2013 Turnover Analysis – Updated thru Akron
Quarterbacks!: In 2011, quarterbacks were responsible for 19 of 21 turnovers. In 2012, quarterbacks were responsible for 23 of 25 turnovers. So far in 2013, quarterbacks are responsible for 8 of 8 turnovers. Any improvement in giveaways is squarely on Gardner's shoulders. If Michigan does not correct the turnover problem, it is likely that turnovers will be the primary cause of 1-2 losses this year.
Synopsis: Michigan's TOM for the game was – 2 and for the year it is now – 2 (– 0.67 per game) which is ranked #96. Turnovers were not a primary factor in determining which team won the game. In fact, the game would have been very close even without the turnovers. Michigan suffered a net disadvantage of just 2.72 expected points due to the – 2 turnover margin (see details below in section on Expected Points).
Countess picked up his third interception of the year and Wilson got his first. The Michigan defense has a total of 5 interception takeaways and is ranked #33 for interception takeaway percentage at 3.9%. Gardner threw three interceptions and now has 6 for the year. M is ranked dead last (#125) for interceptions thrown percentage at 8.3%. Devin also lost his first fumble of the year.
Player Details: Here is the overall summary for all games by player (data in yellow was affected by this week's game). .
National Rankings: All rankings include games between two FBS teams ONLY and are from TeamRankings except for forced fumbles which is from CFBStats. 
The four columns with *** show the best correlation to offense and defense (per Advanced NFL stats).
Turnovers And Winning: This chart shows turnover margin (TOM) at the end of the season versus the percentage of teams with a +4 WLM (8-4 record) or better. [WLM = Win/Loss Margin = Wins – Losses]
Expected Points: The impact of each turnover depends upon the down, the spot the turnover is lost, and the spot the turnover is gained. Although Michigan had a –2 TOM, two of the turnovers occurred on third down. None of Akron's turnovers occurred on third down. As you can see in the table below, both of Akron's turnovers were very costly with an EP of nearly 6 points each. Michigan had one turnover that was very costly at 5 EP, one at 4 EP, and two at just 3 EP. The result is a net disadvantage to Michigan of just 2.72 EP.
This chart shows Expected Points for various yard lines.
This chart shows the basis of EP calculations for each turnover.
September 17th, 2013 at 4:56 PM ^
Normally I feel that the EP metric works pretty darn well to assess the overall value of turnovers in a game but this week I don't think it adequately captures the momentum shifting impact they had.
In general Michigan was able to offset the tactical impact of its own turnovers with the picks, true, but I think Gardner's fumble on the five was not just lost points but a lost opportunity to put an early dagger in to Akron, after which I think the entire game would likely have played out a bit differently.
Either way, we aren't going to win many football games against better teams when we turn the ball over four times per game. Ouch.
September 17th, 2013 at 5:54 PM ^
During the game, I also thought the turnovers were more of an impact. In fact, I was shocked when the calculations came out the way they did.
I think the Mathlete did an analysis last year (perhaps the year before?) that indicated the change in momentum due to turnovers was not very significant. But, it seemed like the TOs hurt M a lot more than the EP indicates.
Comments