Divisional Preview:
The Steelers got a couple breaks in Cincinnati, one, literally, when Carson Palmer blew a knee on the same snap Chris Henry went down. The second, while the main object of desire in last week�s preview, was not so obvious; at least not to the average Bayless. Remember?
� �To win, the Steelers must limit the Bengal O. That means, get Rudi Johnson out of the game.�
Astonishingly, Johnson last touched the ball in the Bengals 5th possession. When Cincy came out for # 6, it was still a 1-score game, 21-17 Pittsburgh. There still were about 25 minutes to go; plenty of time except, evidently, to the Bengal braintrust. With Rudi on the sideline, Cincinnati never again threatened. Facing no run game at all, Coach LeBeau dropped 8 and 9 into coverage. Kitna threw under, or didn�t throw at all, and the Bengal aerial circus, 66.0 YPA after one throw, closed at ~6.1.
Brilliant. Of course, neither Palmer nor Rudi Johnson play defense. Pretty much that was true too of those Bengals paid to do so. The PS tallied 31, little different than their season mark v. Cincy, which rose from 29.0 to 29.66. Yes, Cincinnati stopped the run, to some extent, but as in their final seven (meaningful) games, when they allowed 30 points per, they paid the price. As noted last time, Cincinnati�s is not a playoff worthy defense. As noted last time, takeaways took it. The PS tallied two, officially. A bungled FG try (effectively) added another.
Meanwhile, presumably, the Colts gamboled beneath their practice dome. Indianapolis hasn�t played a consequential game in 4 weeks, not since Shawn Merriman and the Diego Bolts knocked them from the ranks of the unbeaten. It goes without saying that the PSD lacks a Merriman; it�s arguable whether they possess a Luis Castillo. Noteworthy to none, Merriman was my own #1 defender in the last draft and Castillo (who had an unassisted sack that time) my own #1 D-linemen. Of course, the PS had a shot at neither, but sticking to this (self-aggrandizing) theme, these excerpts, lightly updated and (mercifully) compressed, from our Game 11 assessment of the Indy-O.
Colts� O-side personnel:
Per Football Outsiders, the Colt O-line is the finest unit in the League. They are #1 in sack rate (allowed) at 3.4%; the League average is 6.7%. They are #1 in �line adjusted yards rushing�, at 4.71; the League average is 4.07. This is a slap at most draft prognosticators. While LT Tarik Glenn is a former R1 pick, RT Ryan Diem is a R4 guy and RG Jake Scott, R5. OC Jeff Saturday was not drafted at all; LG Ryan Lilja is a former UDFA too.
Indianapolis is strictly a zone blocking team; Saturday is a Hartings-size OC but the Colt OGs are small, 280# and 285#. In contrast, Indy fields twin jumbo high quality OT: Glenn, a finesse player, is 332# and Diem, a mauling type, is 320#. In general, the Colts run to their strength, which is on the edge. The Colt IOL tandem inside, with one scrapping to the second level. The playside OT typically hooks while the offside OT drops into the B-gap, sealing backside. The Colts� IOL do not get much movement, even in tandem; therefore, there is only a limited cutback threat associated with their stretch. Indianapolis is ranked high running between the tackles but most of that is off draw; like their stretch, a pass action run.
You know the principals: Peyton Manning, Edge James, Marvin Harrison and Reggie Wayne. Indy runs a single RB attack, almost exclusively. In years past, the Colts split equally between twin tight and 3-wide sets. This year, with Marcus Pollard in Detroit, they are dominantly 3-wide. Brandon Stokely is #3; he is a smallish WR with good speed. In the Warner-Ram mode, Stokely typically overmatches an opponents� 3rd CB. He and TE Dallas Clark are T-2 at 13.2YPC; Harrison leads the way at 14.0. Of all Colt receivers, Clark has the best YAC numbers; lost among the luminaries in Indy, his big play capacity is generally underrated.
Finally, the Indy O will test the Steelers� endurance. There are 3 components to that:
� Chasing the stretch laterally.
� Manning�s pre-snap antics typically keep D-linemen in stance. In last season�s Divisional round, the Pats countered that with a number of 5-6 LB sets, virtually eliminating down linemen.
� The no-huddle, like the Bills� old Red Gun (also constructed by current Indy GM Bill Polian) limits substitution options. This is going to be a problem for Hampton and, maybe, KVO.
What to expect:
When the Steelers met Indy in 2002, it was, pretty much, Manning to Marvin or nothing at all. Following a catastrophic knee injury, Edge James was slowly rounding back into form. Reggie Wayne had yet to develop and neither Brandon Stokely nor Dallas Clark was on the team. Then, the Colts had a mediocre run game and limited receiving weapons. Now, their run game is near to the top, statistically, and Manning has 4 quality receivers.
The Colts have built an offense to win either on the ground or through the air. Indy has shown a willingness to get into whatever kind of game an opponent�s defense dictates. Early this season, they ground it out against Jacksonville and Cleveland, content to win low-scoring affairs. Cincinnati chose an air war, and went down in flames, as did St. Louis. Week 12, the PS did the same (on the D-side anyway), playing packages from the Colts� 2nd possession. True, Edge rolled but package-heavy mode remains the Steelers best choice.
The Colts� stretch series places considerable pressure on the middle of the field. If the ILB are flowing, then the Colts will be throwing; if not, they�ll try to get Edge on the edge. The stretch is a play action run; Indy�s inside run game, predicated on the draw, is similarly dependent on their air threat. Most teams run to set up the pass but the Colts do not. Take away the air, and, even in packs, the PS may still win the battle on the ground. Not in the 2-4-5, their nickel of choice but, perhaps, in 3-3-5.
No doubt, the packages will give something against the run but here�s the thing. To limit the Colt O, any opponent must pick their poison, and then spit that out. That is: these PS must remove one aspect from Indy�s multi-faceted attack. By lore, that�s the run but in my opinion the Steelers� choice amounts to this: Peyton Manning can beat the PSD but Edge James cannot.
Keys v. Indianapolis O:
Safety deep: Excluding NE, those teams that had best success with Indy (Carolina and Jacksonville) are Cover 2 teams. In the base, the Steelers are not well configured to play that style. That is because their starting safeties are better playmakers than their starting ILB. The PS base D is better fit for single high, or Cover 3. In heavy dime, aka �quarter� pack, the Steelers can play Cover 2; otherwise, not so much.
Disrupt the Colt receivers: As is well known, NE played the Colt receivers physically; that is true of their CB, especially Ty Law in 2003, and their LB, especially in the 2004 Divisional round. The Panther CBs typically play a physical game too. The young Steeler CB, especially Ike Taylor, can overmatch most WR athletically; however, in zone, they do have recognition issues. It figures Taylor will be asked to shadow one Colt WR throughout. Although Marvin Harrison is the main man, Taylor is a better fit against Reggie Wayne. Harrison�s game is predicated on change of speed; Taylor has struggled against those types (reference: Derrick Mason). Wayne is a smaller, slower version of Chad Johnson. With Taylor on him, advantage Steelers.
Punish Edge James: James is it in the run game; in the pass game, he provides Manning a degree of security checking down. The PSD must deny Manning that easy option.
Set the edge: The Colts are a wide running team. The PS OLB can, and must, defeat Colt TE Dallas Clark and Bryan Fletcher.
Bull the Colt IOL: As noted, the Colt O-line works as a unit, moving laterally. Stunts figure to be ineffective, as, in the zone block scheme, O-linemen will block an alley, not a man. However, the PS D-line has a considerable power advantage in this segment. If the PS D-linemen can bow that line, pushing straight upfield, then the PS will win in the run game, whether in base or package sets.
The above outlines Coach LeBeau�s approach last time. That was fairly effective: excluding an 80-yard opening strike, which was all about Ike Taylor�s failure to execute Cover 3, and that short field TD opening H2, when Coach Bill�s long-shot onside didn�t pay out, the Steelers limited the high flying Equine Tots to 4 FG. Do that again and they should win it.
As for the PSO, well, conventional wisdom is: to beat the Dungy Two, beat down the Dungy Two. That may be so, but it does require running efficiently straight out of the gate, not something these 2005 Steelers have accomplished. Here, we tracked the Steelers� run/pass splits all season. Those results were summarized in last week�s Wildcard Preview:
- � The PSO closed in a rush, averaging 185 yards per over their final four games. In three, they gained 190 or better, averaging 204 in the last two. On the face of it, that�s an overmatch. The Bengal run-D is bad and the Steelers are running more effectively than at any point in this season. However, it is worth noting that most of those yards were gained late. In those final four, the PSO rushed for 740 yards, but just 210 in all H1.�
The Steelers run game over their final four was better than at any time in the season. However, even then that was not the component they won with, not from the jump anyway. Dr. Z of SI noted the same thing this week, as per his Divisional Preview:
- �Against Cincinnati, (�), Pittsburgh�s ground game was a modest 12 for 38 in the first half but 22 for 106 after intermission. It�s a pattern that�s been present all season. First half rushing totals have fallen short of those in the second half in each of their last 13 games. In fact, for the whole season, Pittsburgh�s rushing in the first half has averaged a meager 50.9 yards on 14.4 carries, a 3.5 yard average�.�
The good Doctor notes 50.9 per H1 overall, little less than the 52.5 per H1 the PSO gained at their best, over their final four. Of course, Z�s observation begs this question: are all high grade rushing attacks similarly split, struggling through H1 to dominate H2? Well, no:
Lead RB on leading rush attacks, H1/H2 splits:
Team rank |
Team |
Team YPG |
Feature RB or RBs |
H1 Carries/yards/YPC |
H2 Carries/yards/YPC |
YPC +/- By H |
#2 |
Broncos |
158.7 |
Anderson |
125/539 at 4.3 |
114/475 at 4.2 |
Insignificant |
|
|
|
Bell |
81/363 at 4.5 |
92/558 at 6.1 |
+1.6 H2 |
#3 |
Seahawks |
153.6 |
Alexander |
194/927at 4.8 |
169/929 at 5.5 |
+0.7 H2 |
#4 |
Chiefs |
148.9 |
Holmes |
63/237 at 3.8 |
56/214 at 3.8 |
None |
|
|
|
Johnson |
173/1,006 at5.8 |
163/744 at 4.6 |
+1.2 H1 |
#5 |
Steelers |
138.9 |
Parker |
119/462 at 3.9 |
130/738 at 5.7 |
+1.8 H2 |
|
|
|
Bettis |
61/173 at 2.8 |
49/195 at 4.0 |
+1.2 H2 |
#6 |
Giants |
138.1 |
Barber |
185/945 at 5.1 |
163/844 at 5.2 |
Insignificant |
#7 |
Skins |
136.4 |
Portis |
201/852 at 4.2 |
149/663 at 4.4 |
Insignificant |
Notes:
- Atlanta, #1 overall, is excluded due to the Michael Vick effect.
- Only two of nine RB listed have more carries in the 2nd half; they are: Willie Parker and Tatum Bell.
- Five of nine RB listed have significant YPC splits, H1 to H2. Bettis and Parker, along with Bell, have been notably more effective H2. Shaun Alexander, excellent in each, had 2 more H2 yards, but 25 fewer carries. Only Larry Johnson was more effective in H1.
- Still, six of nine averaged more than 4.0 YPC H1; Parker, Bettis and Priest Holmes did not. Of eighteen �halves� noted above, these RB achieved 4.0 or better in fourteen. Holmes missed twice; Bettis and Parker as noted.
Parker�s H1/H2 split is the largest; Tatum Bell�s is next. Bettis is the only RB to average less than 3.0 YPC in any given half. The sum suggests that these Steelers are about one Mike Anderson short of dominance. Next year, Duce Staley may be that guy. Or not. Regardless, it is highly unlikely that these 2005 PS can pound it effectively early. It was what it was; with the exception of the opening two tilts, they haven�t done so yet.
Maybe, the Steelers� early struggles are all about the absence of a big feature back. If so, well, that�s not going to change. More likely, or more optimistically, it�s a matter of the PS O-line (Faneca, Hartings and the three pachyderms) wearing on opponents. In that case, the PS could go all run early (ala AFCC 2004), seeking to gain their attrition threshold by, say, Q2, rather than Q3. However, that�s a bad strategy, overall, considering the Indy O tendency towards fast starts:
- In 14 games that mattered, the Colt O scored 46 TD in 114 possessions. That is a rate of 32%.
- In 33 Q1 possessions, the Colt O scored 14 TD. That is a rate of 42%.
To win, the PS will have to take the crowd out of the game. To do so, they�ll have to score early. There is no reason to expect they will run the ball effectively, not in the early going. Therefore, they�ll have to throw to move it. In other words, SOP, per our yearlong rush charts.
Finally, this key indicator:
- Roethlisberger led the League in YPA. His 8.90 was 0.63 ahead of #2, Peyton Manning. That�s an outsize margin as 0.63 behind Manning clustered 5 QB.
- November in Indy, following a 3-week layoff, Roethlisberger�s YPA was 5.12.
There�s the difference-maker. The Steelers do have one other thing going: punt returns, where they finished #3 overall. In contrast, the Colts� little used punt coverage unit was #30, allowing 10.9 YPR. If the PSO can�t score early, the PS PR group might. However, this, adapted from that concluding our Game 11 preview, still seems to apply:
Conclusion:
In an alternative universe, this Divisional encounter is the Game of the Week. There, the 13-3 Steelers invade Hoosier Land, lair of the greatest of clan Manning, the man around whom the League Office reconfigured the game. In that world, commentators must note that Peyton is the Last Big Thing; that, with Roethlisberger at the helm, the Steelers, not the Colts, have the most potent air attack in all of football. Not the most prolific, but the most potent: YPA being the prime indicator. Roethlisberger leads in that stat; Manning does trail.
We are where we are. SOP v. the Dungy 2 is to clock it, grind time, and win late. That worked out fine against the Dungy Bucs, circa 2001, and those remnants under Gruden too, circa 2002, but there is this crucial difference: these Colts can score but those Bucs could not. Championship teams can win all kinds of games. Consider the Pats: in one Super season, they beat the Colts 38-34; in the next, they beat them 20-3. These 2005 Colts, seeking to dethrone their nemesis, have won big, and they�ve won ugly too. As for these Pittsburgh Steelers: well, they�ve run effectively against some opponents, but more often, not. To be the best, this season, they�re going to have to win by other means. To be the best, within the Manning Rules, they�re going to have to develop the fortitude to get into a shoot-out, as necessary.
Buckle up.
Back of the Book:
As noted above, the PSO gained 36.6% of their rushing yards H1. Last year, well:
2004 PS run game stats, QB rushes are excluded.
Opponent |
H1 Carries for yards |
H2 Carries for yards |
Total Carries for Yards |
Oakland |
19 for 60 (3.15) |
11 for 34 |
30 for 94 (3.13) |
@ Baltimore |
9 for 46 |
15 for 46 |
24 for 92 (3.83) |
@ Miami |
19 for 56 (2.94) |
18 for 95 |
37 for 151 |
Cincinnati |
17 for 73 |
19 for 90 |
36 for 163 |
Cleveland |
15 for 68 |
22 for 89 |
37 for 157 |
@Dallas |
12 for 56 |
15 for 61 |
27 for 117 |
NE |
19 for 81 |
25 for 137 |
44 for 218 |
Philadelphia |
23 for 124 |
27 for 118 |
50 for 242 |
@ Cleveland |
19 for 70 (3.68) |
21 for 72 |
40 for 142 (3.55) |
@ Cincinnati |
14 for 63 |
17 for 72 |
31 for 135 |
Washington |
16 for 33 (2.06) |
16 for 70 |
32 for 103 (3.21) |
@Jacksonville |
13 for 42 (3.23) |
9 for 38 |
22 for 80 (3.63) |
Jersey Jets |
11 for 40 (3.63) |
16 for 82 |
27 for 122 |
@ NJG |
21 for 82 (3.90) |
17 for 79 |
38 for 161 |
Baltimore |
11 for 64 |
25 for 109 |
36 for 173 |
@Buffalo |
20 for 55 (2.75) |
18 for 101 |
38 for 156 |
Jersey Jets (PO) |
12 for 48 |
27 for 115 (includes OT) |
39 for 163 |
NE (PO) |
20 for 51 (2.50) |
12 for 67 |
32 for 118 (3.68) |
Notes:
� Opponents indicate games Roethlisberger started.
� Boldface across indicates 2004 games when the Steelers ran for less than 4.0YPC H1 and for <~37% of all rush yards, as per their 2005 standard. In all, 4 of 18; of those 4, Miami and Buffalo followed a pattern similar to that seen lately (bad run totals early, but strong at the close).
� H1 YPC totals < 4.0 are (shown), as are game totals. In eight of eighteen, the 2004 PS ran for less than 4.0 in H1; in five, less than their 2005 H1 standard of 3.5.
� Adapting Z�s terms: �H1 run totals fell short of H2� in 13 of 18 games. In 3 of those anomalous 5, PS RB/WR totaled less than 100 yards rushing.
2005 PS run game stats, QB rushes are excluded.
Opponent |
H1 Carries for yards |
H2 Carries for yards |
Total Carries for Yards |
Tennessee |
15 for 72 |
23 for 129 |
38 for 201 |
@ Houston |
15 for 73 |
17 for 62 |
32 for 135 |
NE |
10 for 36 |
10 for 26 |
20 for 62 |
@ SD |
17 for 52 |
13 for 37 |
30 for 89 |
Jacksonville |
11 for 24 |
16 for 34 |
27 for 58 |
@ Cincinnati |
14 for 72 |
29 for 154 |
43 for 226 |
Baltimore |
11 for 41 |
13 for 56 |
24 for 97 |
@ GB |
9 for 72 |
16 for 68 |
25 for 140 |
Cleveland |
13 for 45 |
24 for 98 |
37 for 143 |
@ Baltimore |
6 for 9 |
15 for 50 |
21 for 59 |
@ Indy |
14 for 22 |
8 for 43 |
22 for 65 |
Cincinnati |
16 for 44 |
11 for 49 |
27 for 93 |
Chicago |
14 for 64 |
30 for 128 |
44 for 192 |
@ Minnesota |
10 for 20 |
23 for 115 |
33 for 135 |
@Cleveland |
21 for 66 |
14 for 143 |
35 for 209 |
Detroit |
13 for 52 |
28 for 142 |
41 for 194 |
@ Cincinnati (PO) |
11 for 32 |
19 for 109 |
30 for 141 |
Notes:
� Adapting Z�s terms: �H1 run totals fell short of H2� in 13 of 17 games. In 3 of those anomalous 4, PS RB totaled less than 100 yards rushing. The other, Houston, featured their H1 high, 73 yards.
� PS RB/WR have rushed for 100 or better in just 10 of 17 tilts; the last 5, the first 2 and few in between. Last year, that number was 15 of 18.
� Trai Essex started, or saw consequential action, in four games. PS RB/WR broke through to triple digits in one.