Turnover Analysis Updated Thru ohio

Submitted by Enjoy Life on November 29th, 2011 at 2:32 PM

That Should Have Been Easier: Although the official stats indicated a turnover margin of –0-, the blocked punt and the meaningless turnover at the end of the game meant that M really was at a disadvantage in TOM of –2. This calculated to be a disadvantage of 5.5 expected points. Add in the reversed call on M's last touchdown and we all had a far more stressful afternoon than it should have been. In a way, though, it validated that Michigan is a very good team this year because very good teams manage to win even when things don't go your way.

Synopsis for Turnovers: M ended the game with a TOM of zero. For the year, M has had 6 games with a positive TOM, 4 games with a negative TOM and  2 games with a zero TOM. Michigan has lost a total of 21 TOs (ranked #62) but has gained 27 TOs (ranked #19) for a turnover margin of +6 or 0.50 per game (ranked #25). Michigan is ranked #7 in fumbles lost but is #119 in interceptions thrown. The 19 fumbles recovered is ranked #3 and is the reason the turnover margin is excellent instead of horrible.

Avery forced a fumble and intercepted a pass (his 2nd). DRob had the one fumble.

BTW, blocked punts are not considered a turnover.

imageSynopsis for Expected Point (EP) Analysis: Michigan has a net of 1 game won due to TOs. image

(See the Section on Gory Details below for how the adjustment for Expected Points (EP) is calculated.)

imageNational Rankings: OMG, now that the season is over, the NCAA has decided to include the statistics from the WMU game! So, now the chart and table are the same as the (corrected) NCAA stats.


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_thumb_thumb_thumb_thum[1]The 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.




November 29th, 2011 at 4:10 PM ^

Seriously . . what gives?  Why include the stats now?  Is it because they want to justify a 10 win team in a BCS game above other 10 win teams?   I know, that's a lot of question marks.


Edit:  Thank you to Enjoy Life for doing this column all year.


November 29th, 2011 at 4:20 PM ^


I'm a fan of using statistics/math to analyze sports. For example, I like that you can use a collection of relevant data to predict how a single play changes your odds of winning. I like that there exists some ability to rate a team objectively by removing some of the random in-game events that can alter the average follower's perception of that team. Even the BCS algorithms have some good stuff in there.


However, I am not a huge fan of EP analysis for the simple reason that future plays are a direct response to every event occurring previously. Therefore, it doesn't make much sense [to me] to sum EP differentials. I haven't been around the diary section too much, but I would guess that this has been stated previously. Even Brian said something about adding win percentages in his "Get Behind Me" post being a no-no but proceeded to do it anyway.


Regardless, thank you for doing this piece. It is a really nice feature that I've followed with some regularity. Go Blue!

Enjoy Life

November 29th, 2011 at 4:43 PM ^

Yes, any analysis of sports is inherently flawed. And, I have made major changes in trying to understand the impact of turnovers. But, simply counting turnovers or using an "average" value for each turnover seems more flawed than most other analysis. (The only analysis that is worse is looking at turnover margin for an entire year -- yes, I mean you Mr. Steele.)

For now, Expected Points seem to be the most reasonable at reflecting the factors involved in the impact of a turnover by considering down, distance, type of turnover, etc.

I will be doing a year-end review next week and will be making some more changes to better understand turnovers and turnover margin.


December 2nd, 2011 at 9:45 PM ^

enjoy life, what database do you get your turnover information from? I was interested in potentially doing a study to see if there's any correlation between style of offense and turnovers created. thanks in advance.