Turnover Analysis Updated Thru Northwestern

Submitted by Enjoy Life on October 11th, 2011 at 1:48 PM

So, That Was Scary!: For the first 30 minutes of the game, I just sat there in dismay. Despite all the protestations and presentations about how this was NOT 2010, there we were – back in 2010. There was a lot of Gawd Dammits and F this and F that. Then (as the boys at BTN said) "Brady Hoke earned his entire salary for the year at halftime." The space time displacement was not real, Stephen Hawking is right – time only goes in one direction – and we returned to 2011. Whew!

Synopsis for Turnovers: For the second game this year, M had a TOM of -1 (SDSU was the other game). Bad Denard showed up in the first half, threw 3 interceptions, and the TOs were killing Michigan. Good Denard returned for the second half and the defense had 2 takeaways. For the year, Michigan has lost 11 TOs (ranked #84) but gained 17 TOs (ranked #6) for a turnover margin of +6 or 1.00 per game (ranked #15). Michigan is ranked #2 in fumbles lost but is #111 in interceptions thrown. The 12 fumbles recovered is ranked #1 and is the reason the turnover margin is excellent instead of horrible.

Hawthorne added his name to the takeaway list with an interception that I held my breath for the entire duration of the review. There are now 16 different defensive players that have either forced a fumble, recovered a fumble, or intercepted a pass. Thomas Gordon has forced 1 fumble, recovered 3, and intercepted a pass to lead all players in takeaways.

So, What's Up With DRob?: All the data I have looked at indicates that QBs have fewer and fewer TOs as they get older and gain experience with more games played. DRob had a total of 10 interceptions in all of 2010 and he has 9 interceptions in just 6 games this year. Certainly, TOs are the most schizophrenic and irregular statistics in all of football, but Denard's interceptions do warrant some examination. I believe (hope?) that a significant number of the interceptions are a result of the change in coaches and offensive approach. Basically, Denard is learning a new system and, therefore, the learning curve that would be expected in his second full year at QB will be delayed. We can expect more TOs from Denard until he is comfortable with the new offense. Stay tuned.

imageSynopsis for Expected Point (EP) Analysis: Obviously, when the winning team has a negative TOM, turnovers did not impact which team won the game. Michigan turnovers in the first half were the reason for that 10 point deficit, but the two takeaways in the second half resulted in just a 4.06 disadvantage for the overall game. The table summarizes expected points lost by TO's, expected points gained by opponent's TOs, the net EP due to TOs, and the adjusted score without turnovers. BTW, the defense played great in the second half and even without the takeaways, M should have still won the game (but it would have probably been a nail biter).


(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.



The Gory Details

Details for Turnovers: Here is overall summary for all games by player (data in yellow was affected by this week's game).


Expected Point (EP) Analysis: Basically, the probability of scoring depends on the line of scrimmage for the offense. 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. Each turnover may result in an 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 when the TO occurred, the yards the TO is returned, and whether the TO was a fumble or an interception.

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.


image_thumb17_thumbThe expected point calculations are based on data from Brian Fremeau at BCFToys (he also posts at Football Outsiders). Fremeau's data reflects all offensive possessions played in 2007-2010 FBS vs. FBS games. I "smoothed" the actual data.

Here is a summary of the smoothed expected points.




October 11th, 2011 at 2:30 PM ^

I'm not sure we've exactly framed a Denard-Interception theory this way before: is it possible he has the same 'accuracy' this year as last year, but that the 'system' and/or his 'decisions' are responsible for the increase interception rate this year?

To clarify: on passing plays last year, it's not obvious to me that he had to 'read' the defense or 'go through a progression' of receivers. The passing plays were designed to get a guy open based on the threat of Denard's legs, and he simply was instructed to throw it to that guy, who was indeed, typically open. This year he has to worry about dropping back, setting his feet, going through a progression of reads, timing the throw, dealing with actual coverage, etc.

Enjoy Life

October 11th, 2011 at 3:03 PM ^

Yes, this is similar to what I was saying. Denard had a year to get comfortable with the previous scheme (that may have also been simplier) and is now trying to get up to speed on a new scheme (that may be more complicated).

End result is that the learning curve for this year is not as steep and the turnovers will be with us for a while.


October 11th, 2011 at 3:40 PM ^

Why are there more fumbles recovered versus fumbles forced?  Have there really been 4 times where the opposing team simply dropped the ball without Michigan doing anything?  I only remember a handoff during either the Western game or SDSU, and the Rees dropped ball in the ND game.

Enjoy Life

October 11th, 2011 at 5:53 PM ^

The scoop and score by Avery in the Minnesota game was off a botched handoff -- therefore, no force on the fumble.

Also, if there is a bunch of folks tackling the player, often no one is given credit for the force on the fumble.

And, I believe it is sometimes just a judgement call. So, if a tackle is made by one player and a fumble occurs, you would think this would be credited as a forced fumble -- but it sometimes is not.


October 11th, 2011 at 8:45 PM ^

What other game (besides the obvious ND game) did UM win due to TO's? That doesn't seem (intuitively) correct.

WMU +3 TO won by 24

EMU +1 TO won by 28

Minnesota +1 TO won by 58

Was the +3 vs. WMU really worth 24+ points? That seems hard to believe.


Enjoy Life

October 11th, 2011 at 9:44 PM ^

I included the WMU game. Using expected points, the 3 TOs were worth 22 EP. The reason this is so high is because of the 2 TOs that Michigan returned for TDs.

Michigan did win by 24 points (34-10) so based on a purely mathematical calculation, you could conclude that TOs were not a determining factor. However, evaluating TOs is certainly not an exact science and I concluded that without the TOs the game would have been too close to call. Thus, I put the WMU game in the category of TOs being a significant factor in the win.