that is nice bonus change
Improvement '09: Turnovers?
In which your humble blogger grasps at straws trying find a way, any way, Michigan will improve from last year. Previously in this series: Snap Counts.
It's no secret that Michigan was a bad team in several aspects of the game last year. Nearing the top of the list was theteam's propensity for giving the ball to the other team. They finished the year -10 in turnover margin, good for #104 in the nation. They lost an amazing 18 fumbles (only Army, Georgia Tech, East Carolina, and Wisconsin did worse), while throwing 12 picks. It's one of the fundamental predictive tools of this blog to assume that turnovers, especially fumbles, are essentially random, and will not carry over from one year to the next. While Michigan fans are praying that holds true, let's take a more in-depth look at Michigan's turnovers in 2009, and see where the improvement might come from:
|Michigan Turnovers 2008|
|Opponent||Score||Fumbles +/-||Interceptions +/-||Turnover Margin|
|Utah||L 23-25||2-2 (0)||1-1 (0)||3-3 (0)|
|Miami||W 16-6||2-1 (+1)||0-0 (0)||2-1 (+1)|
|Notre Dame||L 17-35||0-4 (-4)||2-2 (0)||2-6 (-4)|
|Wisconsin||W 27-25||2-3 (-1)||2-2 (0)||2-3 (-1)|
|Illinois||L 20-45||0-2 (-2)||0-0 (0)||0-2 (-2)|
|Toledo||L 10-13||1-0 (+1)||0-3 (-3)||1-3 (-2)|
|Penn State||L 17-46||1-1 (0)||0-0 (0)||1-1 (0)|
|Michigan State||L 21-35||3-1 (+2)||0-3 (-3)||3-4 (-1)|
|Purdue||L 42-48||1-1 (0)||0-0 (0)||1-1 (0)|
|Minnesota||W 29-6||0-1 (-1)||1-0 (+1)||1-1 (0)|
|Northwestern||L 14-21||0-1 (-1)||2-1 (+1)||2-2 (0)|
|Ohio State||L 7-42||0-2 (-2)||1-0 (+1)||1-2 (-1)|
|Total||(3-9)||12-19 (-7)||9-12 (-3)||21-31 (-10)|
What can we learn? The passing offense for Michigan was much less apt to turn the ball over when Threet was playing over Sheridan (we already knew that, of course, as 3.65% of Sheridan's passes were picks, compared to 3.50% for Threet). That said, neither of the QBs was at risk of lighting opposing secondaries on fire.
Much more interesting for the future is the decline in turnovers over the year. Part of this comes with experience, and part is just plain luck. However, Michigan's turnover margin through the first 6 games was -8, and it was just -2 over the final 6 games. That's improvement over the course of the year.
Michigan was 1-0 (1.000) winning the turnover margin, 1-4 when tied (.200), and and 1-5 when losing it (.167). Though this isn't a great correlation, it shows that turnovers can help lead to wins (which should be obvious).
PersonnelWhen Michigan's turnovers happened hurt them somewhat, but who committed the turnovers may be even more enlightening. It appears as though the vast majority of turnovers were committed by young players. With an upgrade in experience at every position other than quarterback (where there is hopefully a talent increase), maybe Michigan can see their turnover number decrease correspondingly.
|Michigan Fumbles 2008|
|Player||Class||Number (lost)||2009 Status|
|Steven Threet||RFr||13 (6)||Gone (ASU)|
|Martavious Odoms||Fr||6 (5)||Soph|
|Sam McGuffie||Fr||3 (2)||Gone (Rice)|
|Boubacar Cissoko||Fr||3 (1)||Soph|
|Greg Mathews||Jr||2 (1)||Sr|
|Nick Sheridan||RSo (W)||2 (0)||RJr|
|Michael Shaw||Fr||1 (1)||So|
|Kevin Grady||RJr||1 (1)||RSr|
|Brandon Minor||Jr||1 (1)||Sr|
|Avery Horn||RFr||1 (0)||Gone (JC)|
|Mark Moundros||RSo (W)||1 (0)||RJr|
|Donovan Warren||So||1 (0)||Jr|
There is a discrepancy between the number of fumbles in the first and second charts because the NCAA reports two different numbers, which I couldn't resolve. The NCAA also attributes one fumble to Minor, which I switched to Threet, since it was an inaccurate lateral pass.
As you can see, the vast majority of Michigan's fumbles were committed by freshmen. Steven Threet, in particular, was a fumble-committing machine. Some of the fumbles were based on pressure, but Nick Sheridan took nearly as many snaps as Threet last year, and committed far fewer fumbles. Some of it has to be a particular skill by Threet, which is unlikely to be repeated by whomever takes snaps for Michigan in 2009 (especially with better protection).
17 fumbles (8 lost) were coughed up by players who will no longer suit up for the Wolverines. Among the remaining fumbles, 13 (9 lost) were committed by freshmen. Assuming that they will be more capable of holding onto the ball now that they're used to the college game, Michigan should see a slight improvement in that regard.
And it Means?As Phil Steele (and Brian, and anybody else who has a clue what's going on) says, teams that turn it over a bunch are unlikely to repeat that performance in subsequent years. In Michigan's specific instance, more college experience for the fumblers could mean much, much improvement. If the freshman quarterbacks can be anything resembling competent, there will be a huge step forward.