clock menu more-arrow no yes mobile

Filed under:

Beyond the Box Score: Turnovers

In most of my previous BTBS posts, I've acknowledged that the whole thought behind my EqPts measure (and therefore the PPP and S&P measures as well) is only one part of scoring points.  It's the most important part, but there are other factors involved--namely Turnovers and Special Teams (and luck, but we're not measuring that).

The main questions are a) how much of an impact do TO's and Special Teams actually make, and b) how are we going to measure that?  Well, it's time to look into that.  Today, we'll look at turnovers.

In a previous BTBS glossary, I talked about measuring Turnover Costliness this way:

Each turnover is assigned two values: 1) the point value (see below) of the offense's field position at the time of the turnover, and 2) the point value of the resulting starting field position for the opposition.

Turnover Costliness = (0.75*the higher of the two values)+(0.25*the lower of the two).

I previously had a factor in here regarding closeness of the game, and I'm sure I will again, but for now this is what I'm working with.

Let's throw that idea out.  What happens if we count both values (the value of the offense's field position at the time of the turnover and the value of the resulting starting field position for the opposition) fully, combining the two to gauge the 'points' involved in a given turnover. Where does that take us?

For one thing, it means we're looking at quite a few different numbers here.  For every time your offense turns the ball over, you've got a "Points Lost" number (your own field position at the time of the t/o) and a "Points Given" number (opponent's resulting field position).  For every time your defense benefits from a takeaway, you have a "Points Prevented" number (your opponents' field position at the time of the t/o) and a "points Taken" number (your resulting field position). Obviously Points Lost and Points Prevented are the same number (depending on whether you're on offense or defense), as are Points Given and Points Taken.

Let's quickly look at the best and worst from each category, then try to figure out what this all means.

(And through all of these numbers, realize this--I also count botched punts/field goals as turnovers, so my Turnover Margin figures will likely be different than the official NCAA stats.)

Points Lost - Best (who didn't turn the ball over in scoring opportunities)

1. Clemson (1.69/game)
2. Ball State (1.84)
3. Texas A&M (1.93)
4. East Carolina (1.99)
5. LSU (2.06)

Points Lost - Worst (who did)

120. Western Kentucky (8.52)
119. Troy (7.64)
118. Western Michigan (7.32)
117. Idaho (7.07)
116. Baylor (7.06)

Points Given - Best (whose t/o's didn't set their opponents up with easy points)

1. Boise State (2.34)
2. Iowa (2.38)
3. Kansas (2.52)
4. Clemson (2.72)
5. LSU (2.92)

Points Given - Worst (whose did)

120. Florida International (10.49)
119. Idaho (9.80)
118. Western Kentucky (9.48)
117. NC State (8.66)
116. Baylor (8.38)

Points Prevented - Best (whose defenses' Bend-but-Don't-Break styles included forcing timely turnovers)

1. Western Kentucky (11.32)
2. South Florida (7.96)
3. Cincinnati (6.83)
4. West Virginia (6.65)
5. Central Michigan (6.40)

Points Prevented - Worst (whose didn't)

120. Marshall (1.44)
119. Fresno State (1.97)
118. Georgia Tech (2.14)
117. NC State (2.37)
116. Nebraska (2.38)

Points Taken - Best (who set their offenses up with easy scores via turnovers)

1. Western Kentucky (12.03)
2. Oregon State (10.11)
3. UCLA (8.85)
4. South Florida (8.84)
5. Wake Forest (8.37)

Points Taken - Worst (who didn't)

120. Marshall (1.49)
119. Fresno State (2.31)
118. Tulane (2.52)
117. Northwestern (2.62)
116. Nebraska (2.95)


  • I'll say this for Western Kentucky: their games must have been super-exciting.  There were huge, timely turnovers on both sides of the ball, at all parts of the field.
  • On average, the Points Taken/Given per-game numbers (5.50) were higher than the Points Lost/Prevented numbers (4.43), which suggests that more turnovers took place in the offense's territory.
  • It quickly becames clear that the teams on the 'best' lists were much more likely to be successful than the teams on the 'worst' lists.
  • Where does Missouri fall, you ask?  They were #26 in Turnover Points Margin and #19 in Turnover Margin.  They benefited from turnovers overall, but not so much that they can likely expect a host of bad luck in 2008.

So we can break up a simple "Turnover Margin" figure into four parts--is that really useful?  Think about it this way: thanks in part to Phil Steele's "Turnovers = Turnaround" concept, gauging teams' potential turnarounds from one season to another relies in large part on their past season's Turnover Margin.  That's great--turnovers are partially arbitrary (the nature of playing a game with an oblong ball that bounces unpredictably) and there's data to back up the usefulness of that concept.  But right now, the following two turnovers count for exactly the same:

  • Turnover #1: Illinois is about to score against Missouri in the 2007 season opener.  Backup QB Eddie McGee carries the ball inside the 1 before he's stripped and fumbles...the ball bounces into the arms of Pig Brown, who carries it 100 yards for a TD.
  • Turnover #2: Team A fumbles at their 45 yard line, and Team B recovers.  Team B will immediately go 3-and-out and punt the ball back to Team A.

Using my numbers, Turnover #1 was worth 12.62 points (Illinois having the ball at the Mizzou 1 was worth 5.62 points, while Mizzou returning the fumble for a TD was worth, obviously, 7.00).  Turnover #2 was worth 4.26 points (1.92 points lost/prevented + 2.34 points given/taken).  Is that not a much more accurate read of which turnover truly impacted the result of the game and which did not?  So looking at these point readings can give us a much more accurate feel for teams' "Turnovers = Turnaround" potential in 2008.

So let's look at the teams with the best and worst Turnover Points Margins.

Best Turnover Points Margins

1. Kansas (+8.38 points per game) (tied for #1 in regular turnover margin)
2. Middle Tennessee (+7.48) (#3)
3. East Carolina (+7.43) (tied for #1)
4. LSU (+6.96) (#4)
5. West Virginia (+6.42) (#8)
6. Oklahoma (+6.36) (#19)
7. Ball State (+6.07) (#6)
8. UConn (+5.53) (#5)
9. Western Kentucky (+5.35) (#12)
10. Virginia Tech (+5.32) (#10)

Not a lot of variation from the typical Turnover Margin numbers there, though my numbers do show that Oklahoma benefitted a lot more from their takeaways than they were hurt by their turnovers.  What about the teams with the worst margins?

Worst Turnover Points Margins

120. NC State (-8.87 points per game) (tied for #116)
119. Florida International (-8.67) (#119)
118. Idaho (-8.43) (#113)
117. Baylor (-8.16) (tied for #116)
116. Nebraska (-6.58) (tied for #116)

Okay, so list at the bottom isn't all that different from the 'real' Turnover Margin list.  So let's look at the team whose Turnover Margin and Turnover Points Margin were vastly different (in the rankings)...see what that tells us.

Hawaii: #40 Turnover Points Margin, #85 Turnover Margin (difference: 45)
Akron: #101 Turnover Points Margin, #62 Turnover Margin (difference: 39)
Mississippi State: #89 Turnover Points Margin, #62 Turnover Margin (difference: 27)
Louisville: #74 Turnover Points Margin, #48 Turnover Margin (difference: 26)
Houston: #80 Turnover Points Margin, #105 Turnover Margin (difference: 25)
UL-Monroe: #95 Turnover Points Margin, #70 Turnover Margin (difference: 25)

Other than showing that Hawaii really was pretty lucky to go undefeated (their turnovers were less impactful than their takeaways), this doesn't tell us a lot.  In all, though, the average difference between a team's Turnover Points Margin ranking and its Turnover Margin ranking was 7.68, which isn't a tremendously hefty difference, but it does tell us some things about some teams, like Hawaii (likely due a turnaround in luck), Kansas (almost certainly due a turnaround in lucky), et cetera.

(Is 'impactful' a word?  I don't think it is.)

And beyond all that, we've covered one of the two remaining concepts that impact the scoreboard.  We can now assign an actual point value to different turnovers of different types, which I find pretty damn cool.

Next up: Special Teams.