For example, if on 3rd and 10 a quarterback throws a 40-yard pass that is intercepted and not returned, it goes in the books as a turnover but really doesn't change the game any more than an incompletion, because the offense would punt on the next play.
Some events probably should be counted as turnovers in the sense that they result in a change of possession without any accompanying gain in field possession. Blocked punts and failed 4th down conversions are two good examples of this. But, for a variety of reasons, they aren't so counted.
Thus, turnover margin is not a perfect statistic, but it is an important one. Over the course of a season, a team is likely to have a fair proportion of significant and insignificant turnovers, but in the end being positive on turnover margin is likely to help a team.
So, how much impact do turnovers make?Consider the following data, which I compiled from the official NCAA statistics for 2000 and 2001. The left hand number is a team's turnover margin per game and the right hand number is that team's win rate projected to a 12-game season.
+ 1 and up: 9.26
+ 50 to .99: 8.04
+ 00 to .49: 7.04
- .01 to -.50: 4.86
-. 51 to -.99: 4.24
- 1.00 & lower 3.20
The correlation is striking and quite linear. Teams that average +1 or more per game (+12 or more per season for a 12-game season) win almost 80% of their games and teams that average –1 or worse per game win just over 25% of their games. On average, over the course of a season, every +5 on turnover margin buys a team about one more win. Every –5 on turnovers costs a game.
Often, therefore, teams that have large positive turnover margins are among that year's most pleasant surprises. Here's the list of teams who were positive by double digits on turnovers for the regular season in 2001, along with their regular season record: Miami (+26, 11-0), Fresno St. (+23, 11-2), Bowling Green (+17, 8-3), Maryland (+16, 10-1), Oregon (+14, 10-1), U.S.C. (+14, 6-5), Syracuse (+15, 9-3), Iowa St. (+12, 7-4), Purdue (+12, 6-5), Ohio St. (+11, 7-4), Washington St. (+11, 9-2). The list, of course, includes undisputed national champion Miami and #2 Oregon (neither of which was really a surprise) but also includes some teams that apparently came from nowhere: Maryland, Washington St., Syracuse and perhaps Iowa St. and Bowling Green.
Of course, being positive on turnovers is a good thing and very often they result from good play, but sometimes there's luck involved. Consider that N.D. this year has fumbled 8 times but only lost 4. Duff's fumble of the second-half kickoff against Maryland magically wound up in Yura's arms and Dillingham's fumble against Stanford was alertly recovered by N.D. N.D.'s opponents, on the other hand, have fumbled 6 times and lost every single one of them, including, for example, the collision between two Purdue returners where the ball popped up and fell into Bolen's hands for a touchdown.
Certainly, the huge majority of N.D's forced turnovers have had nothing to do with luck. There was nothing lucky about Watson ripping the ball from the Stanford receiver's arms and returning it for a T.D., there was nothing lucky about Duff jumping the Purdue route late in the game and nothing lucky about the suffocating coverage that Walton has played all year.
For a team to reach the highest levels, it has to have excellent underlying fundamentals (good per rush and per pass margins) and be positive on turnovers. Consider N.D.'s turnover margins in its national championship years: 1966: +8, 1973: +9, 1977: +13, 1988: +14. Those teams had dominant fundamentals and a good turnover margin put them over the top. On the other hand, the 2000 team was +14 on turnover margin but had very ordinary per-play fundamentals and wound up with a record (9-2) that considerably overvalued the team, a fact painfully demonstrated in the Fiesta Bowl.
Let's consider how N.D. compares thus far in the season with the other undefeated teams. Of course, the level of competition has varied considerably, and N.D. has had, by far, the most difficult schedule of any undefeated team. In parentheses after the team's name, I give the ranking of its schedule difficulty according to Sagarin. Here are each team's per rush and per pass margin and the turnover margins:
Virginia Tech (48)
T.O. + 11
Ohio St. (60)
Rush + 2.8
Pass: + 1.1
Pass: + 2.9
Pass: + 1.0
N.C. St. (105)
Air Force (79)
N.D. is towards the bottom end on per-play averages of the undefeated teams. Texas and N.C. State have weaker per-rush advantages and N.D. is at the bottom on per-pass margin, but isn't far behind Georgia, Air Force and Ohio St. in this department. And, of course, N.D. is alone among these teams in not facing a single truly weak opponent, hence the vastly higher schedule difficulty rank.
N.D. is second in turnover margin to Virginia Tech., but Oregon, Texas and Oklahoma are all just one behind. Thus, there is nothing outlandish about N.D.'s turnover margin relative to the other undefeated teams.
At the other end of the spectrum, Miami is even, but has so thoroughly dominated opponents thus far that it hasn't made any difference. Note, however, that Miami''s +26 last year played a large role in winning the national championship. An interception saved Miami in the Boston College game and being +2 on turnovers in the 2-point game against Virginia Tech probably bailed the Canes out there as well. Georgia is actually negative 1 on turnovers, and is probably the team that's the luckiest to be undefeated. They've had three close games against their three quality opponents, beating Clemson by 3, South Carolina by 6 and Alabama by 2.
If you're looking for teams ready for a fall in the bunch, my three nominees would be N.C. State, Texas and Oregon. N.C. State has rolled up big passing margins but very little on the rushing side, and has faced only two teams that even border on being quality opponents: Wake Forest and Texas Tech. The Wolfpack handled Wake Forest fairly easily but was extremely fortunate to beat Texas Tech in a 51-48 come-from-behind affair. Easy victories over New Mexico, Navy, East Tennessee St. and Massachusetts do little suggest that N.C. State will do well it faces North Carolina, Georgia Tech, Clemson, Virginia, Maryland and F.S.U., all of which remain on its schedule.
As for Texas, the Longhorns have amassed less-than-spectacular rushing margins against very weak opposition. Texas's only two opponents that even approach being a quality opponent are Oklahoma St. (a team that lost to Louisiana Tech and whose only victories were against Northern Iowa and S.M.U.) and North Carolina (currently 2-3 including a loss to Miami of Ohio). And, of course, Texas was extremely fortunate to beat Oklahoma St. 17-15. It's hard to imagine the Longhorns beating Oklahoma, Kansas St., Iowa St., Nebraska, Texas Tech. and Texas A&M (all of which remain on their schedule) without stepping up their play considerably.
Oregon's rushing margin has also been weak considering the ease of the Ducks' schedule. They beat Mississippi St. and Arizona, but both of those teams are headed for very low finishes in their conferences, and Arizona gave Oregon a game for quite a while (it was 14-14 with less than a minute to go in the 1st half) before losing 31-14. Oregon also struggled at home to beat Fresno St. 28-24 and their other two games were against I-AA teams.
Oklahoma and Miami have also faced weak schedules on the whole, but Miami has run up huge margins and has convincing victories over Florida and Boston College, and Oklahoma has victories over Alabama and Missouri (though both games were close) to go with its big margins. Next week's Oklahoma-Texas game will be extremely revealing, but I expect Oklahoma to win the game.
But back to N.D.: this is a very good team. It is not a mirage of any sort; N.D. has faced the most difficulty schedule of any undefeated team and the combination of per-play and turnover margin shows that they deserve to have a goose egg in the loss column.