Here are some other sample metrics that I have tracked over the past few years. This is not a thorough list by any means but it is a bunch of stuff that I can fairly easily calculate for comparison. Some of this can be compiled from the official USC stats either on their web page or the NCAA statistical summaries by team and category. The rest has to be calculated from personal data I collect above and beyond this level. Here are six basic charts that I compiled on the USC offense. For a change of pace I wanted to look at something other than just yardage and some of the other typical categories.
The first chart is called first down conversion percentage. This measures the percentage of time that a team gains plus four yards on first down. I have this data in game files so it was fairly easy to go back and calculate. I have never read the actual study but apparently this is one of the very few statistical items that correlates with winning in football. Turnover margin is another. The reason this metric is important is because gaining +4 yards on first down consistently puts the offense in a favorable position on second down (e.g. 2nd and five yards). There are a lot more options to call in the play book on 2nd and five than there are on 2nd and 10 for example. On second and five or even better on second and short the offense has almost a free play to use and they can be very creative. The defense has to defend the entire field and be ready for runs as well as pass plays. USC like a lot teams will sometimes line up in a run formation like a two tight end alignment on 2nd and short. Sometimes they will indeed run the ball but a lot of the time they instead run the play action fake as a counter tendency play and send a wide receiver deep especially if the QB spots a one on one match up. Either way the defense has to defend against the threat of both run and pass and can't "cheat" as much one way or the other.
As you can see the USC offense averaged over 50% on this metric for three straight years which is abnormally high. Of course this has a lot to do with the caliber of the players on those teams. The plus 4 conversion metric fell off in 2006 down to 43.3 % or slightly higher than it was in 2002 coincidentally the last time USC lost two games in a season. I suspect this number will improve again for some simple reasons. Number one a second year starting quarterback always makes better decisions, number two a healthy fullback will make a lot of incremental difference, and number three all the returning starters will make an impact as well. I would hope to see this number average over 50% again in 2007.
A more common metric tracked is third down conversion percentage. This third down conversion statistic is actually compiled and tracked by the NCAA. For recent years you can obtain the data from their website. For older years you have to go back to the USC website and sort through the historical files. Unfortunately that data only goes back a couple of years as well but it is better than nothing. Third down conversion only looks at if the first down was made or not and does not consider how many yards were gained. Success on third and one and third and eight are thus the same which can be misleading at times. Surprisingly to some people I have spoken with the 3rd down conversion rate was the second highest in the Pete Carroll era at USC in 2006. (Note: I did not go back and look at the 2001 data for any of these metrics. I imagine metrics would be overall lower that season). When I listened to Steve Sarkisian give a speech back in March of 2006 at a coaching clinic he joked that with a healthy fullback and a better run game they would have averaged over 70% on this metric in 2006. I'm sure it was a bit of exaggeration on his part but there is no doubt some truth to the matter as well. USC consistently had trouble picking up third down and short yardage in 2006 especially towards the end of the year. If this situation improves in 2006 then the offense will have an easier time moving the ball and keeping drives alive. Interestingly third down conversion was actually lower in 2003 and 2004 than in 2006. With all the vague complaints about "play calling" during the season I find this quite amusing. With all the injuries to the fullback spot and new players in 2006 I also find that achievement quite remarkable.
Here is a similar chart that just looks at 4th down conversion percentage. On third down there are anywhere from 150 to 200 plays in a year making it a much more frequent situation in football than 4th down plays. In general for the years below however there were only about 10 to 20 fourth down plays during the season making it a much smaller sample size. Oddly 2006 was the highest year on record for the USC offense in the Pete Carroll era. Mostly we as fans only recall the painful failed 4th down attempts so again this maybe a surprise to some (it was to me anyway).
Since the game was shorter in 2006 I don't like to compare total yards to total yards since this season is at a 10% disadvantage due to the effect of the clock rules that were implemented. Thankfully those appear to be gone for 2007 although there were some other changes made to speed up the game. That leaves yards per play as a useful metric and average yards per drive as alternative ways to compare things. On a yards per play basis the USC offense was about 10% less effective in 2006 than in 2003 or 2004 for example (5.9 yards per play in 2006 versus 6.5 and 6.3 for 2003 and 2004 respectively). That is mainly due to the capability of the individual players on the field in my opinion such as Reggie Bush and LenDale White for example and of course many others. I suspected the the average drive length (total yards for the year / number of total number of drives) would be worse in 2006 as well by about the same amount. Surprisingly however it was about the same and slightly better in 2006 than 2003 and 2004 by a slim amount.
I did the same comparison by taking all the scoring drives for USC and converting them into an average distance traveled. As you can see the number moves around a bit and offers some insights. In 2003 and 2004 when USC held or shared national titles the team on average had to travel a slightly shorter distance to score. I have traced these drives back individually and the main reason I could identify is because of defensive turnovers. The USC defense created more turnover by far in 2002 through 2005 than in 2006. In the extreme case the number of turnover forced was as high as 42 in 2003 and only 22 in 2006. That is 20 more drives to work with or about one and 3/4's of a game for comparison. Not only was the turnover total higher but disproportionately there were more on the opponents side of the field as well. This is the main reason I could isolate for why the average yards per drive scoring drive was shorter in 2003 for example. On the flip side the 2002 team with Carson Palmer had to move the ball further in order to score. That season was also the leader in some other data I checked on 80 yard sustained drives. That 2002 team might not have been as flashy as the later years but it is still one of my favorites for the simple way it wore out teams and averaged 88 plays per game on offense. In 2006 the offense only had 66 plays per game to work with for the aforementioned clock reasons as well as fewer defensive turnovers.
Comparing points is difficult not only due to the shorter game in 2006 but also of course because of player differences and other factors. Turnovers and field position affect scoring and yet are hard to measure in terms of effect. Defense, kicking, and special teams play also affect scoring as well. The 2003 season where USC averaged 41.1 points per game for example benefited almost a touchdown a game due to scores by special teams and the defense. In 2006 that amount was less than 2 points per game. Any year to year comparison is thus inherently difficult and of course an apples to oranges comparison at the aggregate level. What I thought of as a simpler check to perform was just to take the scoring drives by USC that started on their side of the field and see how many points that generated each year. If the offense has to move that ball that far it is a good proxy I assume for representing sustained offensive capability. Scoring when starting inside the opponents 40 yard line is not as difficult. When I did this calculation the following trend emerged. The total points per season of course varied from a low of 30.5 in 2006 to a high of 49 points in 2005 at the aggregate level. However at this other level I mentioned (scoring based upon starting on your own side of the field) the result is much more even it turns out. 2005 of course is still significantly higher than the other years due to the talent that assembled on the field that year. However the other years are within a much tighter band between 22.2 points in 2006 and 24.1 points in 2002. Turnovers, defense, special teams, etc. are a big impact on the game and much of that did not work in the favor of the 2006 squad.
Anyway I think this is all good news for USC fans. The offense did struggle at times in 2006 but in many ways it was surprisingly effective as noted by these metrics. Add in a second year starting quarterback, a healthy fullback, a more experienced set of running backs, and other personnel and the points per game should pick up nicely in 2007. I would be disappointed if it did not return to the 37 points per game level which is the average for the entire Pete Carroll era at USC. A big key however will be how much special teams and defense can contribute. If USC starts more often at mid-field or on the opponents side of the field due to increased turnovers or other reasons then watch out. We could see the return of a very prolific scoring attack.