Here’s a brief overview of what we’ve learned from our survey of MSU’s statistical trends on offense over the last 12 years, broken down by the “four factors” of tempo-free statistical analysis:
- Turnovers: Minimizing turnovers has never been a strength for MSU under Izzo. Three seasons were particularly bad (1997-97, 2001-02, and 2006-07), with the percentage generally declining in the periods between those peaks. Up until 2006-07, having two starting-quality point guards (Taylor/Hill, Hill/Neitzel) had led to the best results.
- Field Goal Shooting: Izzo-led teams have always taken a below-average percentage of their shots from 3-point range. Field goal shooting has been a strength more often than not. Spikes in 2-point or 3-point shooting have generally offset each other. MSU has been a better overall shooting team the last five years than they were in the previous seven-year period, although effective FG percentage has slowly declined over those five years.
- Free Throw Shooting: Over the last 12 years, the team’s free throw shooting percentage has generally increased over time–becoming a major team strength from 2001-02 to 2005-06. Because of a decline in the rate the team gets to the line relative to field goal attempts, their net free throw rate (FTM/FGA) has declined somewhat over the last five years.
- Offensive Rebounding: MSU was an utterly dominant offensive rebounding team from 1997-98 to 2000-01. They’ve been good, but not great, since. The difference seems to be a decline in the number of perimeter players crashing the boards (Bell, Peterson, Thomas, Klein, Richardson on the early Final Four teams).
The net result of these trends is the graph of offensive efficiency (points scored per 100 possessions) below. (Those of you reading this through a blog reader will need to click through to see the graph.)
Two distinct peaks/valleys are evident:
- Offensive efficiency improved from 1996-97 to 2000-01. This covers the four Big Ten championship teams (and three Final Four teams). The core group of players is Cleaves/Bell/Peterson/Granger/Hutson/Smith.
- After two years of decline, the offense again improved over the 2003-04 and 2004-05 seasons. The core group of players is Hill/Ager/Brown/Torbert/Anderson/Davis.
- Another two years of decline followed, offensive efficiency improved this past season. Hopefully, this marks the beginning of another upswing, with the Lucas/Summers/Allen/Morgan group leading the way.
The worst seasons (all three ranking below #100 nationally) came when there were transitions between the different core groups of players referenced above:
- In 1996-97, the new Izzo guys were playing with the Heathcote holdovers (Kelley, Weathers, Garvaglia).
- In 2002-03, Ballinger and Anagonye were the only remaining contributors from the 2000-01 Final Four team.
- In 2006-07, Neitzel was the only returning player who had played a major offensive role the previous season.
Generally, improved shooting (both from the field and the line) over the 12-year period have offset the (relative) decline in Michigan State’s rebounding prowess. Field goal shooting and the lowest turnover percentage of the Izzo era allowed the 2004-05 team to reach the national-top-ten heights of offensive prowess of the first three Final Four teams, despite not being as dominant on the offensive glass.
If I had to pick one of the four factors that seems to correlate with the overall trend in offensive efficiency, it’s (surprise, surprise) turnovers. Flip this graph upside down and it look a lot like the offensive efficiency graph:
In looking at this graph again, I’m a little surprised to note that this past season’s turnover percentage of 21.0 percent was actually the fourth lowest of the 12-year period. This is fairly remarkable given my comments with just six games left in the conference season:
I’m out of answers. The Spartans have now played 60 games since the beginning of last season. They have consistently given the ball away on one of every four possessions over that time. The odds they will correct this in the 8+ games that remain this season seem exceedingly small.
As is turned out, the team beat those odds, turning the ball over on less than 20% of their possessions in 9 of their final 11 games. Knock on wood that those 11 games are a sign we’re entering year two of another period of improved ball-handling and overall offensive efficiency . . .
Izzo seemed about as frustrated as I’ve ever seen him as last season went on, regarding the team’s enigmatic turnover issues. In the press, he went light on his players, he lambasted them, took the blame himself, admitted he didn’t really know what was happening – he struggled to find that magic bullet to fire at the problem. It sure looked like things had snapped around down the stretch, especially against quality opponents.
Great analysis, and a couple points.
First, this is just a hunch, but I think MSU’s turnover issues have to do with shot selection. They don’t take a lot of outside shots, but teams that do generally enjoy low TO Rates. I don’t know how much messing with the shot selection will affect eFG, OReb, etc. in Izzo’s offense, but I think some gains can be made by keeping it on the perimeter more. It would be interesting to look at how the TO Rate varied with 3PA/FGA, and if there was a significant correlation.
Second, I’ve forwarded by theory on why the TOs went down toward the end of last season. It wasn’t luck, pre-game speeches, or practice. It was personnel, but I’ve already gone into detail about that earlier.
Good stuff.
BTG, it just seems (maybe only to us Spartan fanatics) that too many turnovers over the last two years came with no apparent purpose–as opposed to being the result of trying to create good shots near the basket. But your explanation is a good theoretical argument regarding the longer-term trend of Izzo teams being mediocre in the turnover percentage department.
I agree wholeheartedly that Lucas getting the keys to the offense was the #1 factor in reducing turnovers at the end of the year.
Oh, I realize a lot of turnovers just seem like boneheaded plays, and have nothing to do with what type of shot the team is trying to get. But think of those as the “base rate” of turnovers. Every team has ‘em. They’re (seemingly) unavoidable. Even a backcourt of Derrick Rose, Drew Neitzel, and Mike Conley would manage to turn it over a few times a game. What kills MSU are the “extra” turnovers that they have. Finding the source of these 3-5 turnovers per game would be very helpful.
But again, this is just a hunch of mine, after taking a quick look at the numbers. There might be nothing to it at all. I might revisit this later on this season.
That’s what my intuitive judgment has been–that the 3-5 extra turnovers have been of the boneheaded (or, at least, unforced) variety, above and beyond the base rate of such turnovers every team commits. Here was my attempt to quantify this:
http://www.spartansweblog.com/2008/02/06/breaking-down-the-turnovers/
The problem with trying to figure out what the problem is is that (1) you need more detailed data on the team’s turnovers and (2) even if you have that data, you need data for other teams to compare it against as a baseline.
Here’s hoping the team’s performance this season won’t compel me to undertake this exercise.
“That’s what my intuitive judgment has been–that the 3-5 extra turnovers have been of the boneheaded (or, at least, unforced) variety, above and beyond the base rate of such turnovers every team commits.”
Ah, yes, Raymar’s strange traveling violations come to mind. I really hope he’s worked on that.
[...] of basketball statistical analysis, there’s one tempo-free stat I overlooked in the series on MSU’s historical tempo-free tendencies on offense: assist percentage. Assist percentage is the number of assists a team records divided by its made [...]
[...] series on MSU’s historical statistical tendencies petered out after I got through the offensive [...]