Last week I looked at which coaches have historically played freshmen the most. Unfortunately, this doesn’t completely inform us whether a coach favors freshmen. Circumstances can have a lot to do with whether or not a coach gives time to his newest players. For example, whether a coach gets a lot of Top 10 recruits or loses players to the NBA draft will impact whether or not freshmen play. But it doesn’t really tell us whether a coach favors freshmen.

And from a recruit’s perspective, once the roster is set, that favoritism becomes a key question. Will the coach give me a fair chance to prove myself early in my career? Or am I going to have to wait and earn my playing time through seniority?

Let’s take Matt Painter as an example. In last week’s analysis Painter had historically given the 4th most playing time to freshmen of any power conference coach with at least five years of tenure. But does Painter really favor freshmen? It turns out that circumstance may have had a lot to do with Painter’s decision making. Let’s start by noting that on average freshmen play more for first and second year head coaches than long-tenured coaches. New coaches tend to favor the players they bring in. And thus in 2004 at Southern Illinois and 2006 & 2007 at Purdue, Painter did the typical thing and gave a lot of minutes to young players. This factor was exacerbated at Purdue because Gene Keady left Painter with a highly depleted roster when he departed.

By 2008, Painter should have been playing more upperclassman, but then something remarkable happened. Painter got one of the best freshmen classes in school history as E’Twaun Moore, JaJuan Johnson, Scott Martin, and Robbie Hummel all joined the team at the same time. Then in 2013, the school faced a total rebuild as the majority of the productive upperclassman were gone and Painter went very young again. Those five seasons (2004, 2006-2008 and 2013) mean Matt Painter has played a lot of freshmen in his career. But that doesn’t necessarily mean he favors them. He has simply put the most logical lineup on the floor at various points in time. When I control for the quality of the players and the stage of the rebuilding process, it turns out that Painter still favors freshmen (17^th nationally), but not nearly as much as the raw playing time numbers (4th nationally) would suggest.

In today’s analysis, I essentially attempt to control for circumstances. I predict each player’s playing time based on his stats (his measured production) and the stats of his teammates. If a coach is playing an optimal lineup, the player with the best measured stats (high ORtg given usage rate, high steal rate, ect.) should play the most. But what you see is that most coaches will typically play a freshman less than an upperclassman producing at the same level.

The main new variable is shown in the next table under the heading “Fr Rel to Soph”. This shows how many minutes a freshman will receive relative to a sophomore with equivalent statistics. In the vast majority of cases, the freshmen will play less. For example, under Mike Krzyzewski if a freshman and sophomore have equivalent stats the freshman will play 9.0% fewer minutes. Under Bill Self, if a freshman and sophomore have equivalent stats the freshman will play 9.7% less. Amazingly, under John Calipari, if a freshman and sophomore have equivalent stats, the freshman will play 3% more minutes. This is exceedingly rare.

My methodology is not perfect. We are only using measured production. It is quite likely that upperclassman do a lot of unmeasured things better than freshmen. They probably set better screens. They have fewer defensive lapses. And that is why it isn’t surprising that given equivalent measured stats, the typical coach will play the freshman less.

But relative to the earlier measure (the percentage of team minutes given to freshmen), I find this measure to be remarkably informative. While some coaches favor freshmen by either metric (like John Calipari) and some coaches disfavor freshmen by either metric (like Bo Ryan), in other cases the total minutes given to freshmen can be deceiving.

For example, Lon Kruger looked like a coach who refuses to play freshmen. But that isn’t the case at all. Kruger simply recruited enough high quality D1 transfers that his freshmen were always the weakest link. And he also inherited a veteran Oklahoma team when he took that job. But that doesn’t mean Kruger refuses to play freshmen. He simply has had a top-heavy roster every year.

Anthony Marshall is probably the perfect example. As a freshman at UNLV in 2008, Marshall was by far one of the least efficient players on the team, posting an ORtg of 85 while the rest of the rotation all had ORtgs over 100. But Marshall was a Top 100 recruit and high potential guy, and Kruger stuck with him despite his offensive woes. Marshall played the 5th most minutes that season, despite clearly being the rotation’s least polished player. Circumstance, not in-season favoritism, has caused Kruger to have given so few minutes to freshmen.

On the flip side, Mike Krzyzewski and Bill Self have given a relatively high number of minutes to freshmen. But for those coaches, it has basically been by circumstance, more than choice. They get Top 10 recruits who are so good, you can’t leave them on the bench. But given equivalent stats, both coaches are actually much more likely to stick with the experienced player.

Now this doesn’t mean that Bill Self won’t play Andrew Wiggins major minutes. He will! Andrew Wiggins will be the best player on Kansas next year. But for a player like Joel Embiid, that is where I think this makes more of a difference. If John Calipari was coaching this year’s Kansas team, Embiid could probably be counted on to play 60 percent of the team’s minutes. Under Bill Self, 47 percent seems more realistic.

This also fits a lot more closely with my understanding of these coach’s philosophies. While Mike Krzyzewski isn’t afraid to play freshmen, he has also said that players need to get better in practice, not in the game. Krzyzewski just isn’t going to put a freshman on the court to make that player happy. Players have to earn playing time.

At many points in time, I have called this the “anti-recruiting tool”. If you are hoping to go to the NBA early, you want to ensure you get early playing time. Thus one could use this to argue that given equivalent teammates, a player like Jabari Parker should go to Kentucky and not Duke. But if you think you might need a year or two to develop, it isn’t clear that a coach favoring freshmen is a good thing. If you go to Duke, have elite ability, develop that ability, and stick around for your junior year, you can pretty much be assured that Mike Krzyzewski is going to showcase your game. For recruits that aren’t necessarily hoping to be one-and-done, freshmen playing time might not matter much at all.

Minor Details

- I exclude players that used a low number of possessions on the season. Thus I am basically talking about mid-level or strong freshmen at each program. If a freshman is just hoping to keep a scholarship, none of the above analysis really applies.

- I drop players with injuries. I.e. Nerlens Noel and Kyrie Irving are not included in this calculation.

- Finally, it matters a lot whether a coach is engaged in a rebuilding year. For example, Tim Miles has basically gone from one rebuilding job at Colorado St. to the next at Nebraska. I attempt to control for rebuilding jobs and that moves Tim Miles down some. But even after accounting for this, Miles has still been unusually committed to freshmen.

For example, last year Nebraska’s Benny Parker had dreadful measured stats; but Miles stuck with him. In 2008 at Colorado St. Adam Nigon never shot and wasn’t efficient; Miles stuck with him. Nigon eventually became a productive senior for Colorado St. And Miles really stuck with Pierce Hornung at Colorado St. As a freshman Hornung was undersized and afraid to shoot - he wasn’t even a great rebounder yet - but Miles stuck with him. Hornung eventually became one of the all-time greatest rebounders in college basketball.

Providence’s Ed Cooley has had a similar profile to Miles. He has faced rebuilding jobs at Fairfield and Providence, but he has also been even more committed to freshmen than the typical coach.

Favoring Freshmen, Major Conference Coaches with at least five Years as HC

Fr Rel to Soph - Team - Coach

6.1% - Nebraska - Tim Miles

4.5% - Providence - Ed Cooley

3.0% - Kentucky - John Calipari

1.6% - Texas - Rick Barnes

1.3% - Santa Clara - Kerry Keating

0.8% - Oklahoma St. - Travis Ford

0.4% - Drake - Ray Giacoletti

-0.9% - George Mason - Paul Hewitt

-0.9% - Alabama - Anthony Grant

-1.0% - Colorado St. - Larry Eustachy

-1.1% - South Florida - Stan Heath

-1.2% - Loyola Marymount - Max Good

-1.3% - Fordham - Tom Pecora

-1.5% - Washington - Lorenzo Romar

-1.9% - Duquesne - Jim Ferry

-1.9% - San Diego - Bill Grier

-2.2% - Purdue - Matt Painter

-2.2% - Georgia - Mark Fox

-2.4% - Iowa - Fran McCaffery

-2.5% - Villanova - Jay Wright

-2.7% - Wake Forest - Jeff Bzdelik

-2.8% - La Salle - John Giannini

-2.9% - Vanderbilt - Kevin Stallings

-3.2% - Michigan - John Beilein

-3.4% - Boston College - Steve Donahue

-3.7% - Georgia Tech - Brian Gregory

-3.7% - Seton Hall - Kevin Willard

-3.8% - Central Florida - Donnie Jones

-4.0% - Oklahoma - Lon Kruger

-4.1% - Evansville - Marty Simmons

-4.2% - Indiana - Tom Crean

-4.3% - Wichita St. - Gregg Marshall

-4.4% - Arkansas - Mike Anderson

-4.6% - Ohio St. - Thad Matta

-4.9% - North Carolina - Roy Williams

-5.0% - Southern Illinois - Barry Hinson

-5.1% - Texas Tech - Tubby Smith

-5.1% - UCLA - Steve Alford

-5.3% - Portland - Eric Reveno

-5.3% - Louisville - Rick Pitino

-5.4% - Texas Christian - Trent Johnson

-5.4% - Louisiana St. - Johnny Jones

-5.6% - Brigham Young - Dave Rose

-5.6% - San Francisco - Rex Walters

-5.6% - San Diego St. - Steve Fisher

-5.7% - Florida - Billy Donovan

-5.7% - George Washington - Mike Lonergan

-5.8% - Auburn - Tony Barbee

-5.9% - St. Louis - Jim Crews

-6.0% - Loyola Chicago - Porter Moser

-6.0% - Colorado - Tad Boyle

-6.1% - Saint Joseph's - Phil Martelli

-6.2% - Oregon St. - Craig Robinson

-6.4% - DePaul - Oliver Purnell

-6.6% - Arizona - Sean Miller

-6.7% - Florida St. - Leonard Hamilton

-6.7% - Illinois - John Groce

-7.0% - Oregon - Dana Altman

-7.1% - Maryland - Mark Turgeon

-7.2% - Baylor - Scott Drew

-7.2% - Northern Iowa - Ben Jacobson

-7.2% - South Carolina - Frank Martin

-7.4% - Georgetown - John Thompson III

-7.4% - California - Mike Montgomery

-7.5% - Gonzaga - Mark Few

-7.6% - Richmond - Chris Mooney

-7.8% - Texas A&M - Billy Kennedy

-7.9% - Michigan St. - Tom Izzo

-8.3% - Virginia - Tony Bennett

-8.5% - Mississippi - Andy Kennedy

-8.5% - St. Mary's - Randy Bennett

-8.8% - North Carolina St. - Mark Gottfried

-8.9% - Arizona St. - Herb Sendek

-8.9% - Utah St. - Stew Morrill

-9.0% - Syracuse - Jim Boeheim

-9.0% - Tennessee - Cuonzo Martin

-9.0% - Duke - Mike Krzyzewski

-9.1% - Washington St. - Ken Bone

-9.2% - Creighton - Greg McDermott

-9.2% - Clemson - Brad Brownell

-9.6% - West Virginia - Bob Huggins

-9.7% - Kansas - Bill Self

-10.0% - Cincinnati - Mick Cronin

-10.1% - Stanford - Johnny Dawkins

-10.4% - Pittsburgh - Jamie Dixon

-11.6% - Missouri - Frank Haith

-12.2% - Kansas St. - Bruce Weber

-12.5% - Miami FL - Jim Larranaga

-12.9% - Bradley - Geno Ford

-13.0% - Wisconsin - Bo Ryan

-15.3% - Temple - Fran Dunphy

-15.7% - Massachusetts - Derek Kellogg

-16.0% - St. Bonaventure - Mark Schmidt

-16.3% - Notre Dame - Mike Brey

-16.5% - Marquette - Buzz Williams

Follow @DanHanner