Statistical Review of the 2008 San Francisco 49ers: II. Offense

On Monday, I reviewed the Niners' overall season stats for the team as a whole. Today, in Part 2 of my season review, I'm going to focus on the offensive side of the ball.

As I'm sure some of you have noticed, I was never a big Mike Martz fan, whether it was when he was the head coach for STL or during his tenure as 49er OC. Although I've articulated my Martz-related opinions in other Niners Nation articles and comments, the truth is that my dislike is based more on science than any 49er-related topic du jour. Basically, my problem has to do with how his offense sets players up for failure. It may produce gaudy passing yardage, but in the NFL, gaudy passing yardage isn't enough to win. And winning is the name of the game.

In the area of sport psychology, researchers have developed a unified theory of expertise that covers 5 types of skills: physical, technical, tactical, perceptual, and emotional. Physical expertise in football means having strength, speed, and agility that match the requirements of your position. Technical expertise in football means, for example, a wide receiver perfecting how he runs a dig route. Tactical expertise in football means outsmarting your opponent through gameplans, adjustments, double moves, trick plays, etc. Perceptual expertise in football means reading and reacting effectively. Finally, emotional expertise means effectively regulating stress reactions under pressure.

Martz is definitely great at the technical aspects of the game. Working on QB mechanics and WR route-running is his forte. Just ask Isaac Bruce. However, in my view, Martz has consistently failed (outside of STL ca. 1999) is in his ability to foster expert performance in his players, which, after all, is the main job descriptor of a coach. His primary failures in this regard - which were painfully evident this season - have been in the areas of physical, tactical, perceptual, and emotional expertise. Physically speaking, he installed an offense this season that didn't match the physical attributes of his players (e.g., deep routes for a slow receiving corps; dive plays for outside runners; über-frequent pass protection for big, strong, ham-fisted linemen; 7-step drops for a lumbering QB; etc.). Tactically, Martz seems to outsmart himself by trying to show everyone how much of an offensive genius he is, which leads him to call plays that are clearly wrong given the juncture in the game they were called and what the opponent's tendencies, strengths, and weaknesses are (e.g., passing too much while ahead in the 2^nd half, 7-step drops against sack-happy teams, passing against bad run defenses, etc.). The perceptual aspect of his failure is well-documented: His pass offense is based on automatic, perception-anemic QB decision-making. When a defense figures out the QB's decision-making process, it's especially hard for the now-robotic QB to make an adjustment mid-play or mid-drive (See NYG game). Because a good part of NFL Sundays are about adaptation and adjustment, I never like to see an offense that frowns on its QB calling audibles. Finally, as we saw in the ARI game, and as can be testified to by Rams fans, pressure is Martz's worst enemy. His mental mistakes and my-way-or-the-highway workplace demeanor seem to intensify when inadequate players, dissenting coaches, and crucial game situations put pressure on him.

So, in all of these ways, Martz has had a habit throughout his career - except for a year in which he had one of the greatest groups of offensive talent ever - of failing to put his players in positions where they can perform at the highest level possible. What's that? You don't believe me? Well, let me show you what non-expert performance looks like, and how Martz was a major cause. Ladies and gentlemen, I present to you the 2008 San Francisco 49ers' offense.

REVERSE TITHING

Below is a table showing the 49ers' offensive performance this past season:

Offensive Category	Statistic	Rank
DVOA	-14.6%	27
Weighted DVOA	-14.1%	28
DVOA Variance	4.6%	6
Pass DVOA	-18.5%	29
Rush DVOA	-9.7%	27

How's that for inefficiency? If the Niners were consistently bad overall, then their offense was consistently vomit-inducing. As I said in my Jimmy Raye breakdown, I wonder what Mike Martz must think about that 29^th-ranked passing offense getting freeze-branded onto his résumé. I also wonder what Frank Gore thinks about being the "centerpiece" RB in a 27^th-ranked rushing attack. The weighted DVOA, which I'll explore in further detail a little later, is also interesting because it suggests that the offense led by Shaun Hill was not all that much better than the offense led by J.T. O'Sullivan. But then again, I threw the word "suggests" in there for a reason.

After the jump, I'll compare the 2008 offense to the 2004-2007 offenses, break down the 49ers' offense DVOA, and evaluate their offensive line and skill position play; none of which paint a pretty picture for Mike Martz...

Below is a graph that tracks total offense, pass offense, and rush offense DVOA from 2004-2008:

In 2008, Martz's 49er offense was the 2^nd best since 2003 in terms of total offense DVOA. In addition, the pass DVOA was a 25% improvement over 2007. That's really where the good news ends for Martz though. Going back to my "I wonder what ______ must think..." theme, I wonder what Jim Hostler must think about the fact that his run offense was as good as Martz's, and actually better in terms of ranking (25^th). Also, with, in my opinion, marginally better offensive talent, Martz's pass offense wasn't as good as Norv's was in 2006.

So really, what contribution did Martz's offensive philosophy and play-calling bring to the 49ers? He didn't improve their run game, despite having a RB who led the NFC in rushing yards in 2006, and who he trumpeted during the preseason as being "the centerpiece of his offense." He improved their passing game from 2007 (albeit with superior talent), but didn't return it to the 2006 level despite having better talent than Norv did, and being known as a passing game genius. Looks to me like the stats say Martz didn't bring much of a contribution at all.

Before moving on, here are a couple of other observations about the numbers in this graph. First, as you can tell from the fact that 0% is the upper limit of the graph, the Niners haven't had an above average offense since the roster purge of 2003. Second, the 49ers' rush DVOA has been better than its pass DVOA each season since 2004. I forget. What happened in 2005? Hmmm...January and April seem to come to mind. Oh yeah! Mike Nolan was hired and Alex Smith was drafted.

HEY YOU, GET OFF OF MY CLOUD

So Martz's only success in 2008 seemed to have been improving the 49ers' passing offense by 25%. But should he even get the credit for that? Below is a graph tracking the 49ers' offense DVOA over the course of their 16 games:

As the trendline shows, the Niners' 2008 offense was a tale of 5 seasons. The first season, Games 1-3 was characterized by a fraudulent offense that was underperforming against horrible defenses. The second season, Games 4-7, was characterized by the Martz/JTO-induced Nolan death spiral, with the "Finish him!" command coming @NYG. The third season, Games 8-11, immediately followed the insertion of Shaun Hill as the starting QB. The fourth season, which I'll call, "Where Have you Gone, Frank Gore," was characterized by a bad weather game and Gore's injury (Games 12-15). Finally, the fifth season was really just one game, an all-things-came-together performance vs. WAS. So what does this tell us about Mike Martz? Well, it tells us that his choice of JTO as the starting QB, along with the methods he used to reach that decision, cost the 49ers 4 games. Once Singletary made the move to Hill, there was a definite upward offense trend for the rest of the season, except for a few games in which Gore was hurt. I guess on the bright side, though, at least Martz/JTO got Nolan fired.

More evidence of the Hill effect comes from the following table:

Season Half	Offense DVOA	Rank	Pass DVOA	Rank	Rush DVOA	Rank
Weeks 1-9	-19.2%	31	-35.8%	32	2.3%	11
Weeks 10-17	-10.4%	25	-2.6%	22	-20.1%	32

As this table shows, the offense was nearly twice as good with Hill at QB. More importantly, though, look at the improvement in pass DVOA, and the decline in rush DVOA, during the 2^nd half of the season. From those stats, we can draw a major conclusion that, in one respect, goes against the prevailing wisdom: The 49ers' offense improvement during the 2^nd half of the 2008 season was due far more to the promotion of Shaun Hill than it was to Singletary's insistence on a run-oriented offense. The pass offense was 30% better once Hill was promoted, whereas the run offense was over 20% worse - and last in the league - once Singletary was promoted. In other words, Singletary's reining in of Martz had more to do with passing than running.

Going back to my discussion of expertise, my main point here is that Shaun Hill was the one QB on the roster that could take advantage of the physical skills of the players around him. When Singletary got promoted, he inserted Hill and mandated that Martz start calling plays that better fit his offensive personnel. The improvement we saw in the offense was all about physical (i.e., personnel fit) and tactical (i.e., play-calling) expertise, both of which were forced on Mike Martz by a head coach that, unlike his predecessor, didn't defer to the self-proclaimed genius.

RED MEANS STOP! (REDUXE)

Below is a table showing the 2008 49ers' offense DVOA in each of 5 field zones:

Field Zone	Total DVOA	Rank
SF 1-19	-35.2%	30
SF 20-39	-21.8%	30
SF 40 - OPP 40	7.0%	17
OPP 39 - OPP 20	-1.0%	21
OPP 1-19	-38.0%	30
OPP 1-10	2.9%	19

I know I brought this up during the season, but Martz's offense was atrocious in the red zone, and I think it had to do with his play-calling. As I showed in Part 1 of my season review, the 49ers were a descent team for only 40 yards of the field (i.e., between their 40 and the opponent's 20). Now you see that the offense was primarily to blame for that discrepancy. How does this relate to Martz's play-calling? Well, in that 40-yard window, he's able to open up the playbook and call the intermediate/deep routes that are his passing game staple. When backed up on his own side of the field, these plays can result in costly sacks and interceptions. Similarly, with a lack of space in the red zone, there's simply not enough room to run those preferred routes.

Below is further proof:

Red Zone Play	DVOA	Rank
Pass	-59.3%	31
Rush	-17.9%	29

Hello? A -59.3% pass DVOA in the red zone? A ranking of 29 in red zone rush DVOA? For Martz's playbook and play-calling, that's no doubt an epic fail. To put some perspective on this, the #1 pass offense in the red zone this season (MIA) had a 65.4% DVOA, and the #1 run offense in the red zone (CAR) had a 54.6% DVOA. So that means SF was over 125% worse than MIA in red zone passing, and over 70% worse than CAR in red zone running.

WIN TO RUN, NOT RUN TO WIN

Below is a table showing the 2008 49ers' offense DVOA by quarter and half:

Period	Total DVOA	Rank
1st Quarter	-1.7%	18
2nd Quarter	-8.4%	29
3rd Quarter	-24.1%	29
4th Quarter/OT	-24.3%	29
1st Half	-5.3%	27
2nd Half	-24.2%	29
Late & Close	-22.7%	31

To me, it's not a coincidence that the 49er offense was much better in the 1^st half than in the 2^nd half. Why? Again, it's a matter of physical, tactical, and emotional expertise. In terms of physical and tactical expertise, either the running plays in Martz's playbook, or the running plays that he calls during the 2^nd half, are wildly inefficient. As Maiocco and Football Outsiders have pointed out, good teams run because they're winning, they don't win because they're running. When an offense can't run when their winning in the 2^nd half, it ends up with the DVOAs you see above. In terms of emotional expertise, Martz sure did seem a tad - shall we say - "verklempt" in that ARI game, no?

Here's some more proof of what I'm talking about:

Score Gap	Total DVOA	Rank
Losing > 7	-13.6%	26
Tie/Losing < 7	-12.7%	28
Winning < 7	-18.6%	31
Winning > 7	-17.2%	26

Now that the infamous "Win the West" banner (and its chief advocate) is no longer on display, perhaps Singletary should put up another alliterative banner to remind his players of Martz's dearly departed offense: "Worse When Winning."

HOME, SWEET HOME

Below is a table showing the 2008 49ers' offense DVOA by game location:

Venue	Total DVOA	Rank
Home	-5.9%	25
Road	-23.6%	32

That's right. The Niners had the worst road offense DVOA in the NFL this season. "Verklempt" anyone?

DIGGING YOUR OWN GRAVE

Below is a table showing the 2008 49ers' offense DVOA under different down and distance situations:

Distance	1st Down	Rank	2nd Down	Rank	3rd/4th Down	Rank
1-3 Yds	--	--	4.0%	17	-32.4%	31
4-6 Yds	--	--	-28.8%	29	3.1%	18
7+ Yds	--	--	-19.9%	29	3.5%	18
Total	-13.4%	28	-16.6%	29	-13.6%	23

As Maiocco so astutely pointed out, the 49er offense was tied for 26^th this season in terms of their percentage of 1st down running plays (43.0%). Is it any wonder then that they were (a) 28^th in 1^st down efficiency, (b) 29^th in 2^nd and mid/long efficiency, and (c) 31^st in 3^rd and short efficiency? Ineffectiveness on 1^st down passes leads to unfavorable 2^nd down situations, and too much passing on 3^rd down and short leads to Andy Lee (aka ROBO-PUNTER) stat-padding opportunities.

Here's a different illustration of the same phenomenon, showing offense DVOA by type of play:

Play	1st Down	Rank	2nd Down	Rank	3rd/4th Down	Rank
Pass	-4.3%	22	-32.5%	30	-23.5%	25
Rush	-23.8%	32	-0.1%	17	7.5%	19

Perhaps Martz's reason for calling so many 1^st down pass plays was because he knew his 1^st down run plays were the worst in the league? Or perhaps his 1^st down run plays were worst in the league because he called too many 1^st down pass plays? Sounds like the ol' chicken-and-egg to me. Either way, in Martz's offense, 1^st down passes put the team in a hole, 2^nd down passes dug the hole deeper, and 3^rd down passes covered the grave. Again, this is all about knowing your personnel and putting them in a position to perform at the highest level. If the 49er offense dug its own grave, Mike Martz was the one wielding the shovel.

THE PLAYER'S CLUB - OFFENSIVE LINE

Thus far, I think I've sufficiently proven that Mike Martz was - shall we say - expendable. Maybe I'm being too harsh though. I'm sure some of you reading this would make the argument that the players are at fault for the 49ers' offensive inefficiency. Perhaps the offensive line couldn't run block. Perhaps the RBs couldn't run. Perhaps the WRs couldn't get open or catch the ball. Perhaps...perhaps...perhaps. The rest of this article is devoted to these questions.

Here are the 49ers' offensive line stats for 2008:

Offensive Line	Statistic	Rank
ALY	4.30	8
YPC	3.96	25
Power Success	51.5%	32
10+ Yds	13.3%	30
Stuffed	21.9%	9
ASR	9.4%	31

As you can see from their ALY, the OL could, in fact, run block. In contrast, the OL's ASR tells us that they couldn't pass block. You have to ask yourself, then, "Why on earth was Martz calling passing plays so much?" Answer: A lack of physical and tactical expertise. Something else about these stats is mystifying, though. How can an OL be dead last in power success (i.e., 3^rd and 1 or 2 yards), but top 10 in stuffed percentage (i.e., not allowing negative-yardage run plays)? Both of these stats ostensibly measure strength and push off the ball. I'm still trying to figure this one out. If you have any ideas, feel free to offer them in the comments section.

One last thing I'll say about the stats in this table is that it's pretty clear that most of the 49ers' running game success (albeit infrequent) was due to OL play rather than RB play. Specifically, their ALY (i.e., an OL-focused stat) was substantially higher than their actual yards per carry (i.e., a RB-focused stat), and the percentage of running yardage gained after initial OL blocking (i.e., 10+ Yds) was nearly dead-last in the NFL.

In terms of directional running performance, below is a chart showing how well the OL blocked, and what percentage of running plays were called, in each direction:

Finally, something in favor of Martz. His most-called running play went in the same direction (C/G) where his OL was highly effective at blocking. Oh wait. Forget that. His least-called running play went in the same direction (LE) where his OL was best at blocking. So much for benefit of the doubt.

Looking at specific players, it seems as though RT is a clear limitation for the OL's run-blocking performance. Given that starting RT has turned over more times than JTO on the roster, this shouldn't come as much of a surprise. The Niners definitely need to improve their RT situation if they're going to become a better running team next season.

THE PLAYER'S CLUB - QBS, RBS, WRS, AND TES

So we've determined that, except for RT, the OL wasn't primarily to blame for the Niners' offensive inefficiency this season. Now let's look at the skill position players, starting with the QBs:

QB	Pass DYAR	Rk	Pass DVOA	Rk	Yards	EYds
Hill	156	27	-2.4%	28	1,912	1,864
O'Sullivan	-414	41	-37.3%	40	1,594	707

Obviously, this table reinforces what I said earlier about Martz vis-à-vis JTO. JTO was the 41^st-best QB in a 32-team league who was so inefficient that over half of his passing yards would disappear if he had achieved them in league-average situations. Martz was the OC who chose him to start over the clearly better Shaun Hill.

Here are the RBs (if there's no ranking associated with the stat, it means that you should interpret the stat with caution because of a lack of carries/intended passes):

Player	Rush DYAR	Rk	Rush DVOA	Rk	Yards	EYds	Success Rate	Rk
Gore	111	17	-1.8%	29	1,038	1,045	47.0%	20
Foster	-67	--	-31.9%		238	179	--	--

Player	Rec DYAR	Rk	Rec DVOA	Rk	Yards	EYds	Catch Rate
Gore	67	16	2.5%	28	373	413	65.0%
Robinson	86	--	54.9%	--	202	261	85.0%
Foster	37	--	13.0%	--	133	171	80.0%

As Niner fans, we definitely think Gore is a top 10 RB in the NFL. The stats above show that wasn't the case in 2008. Is it Martz's fault? My gut says yes. As Thomas Clayton fans, we think Foster is definitely expendable. The stats above show that was the case in 2008. Foster was a far better receiver than runner, yet M-Rob was an even better receiver than Foster. Basically, Foster isn't a good backup RB in the run game, and he isn't the best receiving RB in the pass game. Why is he still on the roster, again?

Here are the WRs and TEs (same no-ranking issue applies):

Player	Rec DYAR	Rk	Rec DVOA	Rk	Yards	EYds	Catch Rate
Bruce	154	24	5.9%	28	835	923	56.0%
Johnson	59	52	-3.7%	52	546	574	60.0%
Hill	72	--	9.4%	--	317	369	75.0%
Battle	71	--	11.0%	--	318	362	57.0%
Morgan	-6	--	-16.4%	--	319	265	47.0%

Player	Rec DYAR	Rk	Rec DVOA	Rk	Yards	EYds	Catch Rate
Davis	-55	37	-24.6%	37	358	143	63.0%
Walker	8	--	-1.1%	--	155	108	63.0%

A couple of interesting conclusions come from these stats. First, although Isaac Bruce was clearly their best, the 49ers didn't have any WRs in the top 20 of either DYAR or DVOA. Second, for all the love we give Josh Morgan, it sure seems like he has a ways to go before joining the elite WR ranks. I'm sure some of his poor performance had to do with being injured, and some of his catch rate deficiency has to do with the routes he runs, but there's definite room for improvement nevertheless. Finally, among the WRs, Bryant Johnson appears to be expendable. He was better than an average replacement (See DYAR), and top-64 (i.e., 2 WRs for 32 teams) in both DYAR and DVOA, but it seems as though Jason Hill would make a more-than-adequate replacement in the starting lineup. Indeed, Hill had the best hands on the team; which brings me to my next observation...

For all the crap VD takes about his stone hands, he actually had a better catch rate than every WR except for Hill. Now, degree of difficulty certainly has to be a factor here, but I'd say the grief he gets for dropped passes is disproportionate to the facts. My guess is that his drops come in really important situations, and so they're more magnified in our minds. Wait, why guess? Look at his DVOA. It's 37^th in a 32-team league. Basically, VD's performance in 2008 can be explained this way: He was good on easy plays, but he dropped balls and was all-around inefficient on difficult plays.

BOTTOM LINE

To recap, we can draw the following conclusions from the offensive stats I've presented here:

•The 49ers' offense was woefully inefficient, especially in the passing game.

•The primary reason for their inefficiency was Mike Martz. He chose the wrong QB to start the season, he installed the wrong offense given the personnel, he called the wrong plays given the personnel and its situational strengths/weakness, and he called the wrong plays with the lead and in pressure situations. In general, he simply didn't put his players in the best position to perform effectively.

•The promotion of Mike Singletary improved the offense because he promoted Shaun Hill and forced Martz to call shorter, more efficient pass plays. His promotion did not improve the offense by shifting focus towards the running game.

•The offensive line was pretty good at run blocking, but horrible at pass blocking. The main deficiency in run blocking was at right tackle.

•The choice of JTO over Shaun Hill is right up there with the choice of Jim Druckenmiller over Jake Plummer. It might have cost the 49ers a playoff berth, and definitely cost Mike Nolan his job.

•Frank Gore was not a top-flight RB under Martz, despite Martz's preseason vow to make him the centerpiece of the offense.

•DeShaun Foster is expendable because he's not a good backup runner, and he's not the best receiving RB.

•The 49ers need to improve at WR. Although Bruce performed well, he's not a #1 WR. The same is true, to a lesser extent, about #2 WR Bryant Johnson, especially given how well Jason Hill performed. Josh Morgan needs some more time before becoming a starting WR of the future.

•VD has better hands than you think he does. Truth is, he just happens to suck the most in the most important game situations.

OK. Done with the offense (finally). On Friday, the defense. TO BE CONTINUED...

**DVOA, ALY, ASR, DYAR, and EYds statistics used to produce this article were obtained from Football Outsiders.