The Watchtower: Lions at Bears

Last season, I started a series of posts I called The Watchtower; intense statistical analyses of each opponents’ coaching staff, as they’ve faced their Lions counterparts throughout history.  By comparing the coordinators’ head-to-head performances with their statistical means for the year, I hoped to isolate and expose systemic advantages for either side.  Historians argue about how predictive the process is—but in terms of breaking down the upcoming game, it’s interesting, and unique.  So.

Scott Linehan vs. Lovie Smith

Lin	Smit	Ornk	PgG	YpA	YpC	Drnk	PpG	YpA	YpC	PTS	YpA	INT	YpC	Sack
MIN	STL	6th	26.0	7.60	4.75	17th	20.5	7.60	4.75	17	6.88	1	7.27	8-54
MIN	CHI	6th	25.3	7.16	4.71	13th	20.7	6.49	4.13	27	11.61	0	4.04	4-10
MIN	CHI	6th	25.3	7.16	4.71	13th	20.7	6.49	4.13	14	8.45	3	6.64	5-34
STL	CHI	10th	22.9	6.69	4.26	3rd	15.9	5.36	3.96	27	6.47	1	4.59	3-24
DET	CHI	27th	15.9	5.30	3.95	21st	23.5	6.36	4.33	24	7.45	1	2.46	5-42
DET	CHI	27th	15.9	5.30	3.95	21st	23.5	6.36	4.33	23	7.70	1	4.00	2-13

The massive advantage this year’s edition has over last year’s is the inclusion of actual Schwartz-coached Lions data—we’re not reconstructing the theoretical scheme from what these coordinators have run in the past, these are the actual systems in place today.  Further, when the Lions face opponents they faced last season, the results will be incredibly relevant.  Unless, of course, said opponents do something crazy like replace both coordinators.
Enter Chicago, a team whose 2009 results fell well short of preseason expectations—and as often happens in such situations, the coaches bore the responsibility.  They replaced OC Ron Turner with Mike Martz, and demoted LB coach Bob Babich—the nominal defensive coordinator—back to a pure LB coach.  Still, Lovie did most of the defensive playcalling last season, and Marinelli is a disciple of the same Tampa Two that Lovie runs.  Nothing should be different, schematically, on the defensive side of the ball.
Last season, looking at the data for the first Bears Watchtower, I concluded:

Given greater, equal, or lesser talent, Lovie Smith's relatively aggressive Tampa 2 will surrender a disproportionate amount of yards to Linehan's balanced offense, but also generate high numbers of sacks and turnovers, disproportionately disrupting scoring.
The most likely outcome involves Stafford getting rattled by the Bears, getting sacked 3-to-5 times and surrendering at least two turnovers. Despite moving the ball as well as they have all season, the Lions should score below expectations (currently 19, though a 3-game average is nearly useless).

At the top of the second Bears Watchtower, I lauded my own foresight by pointing out the results:

• Stafford was sacked five times, for a loss of 42 yards.
• Stafford lost a fumble on one of those sacks, and threw an interception.
• The Lions generated a season-high 398 yards of total offense.

The symphony of accuracy ended with a clunk when it came to the points projection though; the Lions scored 24 points on the day, their highest total of offensive points in the the first ten weeks of the season.  However, that was only Week 4, and “normal” scoring expectations hadn’t been established.  After twelve weeks of Daunte Culpepper, Drew Stanton, and injured Matthew Stafford deflating the Lions’ season means, I tried again in the second Bears Watchtower:

If we apply that to the Lions’ current averages, and account for the Bears’ defense’s averages, my projection looks like this: 13-16 points, 6.00-6.25 YpA, and 4.50 YpC.  I have medium to high confidence in this prediction.

Of course, in one last glorious Screw You Ty Moment, Culpepper played the game of his Detroit life.  The Lions put up 23 points, as Culpepper’s stat line looked mysteriously NFL-caliber: 23-of-34 for 262 yards, 2 TDs, and only 1 INT.  The Lions did fumble twice, losing it once, but on the whole, the Lions clearly outstripped my “medium to high confidence” prediction.
My original conclusion, that the Tampa 2 suppresses Linehan offense’s scoring through sacks and turnovers, even while allowing far more yardage than said offense usually amasses, didn’t apply to last year’s games.  The sacks definitely occurred at an above-average clip, especially against Matthew Stafford in the earlier contest.  Turnovers happened, though not at an extraordinary clip.  The yardage came in bunches—but so did the points.
What was different last year?  The talent level.  In prior seasons, Lovie’s defense was ranked 17th, 13th, and 3rd in scoring allowed—but last season, the Bears’ defense was mediocre indeed, ranked 21st.  They allowed the same amount of points to the Lions that they’d been allowing to their average opponent. Since the Lions’ offense was obviously poor, either the Bears were going easy on the Lions, or there’s a systemic advantage.  Given the unlikeliness of the former, we’ll go with the latter and say . . .
Given greater or equal talent, Lovie Smith's relatively aggressive Tampa 2 will surrender a disproportionate amount of yards to Linehan's balanced offense, but also generate high numbers of sacks and turnovers, disproportionately disrupting scoring.  Given mediocre or poor talent, Lovie Smith’s Tampa 2 surrenders disproportionately high yardage and points, respective to the Linehan offense’s talent level.
Given the return of Brian Urlacher, and the addition of Julius Peppers, it’s safe to say that the talent and execution of the Marinelli/Smith defense will improve over last season’s iteration.  Assuming they improve to at least median level, I don’t anticipate this “unless they suck” qualification to be in effect.  So, back to lots of yards, sacks, turnovers, and probably-fewer points.  Now, what of the Lions?
Clearly, the Lions will be much improved over last season in both talent and execution; the additions of Jahvid Best, Nate Burleson, and Rob Sims have led to an obviously more potent attack.  Moreover, the offensive line as a unit looks much sharper, and Matthew Stafford’s decision making is much faster and much better.  I believe the offense will be above-average, ranked 8th-to-14th.  If we project the Lions to score to an average of 24 points per game, and the Bears’ defense to allow an average of 20 points per game, that leaves the Lions far short of what they’ll need to win.
Unfortunately, with zero real data as to the relative talent and execution levels, and a strong pattern suggesting that the Bears will bottle up scoring via sacks and turnovers if they’re notably better than last year, it leaves me high and dry.  But, from the data I have, the Lions should score between 20 and 24 points.  I have low confidence in this prediction.

Mitigating/Augmenting Circumstances

Given that the above section is almost entirely mitigating circumstances, this part seems redundant.  If the Bears are not improved over last season, and the Lions are as improved as we hope they are, the Lions will move the ball at will and score in bunches.  If either of those two conditions are false, this game could be much less exciting, and much more depressing, than we’re expecting.  Sigh.

Mike Martz vs. Gunther Cunningham

Martz

Gun

Ornk

PgG

YpA

YpC

Drnk

PpG

YpA

YpC

PTS

YpA

INT

YpC

Sack

STL

KCC

1th

33.8

8.91

4.81

19th

22.1

6.32

3.83

34

8.48

3

3.30

4-28

The only data point we have is a rough one: deep in the heart of the Greatest Show on Turf, Mike Martz’s offense was the best in the NFL—and one of the best ever.  Scoring 33.8 points per game, averaging an astonishing 8.91 yards per attempt, and Marshall Falk still rolling up 4.81 YpC, this offense was a juggernaut.  Gunther’s defense, meanwhile, was waning in his tenure as Chiefs head coach/DC.  Only the 19th-best scoring defense, allowing 22.1 points per game, they were nonetheless successful in depressing per-play effectiveness in passing and rushing, 6.32 YpA and 3.83 YpC.
When the two teams faced one another, it was a study in averages: Martz’s offense produced exactly to its season mean in points, and very nearly so in YpA.  Now, the Chiefs did manage to bottle up the run game, allowing only 3.3 YpC—and they also snared three interceptions, and garnered four sacks.  Then again, Martz’s offense was notorious for surrendering both picks and sacks in the name of scoring.  So, we only have one data point, and it points toward neither side having a systemic advantage or disadvantage.  The two teams should play to their (relatively unknown) talent and execution levels.
Given that the Lions allowed 30.9 points last season, and given that they’re presumably improved, I’ll presume that they’re presumably good enough to allow the Bears to reach their season averages for 2010—which, of course, is a total guess anyway, since the one thing Martz does everywhere he goes is inflate scoring over the prior year.  Let’s just call it thirty points.  This is a guess and not a prediction, and I have extremely low confidence in it.

Mitigating/Augmenting Influences

Yeesh.  It’d be shorter to list what the data tells us for sure than what it doesn’t.

Conclusion

There are limitations to what past performance can tell us about future performance, and this week I’ve frustratingly hit them.  With a strong trend from last year contradicting itself on Linehan vs. Smith, and no useful data from the other matchup, we’re left with a total crapshoot.  I’ll say that based on extremely weak data, the most likely outcome of the game is close Lions loss, with lots of sacks and turnovers for both sides, and a final score of 24-30.