Bleacher Report - Articles by Arthur Troy

Behind The USC Trojan Scoring Decline: Part II

Arthur Troy — Mon, 28 Jul 2008 06:14:12 -0400

In a previous article I documented the scoring performance trend for the USC Trojans football team over the entire Pete Carroll era stretching from 2001 to 2007. The trend is impressive in multiple respects and deserves accolades. Many Trojan fans have fretted, however, as the offense has averaged a more pedestrian level of 30.46 and 32.62 points per game the past two seasons. A more complete game by game look at the scoring trend yields the following trend:

A polynomial fit through the trend has a nice upward spike towards the end of 2007 when all the players were finally healthy. There were a couple of noted stretches over two most recent seasons, however, when the Trojans failed to approach the overall average of 36.24 points per game for the Pete Carroll era.

Of course, multiple reasons exist when teams struggle to score points. In the previous article I outlined the rather large effect that a decline in turnovers forced by the defense has had on the scoring situation. Fewer turnovers equates to poorer field position and a greater number of long field drives. Without a turnover, USC receives the football via a punt instead of starting near mid-field where the Trojans started so many drives in the past thanks to their stellar defense or occasionally strong special teams play.

Through an intermediary I sent the gist of the turnover analysis to Head Coach Pete Carroll and he was kind enough to look at the document and relay back two comments. Number one: he agreed that the analysis was correct and it accurately reflected the large effect that turnovers and positive field position can have on scoring. Most teams , for example, have a 10% chance (or less) of scoring when starting backed up near their own end zone. Move the ball out to mid-field and the scoring probability jumps well over 50% for most any team.

Secondly, however, Coach Carroll commented that also "big plays" on offense are down for the USC Trojans compared to the past few years, excluding the 2001 season of course. This insight I suspect is much easier to digest and more obvious for fans that have watched the Trojan offense the past two seasons.

Carson Palmer and Matt Leinart were more exciting to watch at QB than the reliable and efficient John David Booty. Reggie Bush was a human highlight reel that recent running backs such as Chauncey Washington and others did not match. And the wide receivers in 2007 did nothing to make fans forget Mike Williams, Dwayne Jarrett, or Steve Smith. 2008 may of course be a whole different story.

Big plays are an important lens for viewing any offense. According to an unpublished NFL report "big plays" are one of four metrics that tend to correlate with winning. The four metrics mentioned are:

First down efficiency (% of 1st down plays that gain +4 yards)
Red zone efficiency (% of scores when inside the 20 yard line)
Explosive plays (+12 yard run plays & +16 yard pass plays)
Turnover margin (turnovers gained - turnovers lost)

Note: Unfortunately I have not been able to track down this report or verify its exact methodology. It is mentioned in passing in the book Developing An Offensive Game Plan.

So big plays apparently are not only exciting for fans to watch but necessary for winning as well, in some statistical sense. Big plays are tough to study, however, since they are not tracked or published online by any service I could locate. Just looking at averages for instance does not give you an indication of the number of big plays contained within that number. Here for example is the recent yards per play performance by USC.

In order to see the big play differential, the data is easier to view when it is converted into a histogram showing the frequency of how many times a certain amount of yardage has been gained.

Here, for example, is the amazing performance of the 2005 USC Trojan offense in total when viewed in this fashion instead of using averages.

As you can see, USC had an amazing 138 plays that season that gained +16 yards either on the ground or through the air. Over 100 other plays gained +10 yards or more as well. The spike centered around the zero value is mostly due to incomplete passes as well as a few zero rushing yardage plays.

For comparison, let's look at a more normal year such as the recently completed 2007 season using the same method. Here you can see clearly that the spike for "big plays" of +16 yards is down from 138 to 102 for a decline of 36 plays or roughly three big plays per game over the course of a season as Coach Carroll points out. In addition, there were a few more zero yard and negative yardage plays as well.

What does the undefeated 2004 national championship year for USC look line in terms of total offense?

Surprisingly the chart is not all that different in shape from the 2007 season. The number of total big plays greater than sixteen yards is essentially even between the two seasons. The number of zero yardage gains, however, is significantly lower in 2004 than in 2007. This ultimately all equates to the half yard per play advantage (6.3 yards versus 5.8 yards) the 2004 team had over the 2007 team.

The NFL study separated out big plays into +12 yard run plays and +16 yard pass plays. Displaying this slightly different breakdown in chart form would take forever so I summed up the 2002 to 2007 seasons into the following table for ease in viewing. (Note: I skipped the 2001 data since it contained nothing remarkable and it would be the lowest mark in each category).

Category / Season	2002	2003	2004	2005	2006	2007
Big Run Plays (+12 yards)	34	46	53	83	27	57
Big Pass Plays (+16 yards)	88	89	72	86	77	66
Big Pass Plays + Big Run Plays Combined	122	135	125	169	104	123
Just +16 Yard Plays Histogram (either run or pass*)	105	115	100	138	94	102

*Note the "Just +16 yard play" data will match the histogram charts above due to the way the run and pass data is combined to generate this number. The above histogram charts will not equal +12 runs added to +16 yard plays column since the two fields are combined into one for the histogram (i.e. the +12 to +15 runs are dropped from the total).

As you can see, each season has a different pattern and strength. None of course live up to the standard set by the 2005 Trojan offense; few teams ever will in my opinion. The number one year in terms of big runs for USC was the 2005 season by a wide margin. The 2007 season comes in second just edging the 2004 squad. The top year for big passes is a close race between the 2002, 2003, and 2005 Trojan offense.

It would take 18 individual charts to show each season since 2002 for big runs, big passes, and total offense. If you are that curious here is a link to a page that contains all charts by category and by year (click here). Conversely if averages are more your cup of tea then this file is of more use (click here).

Looking at the table above highlights that the 2006 Trojan offense was quite deficient in terms of big run plays generating only 27 plays of greater than twelve yards. Conversely the number of big pass plays was significantly down in 2007 to the lowest level since 2001. Of course the loss of starting QB John David Booty for three games no doubt affected this total. Still however it was trending for a slightly down year in the passing department even before his injury.

Pete Carroll and the USC Trojan coaching staff did an excellent job improving the running game in 2007. I suspect that the onus now moves mostly back to the passing side of the equation in 2008. A new more mobile starting QB in Mark Sanchez and a more experienced WR corps in 2008 may be poised for a break out year in that respect. Conversely a relatively new offensive line may drive some caution and a higher percentage of three step passes for the early part of the year until the offense settles in and finds a rhythm.

Keep in mind also however that opponent defensive coordinators are not sitting idly by twiddling their thumbs. They spend their off season studying opponent film and figuring out ways to stop attacks. Either in the form of different fire zone blitzes or slightly different secondary schemes, etc. they will have something new in store in 2008 as always. From casual observation and reading recent off season coaching clinic material it appears that more teams in the Pac-10 seem to be favoring a Cover 4 secondary scheme (aka quarters coverage) on defense than in the past.

Cover 4 Sample Alignment

This scheme is normally tough to throw deep against as it defends deeper vertical passing routes very well. Instead it gives up the flats area forcing a more lateral style passing attack. With down hill safety play it can also be quite effective versus the running game.

This scheme and other similar flavors such as the popular hybrid 1/4, 1/4, 1/2 style coverage schemes versus 3 WR Trips sets may be why YPC is down by 15% over the past half a dozen years in the Pac-10. Click here for a summary of Pac-10 offense trends to locate the YPC trend. Or perhaps it could be the influx of more defensively minded coaches into the league or other reasons such as player ability. That effect is tough to sort out and is good for debate. Of course that does not get the USC Trojans off the hook for the decline in big plays. The high standard of the past few seasons has created annually high expectations.

Regardless 2008 should be an interesting season to watch unfold overall in the Pac-10 and for USC in particular. I'll update the USC Trojan big play trends toward the end of the season to see how the results look in this regard. For USC to have a season that matches 2004 in terms of a national championship a sustained and balanced level of "big plays" in both the rushing and passing department will be required. Mark Sanchez, Stafon Johnson, Joe McKnight, Vidal Hazelton and other skill players get their chance to make big plays for the Trojans in 2008. I doubt however that a USC offense matching the 2005 squad will be been seen again for quite some time.

Note: The data for the histograms was calculated from unofficial play by play stats off of various websites. The charts are accurate to the extent that base data is correct which I can not confirm. Sampling a couple games versus the NCAA data base however did not show any significant deviation.

Cal Bears Power Run Game Concepts

Arthur Troy — Mon, 28 Jul 2008 05:12:19 -0400

If I were to ask you which team on average was the top rushing unit in the Pac-10 over the past six years which team would you guess? If you answered the USC Trojans or Oregon Ducks you'd be close but not correct. The top rushing team in the Pac-10 over the past six years is the University of California at Berkeley. For the past few years California has averaged an impressive 182.8 yards per game to edge out the Trojans and the Ducks. In 2004 and 2005 the Bears averaged an impressive 256.7 and 235.3 yards per game in back to back seasons. No other team in the Pac-10 generated back to back 200 yard averages during the same two year period.

PAC-10 Team Rushing Trend 2002-2007

Rank	Team	2002	2003	2004	2005	2006	2007	Team Avg.
1	California	108.0	168.3	256.7	235.3	162.8	165.5	182.8
2	USC	142.5	155.9	177.4	260.0	128.0	197.2	176.8
3	Oregon	145.6	131.9	158.7	134.3	181.9	251.7	167.4
4	UCLA	127.5	91.9	184.9	160.7	129.8	151.0	141.0
5	WSU	129.2	115.0	128.0	212.0	128.3	115.7	138.0
6	Washington	74.5	119.5	120.2	135.2	127.9	203.1	130.1
7	ASU	89.0	117.3	118.1	145.7	169.7	137.0	129.5
8	Oregon St.	148.7	134.9	70.7	122.6	118.2	174.9	128.3
9	Stanford	133.2	95.8	81.2	92.3	61.1	112.2	96.0
10	Arizona	43.8	123.8	118.4	122.0	84.2	76.7	94.8

Note: Jeff Tedford became the Cal Head Coach in 2002.

Normally when Cal Bears head coach Jeff Tedford's name is mentioned in the national press it is done in conjunction with the passing game and the impressive number of successful QB's he has developed at the college level. Trent Dilfer, Billy Volek, Akili Smith, Joey Harrington, Kyle Boller, and Aaron Rodgers are among his QB's that have gone onto the NFL as early draft picks. True the QB's have not produced much in the NFL as of yet but Coach Tedford and staff have certainly maximized their talents in the collegiate ranks.

Rarely however do I hear the name of Tedford or his staff mentioned in conjunction with their rushing attack in the national news media. Usually the comments center around the QB's and the Cal passing game or some vague comment about Tedford's noted "play calling" ability. Oddly the more credited Cal passing attack is only ranked 7th in the Pac-10 over the same stretch of time averaging around 237 yards per game while the rush attack has ranked consistently near the top position.

What makes the Cal Bears rush attack so successful? Tight Ends Coach Pete Alamar outlined several success factors in a presentation given some time ago to high school coaches. Here are just some of the reasons mentioned.

The Cal rush attack uses angle blocking and power run schemes highly effectively. The coaches believe this provides the best leverage and set of angles for linemen to attack the defense. There is no need to "out athlete" the defender. The blocking scheme makes it easier for technically sound yet not as athletic players to find success.

Also advantageous is the fact that the blocking rules carry over to multiple series of plays. The personnel, formation, and alignment may change however the blocking rules stay fairly constant. This allows for more repetition in practice which helps foster both technique and confidence.

In 2006 Cal went somewhat away from their power game and tried more spread formation sets with new offensive coordinator Mike Dunbar from Northwestern University. The experiment had mixed results and 2007 Cal went back to more 2 back style formations and 2 tight sets in 2007. Mike Dunbar moved onto a new coaching position with the University of Minnesota back in the Big 10 conference.

On the Cal power run plays overall the Bears reportedly averaged an impressive 9 yards per carry over a two season period from 2004-2005. The effectiveness of the play tailed off somewhat the past two years but I suspect it still probably averaged over 6 yards per carry. Stopping this type of run play is key for any opponent that faces the Cal Bears.

Here is the basic Two Back I Formation Power Play favored by Cal. The scheme is similar to the way many teams run it and the details depend upon the exact alignment of the defense and the ability of the Cal players. Here is a typical example.

On this play the right guard and tackle combination block on the defensive tackle. Together they drive him backwards towards the Middle or Mike line backer (M). If the middle line backer goes to the offense's right (i.e. over the top of the play) the tackle will peel off and block him leaving the guard on the tackle. If the MLB tries to come under the play then the right guard peels off to cover him leaving the offensive tackle on the defensive tackle. This double team and release component is very critical to the success of the play and cutting off support.

The center angle blocks down and back to his left on the defensive nose tackle position. The tight end on the right side of the line has a difficult block versus the defensive end. If the tight end can not block the defensive end Cal has some blocking options I'll outline below.

The basic concept behind any power run play is one player "kicks out" and one "leads up". In this basic example above the full kicks out on the strong safety who normally plays near the line of scrimmage versus this formation. The back side guard pulls and leads up aiming for the Strong or Sam Line Backer (S).

The power play is designed to normally attack the strong side A, B, or C gap. The running back has to read the blocks and figure the best open avenue. Normally this particular play does not bounce outside. It is a North South downhill running play.

What happens on the play when the 230 pound pass catching tight end can not block a stud 280 pound defensive end? Multiple options exist and here are two of the most common ways to handle this frequent dilemma. The first is called the "Rail" concept in the Cal running game.

In this version the tight end takes a veer style release and does not attempt to block the defensive end. Instead he targets the middle line backer and attempts to cut him off on the play. The right side guard and tackle work their combination block but this time have to adjust and work one man off onto the Weak side or Will line backer (W).

The so called rail adjustment now forces the fullback to take on the defensive end. If the fullback is a better blocker than the tight end as is the case on many teams this is a better (but still not easy) match up. The back side or left guard is still the lead through player and targets the Sam line backer (S). The back reads and follows his best course of action.

One other alternative for the play is the so called "Arc" version of the power play. This is one more way for the tight end to avoid blocking the defensive end.

On this version instead of veer releasing inside and aiming for the middle line backer (M) the tight end instead arc releases outside of the defensive end and aims for the strong safety (SS) which is normally a better match up for him. The fullback still has to kick out the defensive end on this play while the guard and tackle again work the combination block up to the middle line backer.

The back side guard again pulls and leads through the line up towards the Sam line backer (S). The running back follows as before and either cuts it up inside behind the combo block or follows the guard. In rare cases the play bounces outside.

Which version Cal (or other teams for that matter) use each week is a matter of personnel match ups, game planning and in game adjustments. There are more flavors of the power play than outlined here including a one back version. Also the run plays need to marry up with pass plays that use the same formation and motion. Coaches call this creating complimentary plays out of the formation. Otherwise the formation alignment gives away whether the play is a run or a pass via pre-snap read by the defense.

For those interested in the more details and coaching points of the play, here are several closer looks at Cal's power run game concepts. The first video with the bulk of the coaching material consists of Cal's 2 Back Power Runs and lasts about 40 minutes with explanation and film examples.

The second video contains the 1 Back Power Run examples (no fullback) which I did not explain above and has about 10-12 minutes worth of explanation and game film.

The final video is another 25-30 minutes with the variations which Cal calls the Rail and Arc variations of the play. Which version Cal uses depends upon the skill of their TE versus the opponents DE and the front defender's alignment (inside shade / outside shade, etc.). Take a close look at the execution and you can see one of the key reasons why Cal has been such a strong team in the Pac-10 the past few years on offense especially in the rushing game. For Cal to get back on track after last season's second half slide an effective running game will again be a key ingredient in 2008.

2 Back Power Run Play Examples (click for direct link if not available)

1 Back Power Run Play Examples

Rail & Arc Power Run Examples

Note: At times the video service does not cooperate and display the file right away.

Behind The USC Trojan's Scoring Decline - Part I

Arthur Troy — Thu, 17 Jul 2008 15:57:14 -0400

USC Trojan fans have fretted extensively the past two seasons as the team's offensive production has marginally declined. In hindsight this decline was somewhat inevitable after the 2005 season when the strongest offense in USC history largely departed for the NFL.

Still however the most recent Trojan squads did not match up to the level achieved by the 2002 to 2004 teams fielded by Head Coach Pete Carroll.

Whenever offensive points production falls off the normal fan and media tendency is to put all the blame on either the quarterback and/or the offensive coordinator. While this is indeed often the case there normally is more to the picture than initially meets the eye.

This article takes a closer look at USC football points production using five different categories across several years. The intent of the article is to analyze the points scored by the Trojans as well to identify the main sources of variation.

I did not fully include 2001 or 2002 in this study since the scoring drive charts were not all posted online by USC's Athletics Department for those games. For the few games I sampled however the same trend that I will explain below held true.

For starters lets review USC's total points production over the past seven years.

Year	2001	2002	2003	2004	2005	2006	2007	'01-'07 Average
Points Per Game	26.55	35.77	41.08	38.15	49.08	30.46	32.62	36.24

A more detailed look on a game-by-game basis is contained in this next graphic below as well. The following trend chart below includes the game-by-game average, a linear trend fit, and a sixth order polynomial trend fit as well for the entire period.

The lowest year in terms of points scored during the Pete Carroll era for USC is 2001 at 26.55 points per game in his first overall season at the helm. The highest level is of course the 2005 Trojan edition at 49.08 points per game when Matt Leinart, Reggie Bush, LenDale White, Steve Smith, Dwayne Jarrett, and others lined up on offense.

The average points per game scored for USC over the seven year span is an impressive 36.24 points per game. The past two years however have been on the lower side of this figure.

(Note: Keep in mind that the 2006 season in college football was effectively 10% shorter due to changes in the game clock rules that year. I will not adjust for that effect but any cross year comparison should probably bump those figures up by roughly that amount.)

A big problem however with looking at the scoring data at this aggregate level is that the total combines so many different factors it is hard to sort out with respect potential sources of the variation. For example in football:

The defense scores points at times on safeties, interceptions or fumbles recovered and returned for scores
Special team units can score points on punt and kick off returns for touchdowns or blocked kicks
Teams typically score on offense at a higher percentage after turnovers by the defense
Teams can score on short field (SF) scores (less that half the field) after blocked kicks, or turnovers, or excellent kick returns
Teams can score starting from normal medium field (MF) position starting around the 30 yard line
Teams can score on long field (LF) drives of 85 to 90 yards or more

Each of these categories is indicative of a different relative strength or weakness in a given year. The first four categories for example are reflective of how defense and special teams play directly and indirectly contribute to scoring by the offense.

The latter two cases are more reflective of the true offensive capability in my opinion since those involve sustaining drives the length of the field. It is worth looking at each of these closer in detail for greater insight into how a team actually scores. For simplicity I chose to break things into the five categories listed below.

Category 1 = Defense or Special Teams Scoring.

Category 2 = Scores on offense after a turnover (TO) caused by the defense.

Category 3 = Other scores (Non turnover related) or short field (SF) drives starting on the opponents 1 to 50 yard line.

Category 4 = Medium field (MF) drives for TD by the offense without the benefit of turnovers and starting between the USC 21 and 49 yard line.

Category 5 = Long field (LF) drives for TD by the offense without the benefit of a turnover and starting between the USC 1 to 20 yard line.

Here is the five year summary in terms of a) scoring drives and then b) points:

Category	Scoring Explanation	2003	2004	2005	2006	2007
1	Defense or Special Teams	11	3	4	3	0
2	Scores off TO's	18	22	22	12	18
3	Non TO SF drive (1-50)	12	9	10	4	9
4	Non TO MF drive (49-21)	26	37	28	32	28
5	Non TO LF drive (20-1)	22	9	34	15	15
	Sum Total	89	80	98	66	70

Converting the scoring data into more the familiar points look results in the following numbers:

Category	Scoring Explanation	2003	2004	2005	2006	2007
1	Defense or Special Teams	5.1	1.6	2.1	1.7	0.0
2	Scores off TO's	8.1	10.0	11.2	5.8	8.1
3	Non TO SF drive (1-50)	5.2	4.2	4.7	1.8	4.5
4	Non TO MF drive (49-21)	11.8	17.8	14.1	13.5	12.9
5	Non TO LF drive (20-1)	10.9	4.5	17.0	7.7	7.1
	Total	41.1	38.1	49.1	30.5	32.6

As you can see the sub-totals move around a bit and are pretty interesting once you start comparing the sections with each other across time. I will put them into bar chart format for easier viewing and explanation purposes. Let's start with Category 1—direct scoring by either the defense or special teams.

These initial numbers are not large but they are still significant. The 2003 edition of USC football for example was amazing in that the defense and special teams scored almost a dozen times in one season without the offense ever taking the field.

There were two safeties as well as five interceptions returned for TD (Will Poole, Lofa Tatupu, Jason Leach, Ronald Nunn, and Omar Nazel one each). In addition there were four fumbles returned for touchdown (two by Mike Patterson, and one each by Ronald Nunn, and Kenechi Udeze).

That amount alone accounts for 67 points of the USC total scoring offense that season or an impressive 5.1 points per game on average. Each squad afterwards has scored fewer points in this manner and the 2007 edition was the first team of the Pete Carroll era to be shut out on defense or special teams in terms of scoring.

Even in 2006 LB Brian Cushing scored once versus Notre Dame on an on-side kick off return and CB Terrell Thomas scored once versus Stanford on a blocked kick for example.

Of course direct scoring by the defense and special teams is one thing but contribution of good field position and positive momentum is another to be considered.

In the Rose Bowl game ABC put up the following picture in the game in the fourth quarter reflecting points production after turnovers in the 2008 Rose Bowl game.

Sudden "momentum shifts" after turnovers are forced by the defense clearly benefit the offense and subsequently affect how teams score for a couple of reasons. Of course there is the psychological advantage to the offense in these cases.

In addition, however, is the simple gain in terms of relative field position. The average spot over the past few years after the USC defense forces a turnover is roughly the opponent's 40-yard line.

After a turnover the offense starts with a "short field" advantage and only has to move about 10 yards to get into field goal range and 40 yards to score a touchdown.

USC is about 70% likely to score from such a spot the last time I checked. Combining the points in the first rows from above (direct scores by defense and special teams) along with points scored after a turnover the past five years you get the following result for categories one and two.

As you can see that is a pretty interesting insight. A very significant amount of USC scoring in any given season is a product of either direct scores by the defense or special teams, beneficial short field position, or the momentum created after turnovers by the defense.

This result becomes important when you also consider the trend the past several seasons for USC football in terms of net turnover margin (see below).

For this turnover trend data I'll go back to 2002 since the information is easily obtainable.

The blue columns indicate the turnovers forced on defense by either fumbles or interceptions. The middle bar is the number of turnovers committed by USC on offense. The third and lightest colored bar is the differential for the year (i.e. turn over margin).

Quite disturbingly turnovers forced by USC have fallen the past two years while turnovers by the offense spiked in 2007.

For USC the turnover differential has been extremely small now for two consecutive seasons. That may sound insignificant but remember that the average field position after a turnover is the opponents 40-yard line for USC.

From 2002 to 2005 that meant 38 to 42 drives per year that started *on average* inside the opponents half of the field. From there USC is highly likely to score roughly 70% of the time by my calculations.

When there is no turnover forced by the defense USC or any team instead gets the ball back via punt for example and loses 40 or so yards of field position and starts around or inside its 20 yard line.

Conversely when USC commits a turnover it in turn gives the opponent a shorter field to work with as well. That net effect is large and troubling when I look further at the data.

Of course even where there is not a turnover USC often gets a nice kick off return, punt return or the occasional shanked punt, etc. to help field position.

Those instances all can provide USC with short field position in a variety of subtle ways. I calculated those effects separately as Category 3 instances from the table above. Adding up rows one through three above now provides a complete total of all short scores (i.e. less than 50 yard drives), as well as points after turnovers, and points directly scored by the defense.

The trend now starts to become somewhat alarming when these three components are added together. As you can see below the USC offense the past two years clearly has not benefited from short field scoring opportunities as much as it has in previous years.

(Note: I sampled four games for 2002 and it was roughly the same picture.)

Next here is what is left over when you strip away the first three categories and see what the offense can do when it starts on its own side of the field and has to travel at least 50 yards to score and does not have the sudden momentum change advantage stemming from a turnover.

The following chart reflects Categories 4 and 5 put together and shows what the Trojan offense has done in terms of points production when starting on its own side of the field.

In my opinion this is the most accurate measurement solely of offensive points capability since it nets out most of the help often provided by special teams or defense. The result is surprising flat across four of the years and of course still spikes upward for the amazing 2005 USC offense.

The 2006 figure might be bumped up a small amount as I mentioned above due to the scoring clock rules for that season.

If you look at the breakdown in terms of points scored by USC as organized above the following conclusion pops out that I mentioned above.

The biggest problem the USC offense has had the past two years in terms of scoring has mainly been a decline in short field scoring opportunities which in reality are created as much or more by defense, special teams, and other events.

That message might seems somewhat perplexing at first but here are some additional points to keep in mind that tend to back it up overall.

For starters third down and fourth down conversion percentages have been improving overall the past seven years for USC on offense. The 2005 squad was of course the highest but the next two best seasons have been 2006 and 2007 by a slight margin. Fourth down conversion exhibits a somewhat similar trend.

Next here is a look at the average drive length by USC the past few seasons. This figure is calculated by taking all the yards for a given season and dividing it by the total number of drives for the year.

I did not sort the data and pick out the one play kneel downs etc. before halftime or the end of the game. So this amount probably understates the "real" number by a slight margin and may affect a couple years more than others but the effect should be small (i.e. a yard or two at most).

Average yards per play over the seven year provides another insight that I think is also very revealing in terms of the total picture. USC was indeed better on offense in the years of 2003 to 2004 by about a half yard per play and of course by a larger amount in 2005 than the past two seasons.

Those years correspond directly with the development of Reggie Bush, Matt Leinart, LenDale White, the wide receiving corps and one of the best offensive lines in recent USC history. The years of 2002, 2006, and 2007 in comparison have indeed been more "ordinary" compared to those years.

In conclusion let's return to the total points situation for the USC Trojans. If fans want to see an increase in points scored per game there are of course multiple ways for the Trojans to accomplish this goal.

For one the defense and special teams could come though again and score more some points directly. However I am not optimistic about this possibility given the past two years. 2003 in reality was probably quite unique in that regard for USC.
Alternatively the defense also can contribute in a major way by producing the same level of turnovers accomplished in years 2001 to 2005. That would dramatically help field position and increase the probability of any given scoring drive for USC on offense (e.g. more short field opportunities).
The more conventional angle of course is for the offense to become more explosive in terms of big play potential and increase the number of long field drives that result in scoring.

With skilled players such as Joe McKnight and Stafon Johnson at RB and many other talented receivers returning the third option indeed may be possible in 2008 for the Trojans.

However, the unknown variables at this point in time for USC on offense are of course 1) the QB position with new starter Mark Sanchez at the helm, two) the depth and talent of the 80% new offensive line, and of course three) the inevitable injuries and position shifts that will occur, etc.

With no up tick in turnovers and short field opportunities however I suspect USC will remain "stuck" around the level of 32 points per game on offense. That level is nothing to sneeze at but it feels slightly low compared to the past several seasons (a sign of how spoiled USC fans have already become?).

Keep in mind however that improving the offense is not the only thing required to win a national championship for USC or any other team for that matter.

The Florida Gators for example won the national title with a 13 Wins and one Loss mark in 2006 compiling 396 yards per game (19th in the country) and 29.7 points per game (23rd overall in scoring). That team finished strong however and had a suffocating defense at the end of the year.

In 2007 the Florida Gators improved both offensive marks to 457 yards per game to rank 14th in the country and averaged 42.4 points per game to rank third in scoring.

Despite these improvements however the Gators only finished the year with nine Wins and four Losses. Winning in football still consists of offense, defense, special teams, coaching, player talent, and other intangibles that can't always be quantified.

I'll check this trend again for USC to see how things differ in another 12 months at the end of 2008.

Other overall trends for USC and the Pac-10 are available here for those interested www.trojanfootballanalysis.co m

Coach Mark Richt on the Shallow Cross Series

Arthur Troy — Wed, 18 Jun 2008 07:37:17 -0400

University of Georgia Head Coach Mark Richt: The Shallow Cross Series

This past spring, the featured Saturday afternoon guest speaker at the annual USC Nike coaching clinic was University of Georgia Head Football Coach Mark Richt. Coach Richt flew in just for the day to make the presentation and then turned around to go home for what I imagine was an awfully long one-day trip. Compounding the travel distance and jet lag involved, Coach Richt was also visibly suffering from a slight head and sinus cold.

The crux of Richt's one hour presentation centered on the well-known Shallow Cross Series in the passing game. The play is nothing new and has been around football for quite some time.

However the play, which I'll describe in some detail below with images and video, was made famous through its repeated and highly successful use by the Florida State Seminoles in the 1990s towards the end of their spectacular run of 14 consecutive Top Five finishes.

According to Coach Richt, this single pass play was run by FSU during its heyday as often as a dozen times per game (albeit out of different formations). Bobby Bowden and Richt believed in running it again and again until the defense could stop it. Often the opponent simply could not.

This play was designed for superior execution honed by repetition. The shallow cross play was often half of FSU’s offensive passing yardage gained in any given game.

Coach Richt still uses the play today at the University of Georgia. QB Matt Stafford reportedly runs the play fairly well but is still developing in reading all the different routes and making the required throws with quick timing.

The shallow cross features five eligible receivers (not all WRs) and is run out of the shotgun with only a five-man offensive line protection scheme, meaning it has to have built in hot routes to handle any blitzes or the QB will get drilled. The routes are packaged in a manner such that the play will generally have at least one open route versus whatever defense the opposition plays.

However, the play requires both precise and quick reads by the quarterback. Despite possessing just medium arm strength, for example, former QB Charlie Ward ran this play to perfection and it was a big reason for his success in college football. FSU averaged 520 yards per game and Ward threw for just over 3,000 total yards passing with a 69-percent completion rate when he won the Heisman Trophy in 1993.

This first part of Coach Richt's presentation covered the tactical explanation of the Shallow Cross Series and how it works. This section is the detailed X’s and O’s breakdown on the alignments, different formations, adjustments, and key points of the play versus different defenses.

For those interested in that level of detail I'll link to the 40 minute video I made of the presentation. In the video Coach Richt explains the nuts and bolts of running the shallow cross package from four different 3 x 2 sets for offensive alignment.

For example there is a version called "1 Z" (see diagram below) that features the flanker on a shallow cross, a "1 Y" option that features the Y receiver on a shallow cross, a "1 Ted" adaptation which was Richt's designation for a tight end, and a "1 X" version where the split end runs the shallow cross.

Here is a sample diagram I made of the "1 Z" version with four wideouts and the QB operating out of the shotgun with a single back offset.

The front side of the play (right side) is a combination of three routes. The outside Z receiver runs the shallow cross, the Y receiver (Richt's nomenclature) runs the choice route depending upon the type of coverage (one high or two high safety, etc.) and the halfback runs either an arrow route or a scat route.

On the back side the "Ted" receiver (Richt's nomenclature) runs the hot route and adjusts in case of blitz. Otherwise he tries to split the defense depending upon the number and exact position of the safeties. The outside X receiver runs a bench route for clear out purposes and is not the main option on the play.

Details about pass protection, QB reads, adjustments versus different coverages, etc. are in the video (Click here for direct link to the 40 minute video).

The final part of the presentation highlighted various cut ups of the play, and this might be the easiest and most interesting part for college football fans to follow since it just involves video and Coach Richt's comments. The video shows most of the different reads and throws required for the "1 Ted" version of the shallow cross against different coverages and lasts about 10-12 minutes in length. (Click here for direct link).

I spoke briefly to Coach Richt later during the day when he was taking questions from attendees in another room before he left. I asked him a follow-up question about the plays' somewhat declining effectiveness compared to the Florida State years. Coach Richt replied that he used to think it was easy to make the play work.

However, in reality and with the benefit of 20/20 hindsight, it was special players like Charlie Ward and Warrick Dunn that made it so effective in the past.

Also, Coach Richt noted that he thought the best general defense versus the play is a Cover 2 scheme, which is now so common. This defensive secondary alignment makes the play essentially a Hi / Low or “Smash Route” type read to half the field versus the flat defender. An adjustment or "tag" to the play can also create the smash route concept on the back side as well with minor alterations to the play call.

Still, the Shallow Cross is a very effective play but apparently not quite what it used to be in the past. Interestingly Coach Richt noted that many opposing defensive coordinators learned not to blitz the play (he wishes they would however) as that strategy gave up many big gains and easy TDs in the past.

Notes: 1) If for any reason the video or images do not display try accessing them via this link to a web page I created with the same basic information.

2) Apologies in advance for the level of video and audio quality. Recording and editing this medium is something I am just learning.

Review of Terry Bowden's Analysis: What Stats Matter in College Football?

Arthur Troy — Mon, 16 Jun 2008 04:54:22 -0400

In April of 2008 former football coach turned analyst Terry Bowden wrote a piece for Yahoo Sports entitled "Coaching by the Numbers" that centered upon what metrics mattered the most for college football programs and how best to win.

The gist of the article centers around his chart that depicts how the Top 10 college football teams in 2007 finished in a variety of statistical categories. The matrix highlights to the right how many top programs finished in the Top 10, Top 25, etc. for each category (e.g. Rushing Offense, Passing Offense, etc.)

Bowden provides a more detailed chart as well for those interested (click here).

Per Bowden's analysis he concludes that Rushing Defense is the single most important statistic in explaining the success of a college football program. His rationale is that five of the overall Top 10 teams in 2007 finished in the Top 10 for Rushing Defense and all ten of the Top 10 finished in the Top 25.

Second most in terms of importance (based off of the Top 25 Teams) is Scoring Defense according to Bowden.

The chart is an impressive compilation of data organized in an effective visual manner. I applaud Terry Bowden for compiling the list and sharing it with public.

I even tend to agree with the general conclusion that metrics such as rush defense, scoring defense, and turnover performance "tend" to be a good predictor of success in college football. However those areas are no guarantee of success and of course Bowden does not make that claim.

An aspect of Bowden's analysis that concerned me however was the problem that it only included data for the Top 10 teams and not the entire set of Division I college football teams. Thus the sample size seems too small to draw any conclusions and at the very least opens up some further questions.

In order to be more robust, I thought the analysis should look at the middle and bottom teams as well as see if the opposite is true for the extreme case, etc. Out of curiosity I decided to see what would happen to the hypothesis of Mr. Bowden in two different dimensions.

Number one I wanted to see what would occur if the data set were expanded to include all 119 Division I teams in 2007. Number two I wanted to statistically correlate each of the categories above versus "wins" to see what type of quantitative relationship was exhibited.

In other words, instead of just noting that five of the Top 10 Teams were in the Top 10 for Rushing Defense I wanted to see how strongly that metric correlated with wins using simple regression analysis.

For an example of what I mean, here are the resulting scatter plots of rush offense, rush defense, pass offense, and pass defense versus wins for comparison. The other categories I'll summarize in a table below for more convenient viewing.

1. Regression Analysis Plot for Rush Offense (Yards per Game)

As you can see from the above plot there is a fair amount of variation in the data when rushing yards per game are compared to wins. Teams win with both low rushing totals and high rushing totals. Overall the fit is not particularly good and the r-squared comes out as 0.15 when calculated. (Note: 0 = no correlation, 1 = perfect correlation).

2. Regression Analysis Plot for Rush Defense (Yards per Game)

What happens when the same analysis is performed for rushing defense? The result is indeed a better fit than above for this set of data at least and an improved r-squared of 0.47 in this case. So rushing defense (although still far from a perfect predictor of wins) at least correlates better to wins than rushing offense.

3. Regression Analysis Plot for Pass Offense (Yards per Game)

Passing offense in terms of yards per game is likewise not a very good predictor of wins for the data set in question. Teams again win overall with low passing yardage or with high passing yardage. This wide spread results in a very low r-squared of 0.04 or almost zero correlation.

4. Regression Analysis Plot for Pass Defense (Yards per Game)

Passing defense is only slightly better in this regard than passing offense. There are many teams that won games with both strong and weak passing defense in terms of pass yards allowed per game.

This result is of course possible when you realize that if a team has a strong rush defense, teams may simply opt to pass more against that team. The result is a team that gives up lots of passing yards but may still be a fairly good defense overall.

Of course there are a dozen or so different charts that could be checked and posted (e.g. total offense, total defense, pass yards per attempt, pass yards per completion, pass efficiency offense, pass efficiency defense, etc.). I took the most commonly quoted and available categories from the NCAA database and summarized the contents in a table below.

2007 NCAA Division I Summary Table

NCAA Category	r-squared	Comments
Rushing Offense (Yards P/Game)	0.15	Low correlation to wins. Using rush yards per carry instead of yards per game only moves the r-squared to 0.17
Rushing Defense (Yards P/Game)	0.47	Medium correlation to wins.
Passing Offense (Yards Per Game)	0.04	Extremely low correlation to wins. Often teams behind have to pass to catch up. The result is a lot of passing yards but not necessarily wins.
Passing Defense (Yards Per Game)	0.09	Extremely low correlation to wins. Inverse of the above case. Teams ahead put in back ups and might surrender lots of passing yards yet still win the game.
Passing Offense (Yards Per Attempt)	0.21	Low correlation but more relevant that simply looking at passing yards.
Passing Offense (Yards Per Completion)	0.03	Extremely low correlation. Teams behind throw more passes especially near the end of the game which could account for this result.
Pass Efficiency Offense (Rating)	0.30	Low correlation but higher than other passing statistics.
Pass Efficiency Defense (Rating)	0.43	Medium correlation to wins.
Total Offense (Yards Per Game)	0.29	Low correlation to wins. Gaining yards does not necessarily produce wins.
Total Defense (Yards Per Game)	0.44	Medium correlation to wins.
Scoring Offense (Points)	0.51	Medium correlation to wins. Highest metric on the offensive side of the list. Points of course can be produced by offense, defense, and special teams and that effect is not accounted for here.
Scoring Defense (Points)	0.52	Medium correlation to wins. highest metric on the defensive side of the list. Just edges out scoring offense in the 2007 data set for highest r-squared honors.
Turn Over Margin (Plus / Minus)	0.28	Low correlation overall.

Surprisingly (or not so surprisingly depending upon your point of view I guess) nothing in this list strongly correlates (.8 or above) with winning games. Medium correlation is the best result obtained on a couple of metrics meaning that other factors not included here or in Bowden's chart strongly influence winning in football.

Even combining events such as rush yards, pass yards, and scoring offense together yields only a result around .51 for medium correlation. On the defensive side combining yards per rush, yards per pass attempt, and scoring defense only pushes the r-squared up to .56 as well.

Overall as Bowden's matrix indicated the defensive metrics tend to have more correlation with winning. However, these metrics are still medium in terms of correlation at best.

At the end of the day I suspect a couple of factors are at work here that would take a lot of work to untangle.

First is the problem concerning strength of schedule. The above data is not adjusted to reflect schedule difficulty. A team might rack up more yards versus weaker opponent while another equal team plays a tougher set of opponents. One might win more games (or lose more) accordingly and this could affect the r-squared values observed.

Second is the problem of "garbage time" in games. An excellent defense might look poor in pass defense for example if the starters are pulled and back up personnel play the entire fourth quarter. The team playing from behind may throw for a lot of big plays late in the game and narrow the gap. This could make pass defense appear less important in the end result than it might be if measured at the third quarter mark for example.

Third as coaches will tell you, these sorts of statistics are flawed in some respects when it comes to the actual way the game is evaluated and decisions are made in game. The above statistics reflect merely the end results of plays and ignore the specifics of the play call, down and distance, execution, and other factors.

For example, gaining three yards per rush might not sound impressive to fans. However if a team can average 3.0 yards 90 percent of the time on third and two against an eight man front, this keeps the chains moving and eventually helps a team score. That statistic can become more impressive when considering all relevant factors.

During the season and in the offseason coaches break the game data down much further than the simple tables above to analyze specific plays. Specific run plays (e.g. weak size zone run) out of specific formations (single back set or I formation set) versus specific defensive fronts (under, over, even, odd, and the location of the strong safety) are all charted and effectiveness is judged.

Logic, film observation, data tracking, analysis, and problem solving are all critical skills for football coaching staffs. Unfortunately the information made available to the public is too general to appreciate this aspect of the game. Terry Bowden's article is a step in the right direction though and I hope he explains more details in the future.