Tabling The Issue: Turnovers Over Time

One of the metrics I wanted to analyze as we head into Spring practice is turnovers. Again, I entered the phenomenal nerdatorium at www.und.com and went through the data going back to 1986. Holtz. Davie. Willingham. Weis. Kelly. Is there a correllation between turnovers and success? What is success? Great question. For the purposes of this article, I defined success as nine or more wins in a season. And what’s a turnover? A turnover is a fumble lost or an interception. So, with those definitions in mind, please, if you will, this:

YEAR COACH FUMBLES LOST INTERCEPTIONS
1986 HOLTZ 13 9
1987 HOLTZ 9 11
1988 HOLTZ 13 9
1989 HOLTZ 16 10
1990 HOLTZ 12 6
1991 HOLTZ 11 10
1992 HOLTZ 11 8
1993 HOLTZ 5 5
1994 HOLTZ 10 11
1995 HOLTZ 12 8
1996 DAVIE 23 4
1997 DAVIE 5 8
1998 DAVIE ND ND
1999 DAVIE 15 15
2000 DAVIE 4 4
2001 DAVIE 12 11
2002 WILLINGHAM 15 13
2003 WILLINGHAM 7 19
2004 WILLINGHAM 6 10
2005 WEIS 6 8
2006 WEIS 7 7
2007 WEIS 16 9
2008 WEIS 11 17
2009 WEIS 9 5
2010 KELLY 8 16
2011 KELLY 12 17
2012 KELLY 7 8

One of the reasons I went back so far is because we lionize Lou to such an extent that I wanted his years under the microscope, as it were. But what leaps out from his years, 1986-1995, are the relatively high numbers of turnovers. Take the last National Championship season, 1988. 13 fumbles lost and 9 picks and a 12-0 record. Compare that to Brian Kelly’s 2012 season, also his third, in which he lead the Irish to a 12-0 regular season, but only gave away 7 fumbles lost and 8 interceptions. The horror show that was 2011, a disappointing 8-5 that saw some among us calling for Kelly’s head, saw one fewer fumble slip from Irish hands, 12. But look at those interceptions from 2011….17???

To return to the Holtz years for a moment and round the issue out, those seasons seem anomalous when compared to seasons 1996 through 2012.  The turnover numbers, fumbles lost and interceptions, are consistently higher from 1986 to 1995 and Holtz’s 1989 season (16/10) (12-1) is only approached by Charlie Weis’s 2007 Stalingrad (16/9) (3-9). The hypothesis with which I will go forward is that something in the game changed such that from 1996 to 2012, only one ND squad with more than 15 turnovers in a season, Ty Willingham in 2002 (15/13), had more than eight wins (10-3).

In case you’re wondering how the turnover thing plays out against (a sample of the most recent part of) the historical field, 2012′s champs, Alabama, had 15 turnovers (12/3), 2011′s champs, Alabama, had 12 turnovers (8/4), 2010′s Auburn had 17 (7/10), 2009′s Alabama had 12 (7/5), 2008′s Florida had 13 (5/8) and 2007′s LSU had 16 (3/13).

I’m comfortable that 15 is the magic number for success, at least for the Irish. Somewhere, someone (looking at you Brian Fremeau) has the data on when (and where) the turnovers happen. That, to me, is the key. Like this:

So. What do you think?

 

  • NDtex

    I think another variable in success in the turnover equation is also turnovers forced, leading to turnover ratio.

    Especially in 2012, the Irish had some very timely turnovers to help seal games in our favor.

    Good work.

  • GB

    One thing that struck me was the ’93 season. 10 TOs. Weren’t there 5 TO just in the BC game. It does seem to depend on where and when.

  • Scott

    You’ve got Davie as the coach of the ’96 team