We all know that statistics can be manipulated to say pretty much anything, and to suit virtually any hypothesis, no matter how outlandish.

 But if you assemble enough statistics, and order them logically, the truth tends to rise to the surface.  After crunching some numbers, and possibly inventing a new measure of defensive performance, the verdict on FSU's 2008 defense becomes clear.  It was the run defense, not the pass defense (which incidentially was tops in the ACC in terms of opponents 3rd down completion percentage), with the greatest room for improvement.   FSU's inability to put opponents into third down situations, statistically speaking, can be found after the jump....

As the basic premise for this analysis, it is worth mentioning that I have accepted the idea set forth by FSUncensored in his article Size Matters that FSU's defense would have been even worse if not for the Noles ability to hang onto the football.

In order to test this underlying premise, I dusted off my seldom used copy of Excel and calculated the average number of first downs surrendered  per minute of defensive play (using ACC teams and intra-conference play). I believe this measure is a better metric of defensive prowess for the following reason:  When the game is on the line, and the defense needs to get a stop or prevent a first down, they are out on the field without the benefit of a ball control offense.

In short, if FSU's overall defense was as weak suspected, and were helped by having to defend less plays overall (which  necessarily inflate sdefensive statistics uncontrolled for offensive time of possession), then measuring first downs surrendered per minute of defense should produce a more accurate measure of defensive power.

I don't know how to post fancy charts on here like FSUncensored, so you will just have to bear with me while I explain the numbers. 

FSU was 3rd from the bottom in terms of average number of first downs surrendered per minute of defense, yielding an average of .61 first downs/minute in ACC play.   Only Duke, at .62, and NC State, with a brutal .74, were worse.  The top team, unsurprisingly, was Virginia Tech, at just under .5/minute. 

So, FSU's defense, controlled for efficiency was atrocious, and FSUncensored's observation about time of possession skewing defensive performance appears to be spot on.

Having confirmed that FSU's 2008 defense, at least in terms of efficiency based upon yards surrendered over time, was poor, the great debate necessarily focuses on whether this performance can be attributed to an inability to defend the run e.g. Boston College Drive,  or an inability to cover the pass e.g. any pass to the tight end.

I have suggested, and would continue to suggest, that it was the run defense that was in the roughest shape, and that it was causing problems for our passing defense.  

Referring back again to Size Matters,and thinking about yards per play, I decided to break down FSU's run defense relative to other ACC defenses on 1st and 2nd down.  There are several reasons for excluding third down from this analysis. 

First, I haven't controlled my statistics to remove the effect of sacks (and I'm not sure that it is possible to do so by down), and I suspect that more sacks occur in obvious passing situations like third down. 

Second, third down running plays more so than 1st and 2nd down running plays  (excluding the red zone) tend to cluster around extreme short yardage situations, where a team needs 1/2 a yard or a yard and either barely gets the first down or doesn't.  In fact, FSU's third down run defense surrendered an average of just 1.19 yds/attempt.  In these types of situations, a team often gets or doesn't get around what they needed to pick up, thus making a defense's third down run performance unreliable. 

And third, there are far fewer third down rushing attempts. Per cfbstats, the rushing attempts per down against FSU were 224, 155, and 64, respectively, making third down rushing attempts a relatively small sample size.

In any event, when I compared FSU's 1st and 2nd down run defense to the rest of the ACC, the numbers were shocking.

In all games (I couldn't find just the conference games, so these stats include UF along with the two cupcakes to start the season), FSU surrendered an average of 4.45 yds/carry on first down in 2008.  That was good for second worst in the conference.  To put this number into perspective, the conference average was 3.9 yds/carry on first down.

On second down, the story gets much, much worse.   FSU gave up an eye popping 5.87 yds/carry.  This was dead last in the ACC.  The conference average was 4.09 yds/ carry. 

Without FSU's total lack of run defense on second down, the conference average would have been well below 4 yds per carry, and there would have been no meaningful difference between the average number of yards given up per carry on first or second down by ACC defenses.

To further quantify how bad FSU was at stopping the run in these situations, FSU, gave up an average of 1.42 yds/carry more on second down.  No other team in the ACC had a differential above -1, and no other team, assuming the opposition ran the ball on first and second down, surrendered a combined average of more than 10 yds for those two plays. 

So, what does all this mean?

Conclusions

1) FSU's lack of efficiency, measured in terms of first downs surrendered per minute of defense, appears to be correlated or explained by the lousy run defense on first and second downs.

2) Second down was disproportionately difficult for FSU to stop the run.   Not only was FSU worst in the conference by this metric, they were in a league of their own.   I suspect that 2nd and 5 (after FSU surrendered almost 5 yds on first down) was the most difficult and dangerous down for FSU to try and cheat up from the secondary to help stop the run, since that is a prime time for play action.  Without the benefit of additional run support from the secondary, FSU's front 7 were repeatedly and mercilessly shredded on second down running plays.

3) If I had 1st and 2nd down yds/rushing attempt were controlled for sacks, FSU would probably look even worse, although some of that effect might be offset by a disproportionate amount of sacks coming on third downs.

Edited to correct some spelling and grammar that was bothering me.