Value Add Formula: A Closer Look at the Big Ten

Dylan Burkhardt
on

x350[1]Sports Illustrated’s Luke Winn brought some mainstream attention to John Pudner’s Value Add formula last week. The formula measures the precise offensive value added by any specific player – or more specifically the drop off involved in replacing one player with a generic ninth or tenth man off of the bench. (Detailed explanation here.)

Baseball fans probably realize this is somewhat similar to the wins above replacement metric utilized in baseball. Longtime readers of this blog will also probably notice the similarities between Pudner’s system and KJ’s PORPAG system.

Pudner compiled an entire dataset of all 2500 returning players in college and Winn made several observations regarding the Big Ten. Some that Big Ten fans are all too familiar with (Jordan Taylor is really good) but also just how large of an impact fifth year transfers Sam Maniscalco (Bradley to Illinois) and Brandon Wood (Valparaiso to Michigan State) should have on their respective teams that otherwise have some major question marks.

There’s been plenty of analysis of Marquette, the Big East, New York centric and Chicago based teams but here are a couple thoughts on Michigan and next year’s Big Ten.

Michigan has two players – Hardaway and Novak – that rank in the top-100 of returning players and Morgan comes in at 160th. Smotrycz, Vogrich, Douglass and Horford fail to rank in the top-500.

Rnk First Last Feet Inches Value Add BCS
47 Tim Hardaway 6 5 3.49%
77 Zack Novak 6 5 3.06%
160 Jordan Morgan 6 8 2.39%
616 Evan Smotrycz 6 9 0.83%
633 Matt Vogrich 6 4 0.80%
673 Stu Douglass 6 3 0.72%
947 Jon Horford 6 9 0.32%

Obviously Michigan loses a lot from Darius Morris’s departure but I am a bit more surprised at how well Novak and Morgan performed in this metric. Because Novak played such a large percentage of Michigan’s minutes, I’m going to guess that his number is somewhat just a residual of how Michigan performed on the year.

Hardaway is the player on Michigan’s roster most expected to dramatically increase his production, especially when compared to last season’s numbers including his slow start. It’s tough to see Novak or Douglass making a major jump in production, but Smotrycz, Morgan and Horford all seem like worthy candidates to increase their production.

I filtered Big Ten schools out of the data set to see if anything jumps out, here’s the top 10 returning players in the conference:

Rank First Last Feet Inches Team Value Add BCS
1 Jordan Taylor 6 1 Wisconsin 9.46%
2 Jared Sullinger 6 9 Ohio St. 6.44%
14 John Shurna 6 8 Northwestern 4.46%
25 William Buford 6 5 Ohio St. 4.09%
36 Jordan Hulls 6 0 Indiana 3.60%
41 Draymond Green 6 6 Michigan St. 3.54%
43 Trevor Mbakwe 6 8 Minnesota 3.52%
47 Tim Hardaway 6 5 Michigan 3.49%
54 Josh Gasser 6 4 Wisconsin 3.33%
72 Christian Watford 6 9 Indiana 3.13%

There aren’t many surprises there and it’s tough to argue at the top, where Jordan Taylor and Jared Sullinger are arguably two of the top players in the country, not just the league. Hulls is a devastatingly efficient offensive player so his ranking makes some sense, but perhaps it’s time to be a little more bullish on next year’s Hoosiers with Hulls, Watford and Zeller. Indiana returns the fourth most value add behind Ohio State, Wisconsin and Northwestern. Josh Gasser is the one in the top-10 that really stands out as a surprise – maybe he should have been included on our breakout players list last week.

You can find all of the Big Ten individual returnee data for every team filtered out on this page. (Note: Transfers into the conference aren’t included in this list.)

When looking at data on a team-by-team basis there are a couple of teams that stand out as exceptionally positive or negative. Ohio State and Wisconsin at the top of the list is no surprise but the numbers infer that we should perhaps be a little more bullish on Northwestern and Indiana. The numbers look surprisingly grim for Iowa, a team that conventional wisdom suggests is on the verge of improvement, as just Gatens and Basabe register as valuable returners. The big question for Iowa is whether players like Bryce Cartwright, Eric May, Roy Marble and Zach McCabe can make large jumps in their development. No one expected much from Penn State or Nebraska, so it’s not surprising to see them at the bottom of the list.

  • SamGoBlue

    I love seeing measurements like these done to try to project how players and teams will perform, but in reality it is a far-from-perfect way of going about things. Looking at some numbers put together is never going to tell you what will happen in the future, even though I am a huge proponent of stats. For example, you mention that the formula projects Iowa to finish extremely low. What the formula doesn’t take into account here is that Iowa is now entering the second year of a brand-new system under new head man Fran McCaffery. Wisconsin, on the other hand, should project to struggle a little bit with the loss of Nankivil and Leuer. While this formula probably has a little bit of merit, I would pay much more attention to what someone who has followed the Big Ten religiously over the past few seasons predicts.

  • 5pinbowling

    Hulls is a role player — a very efficient role player, but a role player nonetheless.  You could see when good defensive teams focused on him, he could be taken away.  The same is true for Novak.  (You don’t have to believe me.  KenPom has Hulls as a role player and Novak as a limited role player).  They have to either get help via other players or through effective offensive sets. 

    I haven’t taken a look at the methodology of this system yet, and I’m not saying Hulls and Novak can’t be valuable parts of excellent college teams.  But for Indiana to exceed expectations, they’re going to need players besides Hulls to step up, a la Morris and Hardaway Jr. stepping up last year for us, while Novak and Douglass provided valuable but supporting offensive roles.

    • hoosierdaddynow

      Excellent po

  • http://www.umhoops.com/ Dylan Burkhardt

    These numbers are by no way supposed to be an all encompassing measurement of a player,  especially because they only cover offense. However, they do provide an interesting discussion point with the season around two months away.

  • Mattski

    Just watched Robin Benzing take an Italian player to the hole with a lovely move in the FIBA European championships. Announcer said, “People see him as the next Dirk Nowitski.”

  • Flwolve
    • JBlair52

      Juwan did pretty good in his first year for the Broncos…I’d love to get him but I doubt it would happen.  He’d sit out a year and then be in the same class as Brundidge, Burke, and Bielfeldt

      • http://twitter.com/Alex_MnB Alex Cook

        I don’t know, having him take a non-scholarship spot like Jordan Dumars did couldn’t hurt.

  • Nick

    There are value in these numbers and it does shine light on some players that don’t get enough credit.  However, extrapolation of the data to make predictions about player improvement can be a sketchy exercise.  A scouting component has to be combined with the data to make worthwhile analysis

    For example, Gasser is ranked 54th on this list and is there primarily because he is a good shooter.  He doesn’t do much else and thus it would be unwise to project an expanded role for him.  He is a low-usage player whose value comes in floor spacing, but he will not have the same ceiling and impact as a player with shot creation skills and athletecism such as Hardaway.  Same goes for Novak on the offensive end.