or Inconsistency is both good and bad.  If you're a good team, it's great to be consistent.  If you're a bad team, it's bad to be consistent.

If you're a good team, like Marquette, being inconsistent means you'll probably lose a game or two you shouldn't have.  If you're a bad team, being inconsistent means you might win a game or two you shouldn't have.

This is why you see the "risky" strategies (like a press or lots of threes) employed at the end of games when teams are behind.  It increases the variability of points scored or points allowed.  Lots of time, the end result is the team just digs a bigger hole (if you're already losing, who cares?).  But sometimes the team makes a miraculous comeback.

Of particular note, I actually think that the game plans of Wisconsin and Georgetown are inherently flawed for March based on their notoriously slow tempos.  All that a slower tempo does is increase the probability that a team that's worse you might win.  Sure, these teams might be 85% favorites, but what if the slower tempo means that they're only a 70% favorite in a single-game elimination?  I'd say a team like, oh, Davidson might take those increased percentages gladly.
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: denverMU on August 07, 2008, 12:11:43 AM
I am asking this as a question, not arguing with you, can't a full court press, when run well consistently, be a positive?  I hate going back to Louisville, but they ran it against us twice and did rather well.  Arkansas and Texas are two other teams that come to mind who have run them to there advantage in the past also.  Finally, with all our guards, speed, athleticism and lack of an inside game don't you think we could excel at this?
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: muarmy81 on August 07, 2008, 05:44:17 AM
Quote from: Henry Sugar on August 06, 2008, 10:46:35 PM
Of particular note, I actually think that the game plans of Wisconsin and Georgetown are inherently flawed for March based on their notoriously slow tempos.  All that a slower tempo does is increase the probability that a team that's worse you might win.  Sure, these teams might be 85% favorites, but what if the slower tempo means that they're only a 70% favorite in a single-game elimination?  I'd say a team like, oh, Davidson might take those increased percentages gladly.

This is an interesting take on their style of basketball.  I think that it has actually benefitted them in the past for a smiliar reason you state above.  Sure, they may lower their chances of beating a poorer team but it could and has been a method for them to increase their chances of beating a more talented team.  For example, When WI went to the Final 4 about 10 years ago I would argue that they were probably game in and game out the least athletic and maybe overall talented team on the court yet their slow-paced, defensive style of basketball helped them beat a few teams that were much more talented and operated at a much quicker pace than Bucky.

The point I'm trying to make is...UW and Georgetown SUCK... ;D
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: Henry Sugar on August 07, 2008, 07:43:17 AM
Quote from: denverMU on August 07, 2008, 12:11:43 AM
I am asking this as a question, not arguing with you, can't a full court press, when run well consistently, be a positive?  I hate going back to Louisville, but they ran it against us twice and did rather well.  Arkansas and Texas are two other teams that come to mind who have run them to there advantage in the past also.  Finally, with all our guards, speed, athleticism and lack of an inside game don't you think we could excel at this?

Sure.  A full court press could be a positive.  You'd want to make sure that the press resulted in a much faster pace to combat the volatility, and you'd want to keep a short leash on the threes (or be good at shooting them).

Someone with more basketball insight than me would probably know, but I believe there are different flavors of a press.  Like Pardner mentioned, I think there is a big difference between the Louisville press and the Tennessee press.  I liked it a lot when we'd use the full court pressure on the PG from Acker or Cubillan.  I thought that did a nice job of hassling the opponent PG and speeding up the pace.

I guess my point is that Marquette was already really inconsistent last year, and adding a press only makes that problem worse.  I'm very worried about any strategy that increases inconsistency. 
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: denverMU on August 07, 2008, 08:42:48 AM
OK, thanks for the thoughtful insight.  I just feel there were times last year, particularly when we lost, we got stuck trying to run half court offensive plays that we were not very successful with.  It seems when we can run, with all our quickness and athleticism. we are far better off.
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: jmayer1 on August 07, 2008, 01:02:49 PM
Henry, I understand the regression to the mean factor you included, but does this have much affect if you took historical records over the last 5 years into it, ie-St John's, Rutgers, and USF have been bad over this time and there is no reason to think they won't continue to be bad then next 5?  Same thing on the other side with GTown, Louisville, and Pitt.  On a bigger scale, Duke has been so good over the past 25 years, there is no reason to expect regression towards .500 anytime soon. 

You would expect not too many teams will have either 4 wins or 4 losses in back to back years but if there is no difference in terms of talent is there any reason to expect a large regression back to 9 wins and 9 losses based purely on statistical averages over the long-term?
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: Henry Sugar on August 07, 2008, 01:46:27 PM
deleted.  double post
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: Henry Sugar on August 07, 2008, 01:48:04 PM
First, this notion isn't infallible, but let's look at two teams.

Duke's conference records

2001-02    13-3  (.812)
2002-03    11-5  (.688) 
2003-04    13-3  (.812)
2004-05    11-5  (.688) 
2005-06    14-2  (.875) 
2006-07    8-8    (.500)
2007-08    13-3    (.812)

It's tough for good teams to stay at the top.  Better coaches, like Coach K, are able to resist the trend towards 0.500 for a longer period of time, but they almost always get pulled back towards 0.500.

Rutgers

2002-03    4-12  (.250) 
2003-04    7-8    (.467)
2004-05    2-14  (.125) 
2005-06    7-8    (.467) 
2006-07    3-13  (.231) 
2007-08    3-13    (.231)

Rutgers is bad, but even they get pulled back up towards 0.500 almost every other year.  Last year was a new coach, however.

Remember too that for Rutgers, we're talking about an improvement of 13%.  That would improve their win% to 0.261, or a W-L record of 4.6 - 13.4.  So it's not like they're jumping from 3-13 all the way to 8-8.

I really prefer looking at this in combination with a team that overperformed or underperformed, like GU or ND.  Both teams overperformed AND are dealing with the pull towards 0.500.  danger danger
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: jmayer1 on August 07, 2008, 01:49:41 PM
Ok, thanks for the explanation Henry.
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: wadesworld on August 07, 2008, 02:24:51 PM
Quote from: denverMU on August 06, 2008, 10:02:54 PM
First, thanks for all the great analysis.  Second, I'm not sure how you come up with the statement about a full court press making us inconsistent.  Your other comments are backed up by data, that seems to be opinion.  I recall Louisville pressing us and being very consistent at kicking our asses!
That's just proving the point further.  Louisville KILLED us twice during the year, but lost to teams like BYU, Dayton, Cinci, and Seton Hall (by 10, the game after beating us by 20)...I would consider us better teams than any of those teams, yet we got killed by them and those teams beat them.  It's inconsistency.  Every team has their inconsistencies, including Louisville, but I would agree with Henry, although he did all of his statistics and I'm just basing it off of what basketball knowledge I have.  Some teams know how to attack their press and have the players who can do it well, others don't.  The most important thing about it is adjustments, but some days it works as well as possible and others it'll lose you the game.

Quote from: Henry Sugar on August 06, 2008, 10:46:35 PM
Of particular note, I actually think that the game plans of Wisconsin and Georgetown are inherently flawed for March based on their notoriously slow tempos.  All that a slower tempo does is increase the probability that a team that's worse you might win.  Sure, these teams might be 85% favorites, but what if the slower tempo means that they're only a 70% favorite in a single-game elimination?  I'd say a team like, oh, Davidson might take those increased percentages gladly.
I completely agree with this, except the part about  Davidson.  Their non-conference schedule prepared them for what to expect come tournament time.  It's the opposite of what we do...we have the extremely hard conference schedule to prepare us so we can have cupcake non-conference games...they need togh non-conference games to prepare them for the tournament because they have an easier conference.  Early on in the year they lost by 4 to UNC and were up by 20 AT UCLA before losing (both teams went to the Final Four).  They lost to Duke by 6 and to NC State by 1.  They ended up losing by 2 with a shot to win the game to the eventual National Champions in the Elite Eight and Kansas does not slow down the tempo like Wisconsin and Georgetown.  There was a method to their madness in their non-conference schedule...they lost all of those big games, but they found out that they could at least hang with those teams, so when it came tournament time they had confidence.  Quite honestly I don't think there was anything flukey or any sort of upset about it.  Those two games were upsets in that they were seeded lower and don't have the program history of UW and GT, but I believe they were the better team in both those games.
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: jce on August 08, 2008, 10:07:08 AM
Quote from: denverMU on August 07, 2008, 12:11:43 AM
I am asking this as a question, not arguing with you, can't a full court press, when run well consistently, be a positive?  I hate going back to Louisville, but they ran it against us twice and did rather well.  Arkansas and Texas are two other teams that come to mind who have run them to there advantage in the past also.  Finally, with all our guards, speed, athleticism and lack of an inside game don't you think we could excel at this?


I think the press is inherently a weaker defense that good teams know how to handle.  You used to see teams (like Pitinos early teams) use it constantly, but now they only use it "strategically," to quicken the tempo of a game or cause an unexpected turnover.  The "40 Minutes of Hell" defense that Richardson used at Arkansas isn't really run any more with the exception of Missouri.  And a good team like Kansas shredded it.
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: Blackhat on August 08, 2008, 11:41:41 AM
I agree the full c press should only be used strategically and mostly in smaller spurts.   With the amount of games we now play a liberal use of the fc press would work against us. 
Title: Re: [Cracked Sidewalks] Big East Preview - Part Two
Post by: RawdogDX on August 08, 2008, 12:14:10 PM
Quote from: Stone Cold on August 08, 2008, 11:41:41 AM
I agree the full c press should only be used strategically and mostly in smaller spurts.   With the amount of games we now play a liberal use of the fc press would work against us. 

The amont of games we play?  I don't see why people would want to play full court against us, we have a all beast pg and capable back ups.  Love to play up tempo.  It seems like it would be playing into our hand.