2013 Turnover Analysis – Updated Thru Central Michigan
Why Bother?: In Brady Hoke's first year, Michigan was +1 in games where TOs were a primary factor. Last year Michigan was –1 in games where TOs were a primary factor. The difference between 10-2 and 8-4 was basically due to turnovers.
Recap of Last Year: Michigan ended 2012 with a Turnover Margin (TOM) of –0.67 per game ranked #101 nationally. Using Expected Points (EP) as the basis, TOs were a primary factor in one win (MSU) and two losses (ND & Ohio). Interceptions were THE problem with M ranked #123 in percentage of interceptions thrown and #75 in percentage of passes intercepted. Fumbles were NOT the problem with the offense ranked #27 in total fumbles and the defense ranked #50 in forced fumbles.
Once Is An Accident?: It's only the first game and, unlike many of the talking heads, I will not pretend any of this is a trend. That said, interceptions thrown percentage for the first game was just terrible, horrible, ugly at 14.3%! (Last year Int% was 6.0% and that was #123 nationally.) Watch this space to see if next week represents a trend and week #3 becomes a problem.
Synopsis for Turnovers: 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).
September 3rd, 2013 at 6:04 PM ^
Great to see this feature back this season - thanks for the effort!
Fumbles I take to be relatively random events but interceptions clearly are not. The big question here is whether Devin Gardner is essentially a really good first year quarterback who will throw almost as many picks as TDs or more like a second year QB who shows improvement.
Small sample size, of course, but the early returns are not promising. I will hold out hope that he was just super jazzed on the first one - whenever you saw him come in to spell Denard before he got the job he would always dance around like crazy when he left the field - I'm sure he was nervous/uber pumped, etc. And I can believe that the second one may have been at least partly caused by the hit, but I have to say I still didn't feel as comfortable with him passing as all the off season hype had led me to hope. We could be in for a couple more games decided by turnovers this year against the better defenses (assuming there are any in the B1G).
(Also, not to nitpick but you have 'WMU' instead of 'CMU' in your chart.)
September 3rd, 2013 at 7:53 PM ^
Thanks for the catch. I'll fix the WMU goof.
September 4th, 2013 at 9:58 AM ^
Norfleet lost a fumble? Was this the botched special teams play? If so, it wasn't a lost fumble as Delano Hill recovered.
September 4th, 2013 at 11:19 AM ^
In the chart the first column (FMB) shows all fumbles by player and the second column indicates if the fumble was lost (FL).
September 4th, 2013 at 12:16 PM ^
The header confuses that, as it is labeled TO LOST. Interesting data, but as is the case when I crunch numbers, sometimes you have to get some proofing... I'm not complaining though, it's a good diary.
September 4th, 2013 at 1:34 PM ^
Sorry, reading is fundamental.
Comments