collapse

* Resources


UDM 4

* Butler SOTG

Theo John

15 points, 11 rebounds,
1 assist, 3 blocks,
32 minutes

2018-19 Season SoG Tally
Howard10
S Hauser4
John4
Anim2
Chartouny2

'17-18 * '16-17 *
'15-16'14-15 * '13-14
'12-13 * '11-12 * '10-11

* Big East Standings

* Recent Posts

Trap Game Tomorrow by Jay Bee
[Today at 02:32:02 AM]


NCAA Tournament Hotels by Jay Bee
[Today at 02:25:44 AM]


MAKE YOUR PROVIDENCE PREDICTIONS HERE by The Thing
[Today at 01:16:01 AM]


What? Is Providence.... by rocky_warrior
[Today at 12:43:39 AM]


MAKE YOUR BUTLER PREDICTIONS HERE by denverMU
[Today at 12:36:04 AM]


KenPom's Big East predictions vs. reality by Mr. Sand-Knit
[February 22, 2019, 11:55:28 PM]


Marquette NBA Thread by MU82
[February 22, 2019, 11:26:19 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: @ Providence

Marquette
79
Marquette @

Providence

Date/Time: Feb 23, 11:00 a.m. CT
TV: Fox
Schedule for 2018-19
Butler
69

Author Topic: Recent Marquette Historical Stats (long)  (Read 3585 times)

Henry Sugar

  • Registered User
  • All American
  • *****
  • Posts: 1977
  • There are no shortcuts
    • Cracked Sidewalks
Recent Marquette Historical Stats (long)
« on: October 16, 2007, 02:16:31 PM »
Analysis of Recent MU History and a Comparison of our last three coaches.

I've been playing around with some data for a while, picking it up and then putting it down on a fairly regular basis.  (My original goal was a relative comparison of Marquette and Illinois, but that is a post for another time).  Regardless, the previous retrospective post here http://www.muscoop.com/index.php?topic=4036.0 spurred me to finally run the final numbers and share with the board.  Interestingly enough, I was already (half-heartedly) looking at MU from the idea of stock returns to see if there is any correlation between win % and NCAA wins.

The comparison is for the recent history of Marquette, which I date as starting with O'Neill in 1989.  The two basic factors that I began with were "Win %" and "YoY return".  In other words, what percentage of the games did the basketball team, and by how much did the team improve over the previous year?



The graph of Win Percentage (above) is a nice starting point, although the Crean emphasis in the last year is cropped by the picture.  O'Neill's years are in orange, Deane's in Red, and Crean 's years are in green.  A cursory view at the graph bears out some high-level general impressions of the different coaches.  O'Neill started at a low point and built the team up in a remarkably short period, improving every year but one.  Deane had some excellent win percentages over the first four years, but also had the team get worse the last three years.  Finally, under Crean, the program experienced two mundane years, an early meteoric rise with Wade on the team, and then a decline followed by solid improvement the last two years.

Digging into the numbers a little more:

O'Neill inherited a team that had gone 0.464 the previous year from Dukiet.  The team improved every year under his tenure, including a high return of 45% and a high win % of 0.727.  Coach O'Neill won one conference championship, made the NCAA's twice, and the NIT once.  During this five years, MU had two NCAA wins over five years (which both happened in his final year).  O'Neill averaged a win % of 0.578, a YoY annual return of 12%, and 0.4 NCAA wins.  O'Neill deserves all of the coaching respect he gets for restoring our program to prominence.

Deane inherited a team that had gone 0.727 the previous year.  Under Deane, the team only got better once over five years (Deane's second year), with a high win % mark of 0.742.  Marquette reached the NCAA twice and the NIT twice.  Deane had one NCAA win, which also occurred in his second year, although there were also six NIT victories.  Deane averaged a win % of 0.643, a YoY annual return of -7%, and 0.2 NCAA wins.  There were clearly some good times and winning programs under Deane, but it is obvious from the numbers that the program was consistently declining under Deane.

Crean inherited a team that had gone 0.483 the previous year.  The team has improved over the previous year in five of eight seasons, with two additional times that Crean had 0% returns.  The high win % for Crean is 0.818 (the final four year), with a high YoY return of 52% (Wade's first year).  Marquette has won one conference championship, reached the NCAA's four times and the NIT three times.  Under Crean, the team has only won in the NCAA's in one season, where they got four wins in the FF season.  Crean has averaged a win % of 0.652, YoY annual return of 7%, and 0.5 NCAA wins.  I think that the most telling view is the YoY annual return over eight years, which if consistent for 2007-2008, means about 26-27 wins this year.

What does this mean to the ultimate goal of winning an NCAA title?  Well, with further analysis, not a whole lot.  I ran a few regression analyses, and there is not a lot of correlation between Win %, YoY Return, or Conf win %.  In other words, how a team does in the regular season does not predict how they will do in the post-season.  There is, however, a slim statistical correlation between conference championships and NCAA victories.  Unfortunately, there's so few data points that it is very difficult to say how win %, YoY improvement, or Conf Win % will result in NCAA wins.

Besides sharing some data and a chart, so what?  The point is simply that under Crean, the program has averaged over twenty wins a season while still improving each season at an average of 7% per year.  I thought it was a nice complement to the previous post.  While I agree with Crean's flaws, he certainly is doing an excellent job of building MU basketball into a program and I'm not interested in trading him for any other coach.  Not bad for a 41 year old.  Let's just hope that we are able to win a conference championship this year.   ;)
A warrior is an empowered and compassionate protector of others.

Canned Goods n Ammo

  • Registered User
  • All American
  • *****
  • Posts: 5008
  • Ammo, clean shaven Ammo.
Re: Recent Marquette Historical Stats (long)
« Reply #1 on: October 16, 2007, 02:29:32 PM »
Analysis of Recent MU History and a Comparison of our last three coaches.

I've been playing around with some data for a while, picking it up and then putting it down on a fairly regular basis.  (My original goal was a relative comparison of Marquette and Illinois, but that is a post for another time).  Regardless, the previous retrospective post here http://www.muscoop.com/index.php?topic=4036.0 spurred me to finally run the final numbers and share with the board.  Interestingly enough, I was already (half-heartedly) looking at MU from the idea of stock returns to see if there is any correlation between win % and NCAA wins.

The comparison is for the recent history of Marquette, which I date as starting with O'Neill in 1989.  The two basic factors that I began with were "Win %" and "YoY return".  In other words, what percentage of the games did the basketball team, and by how much did the team improve over the previous year?



The graph of Win Percentage (above) is a nice starting point, although the Crean emphasis in the last year is cropped by the picture.  O'Neill's years are in orange, Deane's in Red, and Crean 's years are in green.  A cursory view at the graph bears out some high-level general impressions of the different coaches.  O'Neill started at a low point and built the team up in a remarkably short period, improving every year but one.  Deane had some excellent win percentages over the first four years, but also had the team get worse the last three years.  Finally, under Crean, the program experienced two mundane years, an early meteoric rise with Wade on the team, and then a decline followed by solid improvement the last two years.

Digging into the numbers a little more:

O'Neill inherited a team that had gone 0.464 the previous year from Dukiet.  The team improved every year under his tenure, including a high return of 45% and a high win % of 0.727.  Coach O'Neill won one conference championship, made the NCAA's twice, and the NIT once.  During this five years, MU had two NCAA wins over five years (which both happened in his final year).  O'Neill averaged a win % of 0.578, a YoY annual return of 12%, and 0.4 NCAA wins.  O'Neill deserves all of the coaching respect he gets for restoring our program to prominence.

Deane inherited a team that had gone 0.727 the previous year.  Under Deane, the team only got better once over five years (Deane's second year), with a high win % mark of 0.742.  Marquette reached the NCAA twice and the NIT twice.  Deane had one NCAA win, which also occurred in his second year, although there were also six NIT victories.  Deane averaged a win % of 0.643, a YoY annual return of -7%, and 0.2 NCAA wins.  There were clearly some good times and winning programs under Deane, but it is obvious from the numbers that the program was consistently declining under Deane.

Crean inherited a team that had gone 0.483 the previous year.  The team has improved over the previous year in five of eight seasons, with two additional times that Crean had 0% returns.  The high win % for Crean is 0.818 (the final four year), with a high YoY return of 52% (Wade's first year).  Marquette has won one conference championship, reached the NCAA's four times and the NIT three times.  Under Crean, the team has only won in the NCAA's in one season, where they got four wins in the FF season.  Crean has averaged a win % of 0.652, YoY annual return of 7%, and 0.5 NCAA wins.  I think that the most telling view is the YoY annual return over eight years, which if consistent for 2007-2008, means about 26-27 wins this year.

What does this mean to the ultimate goal of winning an NCAA title?  Well, with further analysis, not a whole lot.  I ran a few regression analyses, and there is not a lot of correlation between Win %, YoY Return, or Conf win %.  In other words, how a team does in the regular season does not predict how they will do in the post-season.  There is, however, a slim statistical correlation between conference championships and NCAA victories.  Unfortunately, there's so few data points that it is very difficult to say how win %, YoY improvement, or Conf Win % will result in NCAA wins.

Besides sharing some data and a chart, so what?  The point is simply that under Crean, the program has averaged over twenty wins a season while still improving each season at an average of 7% per year.  I thought it was a nice complement to the previous post.  While I agree with Crean's flaws, he certainly is doing an excellent job of building MU basketball into a program and I'm not interested in trading him for any other coach.  Not bad for a 41 year old.  Let's just hope that we are able to win a conference championship this year.   ;)

AWESOME POST.

Numbers don't lie, but they can we slanted. However, I agree with your perspective and analysis, especially if you add in all of the nice marketing that Crean has helped facilitate. I know "marketing" is a 4 letter word to some people, but it has helped bring this program back to national relevance.

Great work!



Djgoldnboy

  • Registered User
  • Starter
  • ***
  • Posts: 190
Re: Recent Marquette Historical Stats (long)
« Reply #2 on: October 16, 2007, 03:04:46 PM »
Nice work...the one thing the chart fails to realize though is quality of opponents, 10 wins in the Big East, holds a lot more weight than 10 wins in what was the C-USA conference, at least to me it does.

27 wins sounds good to me though!!


rocky_warrior

  • Global Moderator
  • All American
  • *****
  • Posts: 6091
Re: Recent Marquette Historical Stats (long)
« Reply #3 on: October 16, 2007, 03:24:11 PM »
Very nice analysis!

10 wins in the Big East, holds a lot more weight than 10 wins in what was the C-USA conference

Good point - maybe an easy way to do that would to just multiple each years win % by the SOS for that year to create a weighted win %???

tower912

  • Registered User
  • All American
  • *****
  • Posts: 13833
Re: Recent Marquette Historical Stats (long)
« Reply #4 on: October 16, 2007, 03:29:59 PM »
I really enjoyed this post.   Fascinating statistical analysis.   Great work and thank you.  In a somewhat related topic, are you able to create a graph showing the team's record after February 1, cross referenced with the SOS of the opponents?   Yes, I know that this is about Murf, but it would be interesting to see how it comes out statistically with SOS factored in.    Maybe he is right.
Real Warriors don't whine.

dwaderoy2004

  • Registered User
  • All American
  • *****
  • Posts: 1505
Re: Recent Marquette Historical Stats (long)
« Reply #5 on: October 16, 2007, 03:31:35 PM »
the problem with saying that based on an "expected" Yoy we will have 27 wins is that it has a maximum value.  We cannot continually see returns, otherwise we will evetually be undefeated, and then there is no possible improvement.  in order to improve, we must have a down year.  I think this works in the short term, if a coach is only there for 5-10 years as in these cases.  but for someone who has been at a school for 30 years, i.e., boeheim or lute olson, not sure it would work so well...

Henry Sugar

  • Registered User
  • All American
  • *****
  • Posts: 1977
  • There are no shortcuts
    • Cracked Sidewalks
Re: Recent Marquette Historical Stats (long)
« Reply #6 on: October 16, 2007, 07:55:51 PM »
Just wanted to say thanks to the people that provided some critical feedback.  I recognized going into the analysis that it was basically missing on two areas.

#1 - SOS.  I looked around (albeit not very hard) for historical SOS and couldn't find anything.  If anyone has a good reference to find this information, I'd be really grateful.

#2 - The concept of "returns" definitely has the upper limit of diminishing returns.  To view it fully, consider someone like Coach K, who already has his team at a very high win % every year and then averages (I'm guessing) 0% YoY.  I thought about running benchmarks against some high-profile coaches, but then that ran into the "too much work for a silly spreadsheet" area.  Essentially, the idea of "returns" for a basketball team is merely intriguing, but useless unless paired with the Win % or other relevant data.  As an example, when I regressed NCAA wins against Win % and YoY Return, only the Win % was statistically significant. 

However, as dwaderoy2004 posts, it does seem to make sense for a short term period to help validate the direction of a program.
A warrior is an empowered and compassionate protector of others.

mu_hilltopper

  • Warrior
  • Global Moderator
  • All American
  • *****
  • Posts: 6124
  • Eat Arby's
Re: Recent Marquette Historical Stats (long)
« Reply #7 on: October 16, 2007, 09:55:18 PM »
I think everyone has a different yardstick for success.   Win counts can be manipulated like crazy, with how OOC games are chosen, not to mention the quality of the opponent you face in conference. 

About the only win count/percentage that would make any modicum of sense is to weed out games versus a mendoza line of RPI, like 150 or perhaps 200.    In the above analysis, a win versus Savannah State is counted as highly as a win vs. Kentucky and Duke.

But more to the point .. win counts aren't everything to everyone.  If they were, we could move into the SWAC and go undefeated every year. 

Everyone has a different formula for judgment.  Many elements are involved.  Wins, sure.  Quality wins.  Wins vs. UW!  Pre-season tourneys, as well as Conference tourneys.  NCAA wins. -- And of course, the intangibles.  Personality.  Alumni and fan relations.  Marketing sense.  Etc.

Each of the coaches listed above were not total failures, nor do they get perfect marks in all categories.

77ncaachamps

  • Registered User
  • All American
  • *****
  • Posts: 8451
  • Last of the Warrior Class
Re: Recent Marquette Historical Stats (long)
« Reply #8 on: October 16, 2007, 10:29:15 PM »
You've got too much time on your hands.

But in this case, that's a good thing. ;)
[This space reserved for a Jordan King highlight.]