Looking back at FEI's preseason projections, we find that most teams did about what they were supposed to do -- but not in the Big Ten, where things got screwy.
19 Jan 2013
by Danny Tuccitto
In the lead up to this game, much has been made of the fact that, at +4.5 points (or +4 in some places), the Falcons are the largest underdog of any No. 1 seed in any round since the 1978 playoff expansion. DVOA doesn't go back to 1978 yet, but we can tell you a couple of things about how this mismatch ranks over the past 22 seasons.
Limiting things to conference championship games, Atlanta's deficit of 20.8 DVOA percentage points as a No. 1 seed this week bests (or worsts) them all, nearly doubling the previous high (or low). On their way to a third consecutive Super Bowl loss, the 1993 Buffalo Bills hosted the AFC Championship game against a Kansas City Chiefs team that was 11.2 percentage points better than them during the regular season. In the game, Buffalo took a 20-6 halftime lead, but Kansas City had one of the few quarterbacks on earth at the time that would prevent a fan base from spending the break hovering over a porcelain goddess. Unfortunately, Joe Montana was the meat of a buffalo sandwich early in the third quarter, and left for good with a concussion.
Being that big of a DVOA underdog as a No. 1 seed is bad enough, but it turns out that this year's Falcons are the second biggest underdog of any championship game host since 1991, regardless of seed. You don't have to explore the deep recesses of your memory to come up with the biggest underdog, unless you're an Eagles fan and blocked out the 2008 NFC Championship game in Glendale. In that one, the Cardinals won despite a 36.9 percentage-point DVOA deficit. I'd say that's a record that may never be broken, except the three biggest deficits faced by a championship game host have all happened in the past five seasons. (The 2010 Bears lost to Green Bay in Chicago as a 20.6 percentage-point DVOA underdog.)
The good news for Falcons fans is that they may have noticed two of the three teams I've mentioned thus far advanced to the Super Bowl despite being a championship game underdog. Indeed, although (perhaps especially because) small sample size caveats apply, this hasn't been much of a predictor since 1991. Overall, the No. 1 seed has faced a DVOA deficit only seven times since 1991, they've gone 4-3 in those matchups, and two of the four winners took home the Lombardi Trophy (the 1994 49ers and 2003 Patriots). Meanwhile, championship game hosts (regardless of seed) have gone 7-7 as a DVOA underdog, and four of the seven winners prevailed in the Super Bowl (1994 49ers, 2003 Patriots, 2006 Colts, 2008 Steelers).
For those who may be unfamiliar with the Football Outsiders stats, they are explained at the bottom of the page. Scroll down or click this link. Please remember that all stats represent regular season only, except for weighted DVOA and anything else specifically noted.
Any game charting data that appears with a asterisk appears courtesy of the ESPN Stats & Information Group and is complete through the end of the season. All FO game charting data is now complete for the four remaining playoff teams.
All readers can click here for in-game discussion on our message boards. If you have FO Premium, you can click here to see all the matchup of DVOA splits for this game. Don't forget to also check out our scouting preview in Andy Benoit's Film Room.
Another major story line of the week was the idea that Atlanta's defense has had trouble -- to put it mildly -- stopping offenses that feature read option. And it's true: Two of the Falcons' four worst single-game DVOAs on defense came against Carolina in Week 14 (29.0%) and against Seattle last week (24.0%). The game that people cite as evidence against the conventional wisdom was when Atlanta shut down Robert Griffin in Week 5. However, looking back at that one, Washington only ran the ball from a read option look twice before Griffin left the game due to injury, and nearly all of Alfred Morris' 115 rushing yards came out of the typical zone-blocking running plays that Mike Shanahan has been calling for decades. Griffin ended up only running the ball once on a scramble out of naked bootleg action.
The 49ers actually don't run as much read option as you might think. They kept going to it against Green Bay because, as Ben Muth detailed on thursday, the Packers defense played as if they had never seen read option in their lives. The Falcons may have been thrashed by it a few times, but they have seen it plenty. If they show a propensity for stopping it on Sunday, San Francisco will likely try something else.
One indicator of this "We're not a read option team, per se" idea comes from our shotgun stats, since that's where read option comes from. Since Colin Kaepernick became the 49ers starting quarterback, he's been in shotgun (or pistol) on 44 percent of snaps, which ranks only 22nd in the league. Granted, that's nearly twice as often as when Alex Smith was the starter (25 percent), but Kaepernick takes a surprisingly large percentage of his snaps from under center.
So, if Atlanta can just find a way to slow down the read option, what does the rest of the run game matchup look like? Surprisingly, it doesn't look that bad. Yes, the Falcons rank 20th in run defense DVOA, and the 49ers rank third in run offense. However, according to our line metrics, Atlanta's front seven ranks 10th in ALY (San Francisco's offensive line ranks first) and 11th in stuffed rate (San Francisco ranks seventh). The Falcons front seven also excel on runs to the offensive left (seventh in left tackle ALY, eighth in left end ALY), two directions where San Francisco ranks near the top of the league in run frequency. Furthermore, as Rivers McCown pointed out in last week's divisional round preview, the Falcons play much better on run defense at home than they do on the road, and have been a far more efficient unit with Sean Weatherspoon on the field and Ray Edwards off the team.
Given how surprisingly neutral the run game matchup is, the 49ers may find themselves in the exact situation Jim Harbaugh and Co. elevated Kaepernick to deal with: having to win a game via the pass. Luckily, there are several stats that favor San Francisco here. First, in yesterday's Film Room, Andy Benoit mentioned two tactics San Francisco can use to mess with Atlanta's zone pass defense: (1) lining up Michael Crabtree and Randy Moss on the same side of the formation, and (2) using 3x1 sets.
Our charting stats agree. Against personnel groups with two wide receivers and no other skill players flexed or split wide, the Falcons allowed 5.0 yards per pass and had a 64 percent defensive success rate when the receivers were in a "balanced" formation (i.e., on opposite sides of the field), but allowed 6.3 yards per pass and had a 54 percent success rate versus a "twin" formation (i.e., on the same side). Even more pronounced are the Falcons' stats against "trips" formations vis-à-vis "slot" formations. Namely, Atlanta has only faced eight passes all season against trips formations, while San Francisco is 0.8 yards per pass better, and are three percent more successful, in trips than slot.
Focusing on these 3x1 sets, Andy also noted that the main Seahawks beneficiary was tight end Zach Miller: On three targets in the middle of the field, Miller had three catches for 51 yards, with all three resulting in successful plays. Overall, Atlanta ranked 21st in pass defense DVOA versus tight ends this season (3.0%). However, there's a huge disparity when we split things out by pass direction. On passes to tight ends in the middle of the field, the Falcons defense had a 45.8% DVOA (ranked 28th), which was worse than on tight end targets to the left of the offense (-2.0%, 17th), and much worse than those to the right (-33.5%, fifth). San Francisco seems to put Vernon Davis in hibernation, and then feature him at the perfect time; Sunday may be one of those times.
So if the 49ers passing game is going to be a vital source of points on Sunday, then I'd be remiss if I didn't talk about pressuring Colin Kaepernick. The big-picture stats say that San Francisco's offensive line ranks 29th in Adjusted Sack Rate (8.5 percent), while Atlanta's front seven ranks 26th (5.8 percent). And certainly, John Abraham's ankle issue doesn't help the situation for Atlanta. However, digging deeper shines a light on a few interesting splits.
Although the Falcons have a hard time sacking opposing quarterbacks, Atlanta does hurry them an awful lot. Against Seattle, they only sacked Russell Wilson twice, but hurried him nine times; that continued a season-long trend. This year, despite their lowly ASR ranking, Atlanta had the third-highest hurry rate in the league (18.3 percent).
Similarly, the San Francisco's ASR is misleading. When Alex Smith was the starting quarterback, the 49ers offensive line had an ASR of 9.8 percent; it was 6.8 percent during the regular season with Kaepernick. Their hurry rates were similar, but, like most young quarterbacks, Kaepernick was an abomination under pressure. On passes without pressure, he averaged 9.2 yards and had a 52 percent success rate. With pressure, though, his yards per pass dropped to 3.8, and the offense's success rate dropped to 28 percent. If extended to a full season, that success rate difference of 24 percent would have been the eighth highest in the league.
One final question to ask is, "How is the Falcons pass rush going to get pressure?" The irony of a 49ers-Falcons matchup is that former San Francisco head coach Mike Nolan has a defensive playbook that rivals Jim Harbaugh's offensive playbook in it's expansiveness. Given his personnel, Nolan has moved away from hybrid fronts (Atlanta's used nearly four times as many 4-3 alignments as 3-4 alignments this year), but his array of zone blitzes is alive and well. A major problem this week, however, is that San Francisco's main difficulty is against five rushers, not the three or four Atlanta sends 72 percent of the time (17th-most frequent).* What's worse, the 49ers rank second with 7.6 yards per play against three or four rushers, while the Falcons are dead last in the NFL (7.3 yards allowed per play) when rushing three or four.
If the matchup between San Francisco's offense and Atlanta's defense had counterintuitive stats, the matchup on the other side of the ball is as "duh" as it comes.
The most obvious is that Atlanta will have minimal success running the ball. The Falcons ranked 29th in rush offense DVOA (32nd when weighting recent games), while the 49ers ranked second in rush defense (eighth if weighted). Even the splits are horrible for Atlanta. They ranked 27th on first down (49ers were fourth), 26th on third down (49ers were third), and 28th in the red zone (49ers were second). Atlanta was also dead last in power-running situations, and their third-and-short running matchup is 25th versus fourth.
Hey, why stop there? The 49ers front seven led the league in second-level yards, while the Falcons offensive line was seventh-worst. In terms of where Atlanta likes to run the ball, San Francisco's front seven also ranked eighth in ALY on runs behind right tackle.
Therefore, to say that Atlanta's route to points is through the air registers as a massive understatement. Unfortunately for the Falcons, the statistical matchups for their pass offense are a mixed bag at best. On the plus side, at least Roddy White and Julio Jones don't have to face the best starting cornerback duo in the league this week: Carlos Rogers and Tarell Brown rank 54th and 62nd in coverage success rate among cornerbacks, respectively.
In addition, the 49ers ranked 17th in ASR and 20th in hurry rate (with 13 healthy games from Justin Smith), which should help Matt Ryan immensely. As was the case with Kaepernick, the Ryan-led Atlanta offense has one of the largest discrepancies in the league this season between passes with pressure and those without it. On non-hurried throws, Atlanta averaged 8.3 yards and was successful on 58 percent of plays; they averaged 4.2 yards and a 34 percent success rate on hurried throws.*
Now for the bad. The problem for Atlanta's passing game this week is more strategic than tactical, more about what they installed six months ago than what they game-planned for six days ago. It's simply the case that the Falcons offense is a unit that lines up, plays their game, and dares defenses to stop them. For instance, Atlanta had 11 personnel on the field 51 percent of the time this season, about 2.5 times more often than their next highest grouping (21 personnel).* Using that "base" offensive personnel, they passed the ball on a whopping 78 percent of plays. This works in the San Francisco's favor, as they're in their base 3-4 only 34 percent of the time.
Furthermore, as I pointed out last week in the divisional preview, San Francisco's nickel personnel group sees Chris Culliver (37th in success rate among corners) come onto the field, which shifts Rogers into the slot. Who was the 49ers worst cornerback in coverage this season? Rogers. Who is the forgotten player in Atlanta's 11 personnel grouping? No. 3 wide receiver Harry Douglas. In other words, simply by "playing their game," the Falcons offense renders a favorable matchup for them moot (i.e., Culliver vs. White or Jones instead of Rogers vs. White or Jones).
When the Falcons use their 11 personnel, it's obvious that Tony Gonzalez (second in tight end DYAR) and Jacquizz Rodgers (fourth in running back receiving DYAR) are the third and fourth receiving options, respectively -- not Douglas (75th in wide receiver DYAR). However, once again, a typically advantageous situation won't be so advantageous against San Francisco. That's because Patrick Willis (3.8 yards allowed per target, 63% defensive success rate) and NaVorro Bowman (4.8, 60%) are two of the best coverage linebackers in the league, and they have man-to-man responsibilities against the opponent's tight end and running back in nickel. Indeed, the 49ers ranked seventh in pass defense DVOA on tight end targets this season and eighth in running back targets, as opposed to 21st in targets to "other" wide receivers.
Overall, these are two mediocre special teams units: Other than Atlanta's kickoffs and San Francisco's punts, none of the component units rank in the top quartile of the league. We already detailed David Akers' travails last week, but he made his only attempt (from 36 yards out) against Green Bay, and should benefit this week from kicking in a dome. And clearly, the 49ers didn't heed my advice about taking advantage of Billy Cundiff's penchant for booming kickoffs through the back of the end zone.
As I stated at the outset, Atlanta's underdog status in this game gives no guidance whatsoever about the outcome on Sunday. Similarly, if the Falcons are able to defend the read option better than Green Bay did (which isn't asking for much), neither will the stats related to each team's running game. That leaves the passing game as this week's most likely arbiter. It's become almost cliché in the current offensive environment, but, given how much worse Ryan and Kaepernick throw the ball when hurried, whichever defense pressures the opposing quarterback more regularly will probably win the game. If neither pass rush gains the upper hand -- and both offensive lines have improved in pass protection recently -- then it will come down to which quarterback that was hand-picked by his head coach to win this very game performs more efficiently.
In light of the statistical matchups, unless White or Jones has a legendary performance against Brown, I just don't see how Atlanta will be able to score enough points with a top wide receiver, Gonzalez, and Rodgers being covered by Culliver, Willis, and Bowman.
Then again, stranger things have happened to superior DVOA teams visiting inferior top seeds in conference championship games; just ask Donovan McNabb.
DVOA (Defense-adjusted Value Over Average) breaks down each play of the season and compares it to the NFL average based on situation and opponent. You'll find it explained further here. Since DVOA measures ability to score, a negative DVOA indicates a better defense and worse offense, and a positive DVOA indicates a better offense and worse defense.
SPECIAL TEAMS numbers are different; they represent value in points of extra field position gained compared to NFL average. Field goal rating represents points scored compared to average kicker at same distances. All special teams numbers are adjusted by weather and altitude; the total is then translated into DVOA so it can be compared to offense and defense. Those numbers are explained here.
Each team is listed with DVOA for offense and defense, total along with rush and pass, and rank among the 32 teams in parentheses. (If the DVOA values are difficult to understand, it is easy to just look at the ranks.) We also list red zone DVOA and WEIGHTED DVOA (WEI DVOA), which is based on a formula which drops the value of games early in the season to get a better idea of how teams are playing now (explained here).
Each team also gets a chart showing their performance this year, game-by-game, according to total DVOA.
There are separate charts for offense and defense. In addition to a line showing each game, another line shows the team's trend for the season, using a rolling average of the last five games.
12 comments, Last at 28 Jan 2013, 9:33pm by deep64blue