The Effects of Coaching Changes

In a recent post, I went back and looked at a Beilein West Virginia team. Both Bob Huggins and John Beilein have had success coaching the Mountaineers, but they have done so with markedly different offensive strategies. The following graph visualizes these differences:

There once was a time when West Virginia wasn’t a perennial offensive rebounding powerhouse. In fact, Beilein’s teams were commonly towards the bottom in this category. It’s no coincidence that Huggins tenure overlaps with Kevin Jones’ career.  Jones was a force on the glass throughout his WVU career. Still, it’s definitely safe to say that Huggins emphasizes crashing the boards.

Beilein’s offenses had a far different DNA. They shot the ball with high efficiency and avoided turnovers. This type of team, in theory, should not need offensive rebounding to be successful.

The above analysis got me thinking more and more about the effects of a new coach on a program. When a new coach is hired, the team is more than likely still somehow impacted by the old coach. Intuitively, it seems that a first year coach’s team should be a mixture of the old coach’s foundation plus the new coach’s adjustments. I decided to stay with West Virginia and look at the Beilein to Huggins transition. The following graph highlights the tempo, offense, and defense of Beilein’s 2007 WVU team and Huggins’ 2007 Kansas State team. It also shows Huggins’ 2008 WVU team, which allows you to get a sense of the impact of Huggins on the program.

Pace

To me, pace is the most interesting variable to evaluate a coaching change. Pace is presumably less dependent on players and more dependent on coaching than offensive and defensive efficiency. As you can see, Beilein ranked around 275 in tempo his final year at West Virginia. At Kansas State, Huggins ranked around 175. I suspected that Huggins first year in Morgantown would fall in between these two ranks and that is how it happened to turn out.

Offense

I used Ken Pomeroy’s AdjO statistic for this variable. Here it starts to get a little tricky. I would like to think that any great coach would admit the importance of having good players. Thus, the best offensive coach in the country is obviously not going to necessarily have the best offensive team in the country. Talent and skill will determine offensive ranking probably just as much as coaching. Kansas State and West Virginia is not a particularly lopsided talent comparison, so for this example it would be reasonable to expect a 2008 West Virginia data point in between Huggins and Beilein from the previous year. The graph does in fact show that Huggins kept West Virginia an elite offensive team, with fair improvement over his K State team from the year before.

Defense

The three data points for AdjD are all very close to each other. Huggins’ 2007 Kansas State team was slightly better than Beilein’s WVU team. Furthermore, the2008 Mountaineers were slightly better than both of the aforementioned teams on defense.

I extended the above study to four other programs with coaching changes in recent years. The graphs for these four changes are shown below:

There’s a lot of information to take away from all of these graphs. Each team has a unique set of circumstances the new coach entered into. Here are my four takeaways from the graphs:

Steve Alford knows how to coach defense. Alford returned nearly New Mexico’s entire roster from McKay’s final season. It’s therefore not all that surprising that the team’s tempo was very similar despite the coaching change. We should also expect improvement on both sides of the ball with a more experienced roster. The Lobos went from an okay offensive team to a good offensive team, as would probably be expected. However, Alford changed New Mexico’s defense around completely. Since 2003, the worst defensive ranking an Alford led team has had is 90^th.

Ken Bone does not know how to coach defense. Harsh, I know. The graph shows that Washington State went from one of the premier defenses in the country to an extremely average defense upon the arrival of Bone. However, I should note that the Cougars did graduate Taylor Rochestie, Aron Baynes, Caleb Forrest, and Daven Harmeling from the 2010. Plus, Tony Bennett has consistently produced effective defensive teams throughout his coaching career. It’s worth noting change in pace from Bennett to Bone. This provides some evidence for my initial assumption on the dependence of coaching on pace.

Buzz Williams took over a very stable (and stacked) Marquette team. Buzz made his head coaching debut for New Orleans in 2007, but then went back to an assistant in 2008. I used his 2007 team for the graph. Not often does a coach take over a top 15 team with everyone back from the previous year. Buzz inherited a lineup of McNeal-Matthews-Hayward-James-Butler.  The team got slightly better on offense and slightly worse on defense, but Buzz had the fortune of stepping into Marquette with a loaded gun.

Gregg Marshall’s team hasn’t always been a tempo free darling. In the last two years, Wichita State has been known for their high rankings by the computers. The 2008 Shockers struggled to an 11-20 overall record. Wichita State and Winthrop were very similar teams in 2007, but the coaching change and a fair amount of roster turnover led to a rebuilding year in 2008. This one is tough to gain any real insight from as my measures don’t significantly distinguish differences between Marshall and Turgeon.