Expected Point (EP) Calculations Dramatically Enhance Turnover Analysis!: This year I added EP calculations because I thought it would provide a more reasonable measure of the net effect of turnovers. (In previous years I had used an average of 4-5 points per TO based on sites like the Football Outsiders.) The benefit of using EP calculations for turnover analysis has been far greater than I could have ever imagined.
In the WMU game, M returned two TOs for touchdowns and the EP calculations reflected an additional 7.7 point advantage to Michigan (versus using just an average of 4-5 points per TO). This week, the EP calculations actually give Michigan an advantage of 1.91 expected points even though we had a negative one TOM! How could this be? Well, in retrospect, it is fairly obvious. Each turnover may result in a immediate lost opportunity for the team committing the TO and a potential gain in field position by the opponent. Both of these components can vary dramatically based upon the down, the yards the TO is returned, and whether the TO was a fumble or an interception.
For example, in the SDSU game, Denard threw an interception on third and 6 from the M28 and SDSU took over on the M39. If the pass had been incomplete, M would have punted on fourth down and SDSU would have gotten the ball at the SDSU 36 (based on a net punt of 36 yards from the M28). Therefore, the total effect of this TO was only 25 yards of field position which is about 1.00 EP. Contrast that to the SDSU fumble after a 30 yard run to the M07 yard line. SDSU would have had a first a goal at the M07 with an EP of 5.32. Michigan did have an EP of –0.72 because of recovering the ball on the M07 (poor field position) but this still is a net of 4.60 EP. The bottom line is that the M turnovers were far less damaging than the SDSU turnovers and this is reflected in the EP calculations.
In general, interceptions will have less impact than fumbles because an interception usually results in a smaller difference of field position. Also, turnovers on third down will have less impact than TOs on first or second down.
Synopsis for Turnovers: For the first time this year, M had a negative turnover margin. The game had a total of 7 turnovers. Michigan lost 4 and gained 3 from SDSU, leaving M with a TOM of -1 for the game and +5 for the year.
Turnovers did not impact which team won the game (TOs did impact which team won the WMU and ND games).
(See the Section on Gory Details below for how the adjustment for Expected Points (EP) is calculated.)
National Rankings: Remember the table below includes the WMU game and will NOT be the same as the (incorrect) NCAA Rankings. DRob continued his TO problems with 2 interceptions and a total of 6 for the year. Through the first four games last year, Denard had just one interception. Interceptions are ranked #104 Nationally. Overall, the –1.0 TOM for the week dropped M to #12 in TOM for the year.
The Gory Details
Expected Point (EP) Analysis: Basically, the probability of scoring depends on the yard line that the offense is at (which seems fairly obvious in retrospect). Therefore, the impact of a TO also depends on the yard line where the TO is lost and the yard line where the TO is gained.
Here are the details for the game.
The analysis is a bit tricky because: (A) the TO may directly result in lost EP for the offense but (B) only modifies the EP for the team gaining the TO because the team gaining the TO would have gotten another possession even without the TO (due to a punt, KO after a TD, KO after a field goal, etc.). The Net EP Gain must take into account the potential EP gain without the TO. The EP gain without the turnover is based on where the field position would have been for the next possession if the TO had not occurred.
The expected point data has been revised this week to reflect updated data from Brian Fremeau at BCFToys (he also posts at Football Outsiders). Fremeau has updated his data to reflect all offensive possession played in 2007-2010 FBS vs. FBS games. I "smoothed" the actual data.
Here is a summary of the smoothed expected points.
Details for Turnovers: Here is overall summary for all games by player (data in yellow was affected by this week's game).