As good as Holtz's teams were at run blocking and running the ball, they could not have won consistently running the ball into eight and nine man fronts all day. Holtz understood this and he had plays that stretched the field both horizontally and vertically. His preferred method of stretching the field horizontally was the option.
But often overlooked was that he stretched defenses vertically with deep passes. Although Holtz's teams didn't throw often (about 30% of the time over his career at ND), they threw enough and threw it effectively.
The most important statistic in evaluating the effectiveness of a passing attack is yards per passing attempt. Passing is always a risky proposition. The chances of a turnover are about three times that of a running play (if one factors in fumbles on sacks) and the chances of a zero gain (incompletion), loss (sack) or interception are over 40 percent for even the most accurate of passers. So, there needs to be something in return in terms of gained yardage.
An effective running attack and a "stretching" passing attack complement each other. For example, in ND's national championship year of 1988, the Irish averaged 4.8 per rush and 8.4 per pass.
In 1989, they averaged 5.1 per rush and 7.9 per pass. In 1992 they averaged 5.6 per rush and 8.1 per pass. In 1993 they averaged 5.5 per rush and almost 10 yards per pass. All of those teams won major bowls and finished 1, 2, 4 and 2 in the polls respectively.
To say things have fallen off a bit since then is an understatement. In 2001, Davie's last year, ND averaged 3.9 per rush and 5.1 per pass (dead last in the nation in the latter category).
In 2002, Willingham's first year, the passing average improved to a respectable 6.6 per pass (6.9 is about the NCAA average) but only 3.4 per rush (3.8 is about the NCAA average).
In 2003, N.D. improved to 4.0 per rush but was a dismal 5.2 per pass (though Quinn was a slightly better 5.5 per attempt).
The thing that those offenses shared (with some exceptions in 2002) was that they simply could not make teams pay for loading up with eight or nine men in the box to stop the run.
Consider, however, the two best games ND played in 2002: the victories over Michigan and FSU. Against Michigan, ND (Holiday) was 8 of 17 for 154 yards. It may not the sort of day that gets Heisman votes, but notice the nine yards per attempt. ND took and hit some deep shots and magically was able to run the ball with reasonable effectiveness (157 yards to Michigan's 91 yards) and the Irish had the win.
Against FSU, the Irish (Holiday again) was 13 of 21 for 185 yards. Again not headline-grabbing stuff, but notice again the 9 yards per attempt (including a deep shot to Battle for a T.D. on ND's first offensive play). And, again, Notre Dame was able to mount the slightly better rushing attack (116 yards to 92) and the Irish had the win.
Now, let's consider what went wrong even in games in which ND was competitive or won narrowly last year.
Against Washington State, the Irish threw 34 passes for 149 yards (4.4 per attempt), against MSU, 42 passes for 202 yards (4.9 per attempt), against Purdue 62 passes for 297 yards (4.8 per attempt).
The Irish also threw 17 passes for 33 yards against Pitt (1.9 per attempt), 27 passes for 139 yards against Navy (5.1 per attempt) and against Syracuse 34 times for 199 yards (5.9 per attempt).
Of the games that the Irish won narrowly or could've won (and I'm being generous in including Syracuse in this group), only in the BC game (39 attempts for 350 yards, 8.9 per attempt) did ND really have a decent per-attempt figure.
Many commentators place too much emphasis on completion percentage and simply assume that a quarterback who completes 60% of this passes is inherently better or more "efficient" than one who completes between 50% and 55% of his passes.
Let's consider Quinn's three games so far this year.BYU: 26 of 45 for 265 yards
Michigan: 10 of 20 for 178 yards
MSU: 11 of 24 for 215 yards
Now, most would look at the BYU game and think that the most impressive game so far for Quinn. The two statistics that jump out at most people are completion percentage (58% in that game) and total yardage. In his last two games he was at or under 50% and the total yardage was less.
But let's reconsider the games in terms of yards per attempt. In the BYU game Quinn was at 5.9 per attempt but in the last two games at 8.9 per attempt. And, not coincidentally, ND won those games and was able to run the ball.
While the Irish ground game was not devastating against either Michigan or MSU, at 3.4 and 3.7 per carry respectively it was good enough to carry the day, and a huge improvement over the under one yard per carry performance against BYU.
Note as well that the passing offense that chewed up large bits of the field against Michigan and MSU allowed Notre Dame to run the ball more. Against Michigan, the Irish ran the ball 40 times against 20 passes, and against MSU, it was 48 rushes against 24 passes. Against BYU, however, it was 47 passes against only 21 rushes. The result was defenses that were clearly softer in the second half of the Michigan and MSU games.
In practical terms, the possibility of completing some deep passes has meant that defenses can't stack the line and have safeties committed mostly to run support. Against BYU, the Irish had only two pass plays over 15 yards. One was the slip screen to McKnight that went for 54 yards and the other was a 27-yard pass to Fasano in the second half.
But in neither case was the ball caught more than 15 yards downfield; in fact, McKnight actually caught the screen behind the line of scrimmage.
Against Michigan, however, Quinn completed a 49-yard pass to Stovall (to set up what could've been a touchdown in the first half), a 46-yard touchdown pass to Shelton that cut Michigan's lead to 9-7 and a 27-yard pass to McKnight that set up the touchdown (on the pass to Powers-Neal) that made it 28-12.
Those plays helped to change the complexion of the game (just as the 41-yard completion to Stovall on the opening series against Michigan in 2002 and the touchdown bomb to Battle to open the FSU game in 2002 changed those games).
Suddenly Michigan had to defend the field and with a credible rushing attack emerging in the legs of Darius Walker, the offense started to work.
Against MSU, Shelton averaged 41 yards per catch and Samardzija averaged 16 per catch. Those are the sort of averages that will get a defense's attention and Notre Dame was able to mount the crucial drive to go ahead 28-7 without ever throwing a pass.,/P>
If Notre Dame continues to average 8.9 per attempt in the forthcoming contests, the Irish are likely to win a lot of football games.