In a game of adjustments, the Ravens failed to find a way to adjust any of the offensive line, pass rush, or coverage and John Harbaugh tossed in an unusual game-management mistake.
I’ll go through the offensive line in a separate article and podcast, but with 6 flags, including 5 holding calls, all to center and left, that group is in flux.
Carr dropped back 35 times. For those plays, I’ve summarized the pass rush results:
A few notes on this absurd time in the pocket:
–The expected yards are based on ATS opportunities for Joe Flacco from 2010-15. Carr significantly underperformed on expected yardage for his opportunity set, but he did not have a turnover and threw 4 TDs. In this case, Carr’s 123.4 QBR was a more telling measure of his success.
–The pressure was less intrusive because the Ravens did not deliver a single QH (knock down) of Carr. The Ravens have exerted a significant physical toll on each of the first 3 QBs they faced.
–Many of the plays were significant extensions where Carr should have been able to pick apart the Ravens secondary/linebackers. Said otherwise, not all ATS opportunities are equal and Carr could easily have had more big plays.
–The Ravens didn’t have any deceptive blitzes. I use a complex definition I’ve posted before in my articles for this, but the easiest way to run a deceptive blitz is to have 2 blitzers come from off the LoS/off the slot receiver. Otherwise, pre-snap movement, drops to cover, stunts/twists, and overloads all contribute. It’s common that a team will rush 4 every snap with a comfortable lead. It’s extremely rare that a team will continue to rush unsuccessfully with a consistent scheme in a close game.
–What little pass rush success the Ravens had was at the expense of RT Vadal Alexander, number 74. He is normally a guard, but filled in for the injured Menelik Watson. Alexander was flagged 3 times for holding and once for a false start.
Results by number of defensive backs:
Some notes on the DB scheme:
–The Ravens lined up in the first quarter set (7 DBs) of the season in a 3rd-and-6 situation (Q2, 2:00). Carr ran for a 7-yard gain but the play was negated by an offensive holding penalty, so it doesn’t show above. It was, nonetheless, a good scheme and good result. The 7 DBs included the starters plus Young, Powers, and Levine.
–The lone dime (Q4, 7:01) was also the Ravens’ first of 2016. Young and Powers joined the starters and the secondary maintained coverage for an extended time as Carr left a clean pocket and rolled to the right sideline looking for a target. He eventually threw the ball away.
–Powers was activated for the first time and took over for Levine in “big nickel” situations. Levine lined up defensively on just the single (negated) snap with 7 DBs.
–Zach Orr remains unable to contribute on passes of intermediate depth. On the Raiders final drive, Carr was picking on Orr, first with the pass to Richard (Q4, 2:53) which the RB dropped, but was set for a big gain after the catch with Zach trailing. On the next play (Q4, 2:49), Carr again found Walford for a 17-yard gain covered by Orr. This was an outstanding game for Orr overall, but his plays came close to the LoS. He still appears miscast as a 3-down player.
–If nothing else, it was good to see the Ravens try to confuse the Raiders with a package other than the vanilla nickel. However, I can’t come up with a good reason why we didn’t see more dime or quarter with the game on the line.
There were several coaching decisions/oddities of note:
–The Ravens decision to go for 2 down 14-12 was correct in my view. At the time there was only 17:46 left in the game, so all even FG exchanges or lack of scoring would have resulted in a loss if the team went for 1. That set of outcomes is too likely in 17:46 and significantly outweighs the possibility of an 8-7, 7-6 or other 1-point differential in the remaining time. The difference in expected points is also very small (perhaps .06 to .08). This could be modeled a number of ways, but I’m convinced that was the right choice. It’s also likely the Ravens would have simply given back a point later when they successfully converted after Wallace’s TD.
–The failure to decline Jackson’s unnecessary roughness penalty (Q4, 13:29) was a poor choice. Everybody is saying it, so you probably don’t need much convincing, but let’s do a little math. Janikowski since 2011 by distance group:
From a pure expected points perspective (ignoring other important factors like time), using an optimal outcome (incomplete on the following play), the 38 and 53-yard FGs have a different outcome of approximately .75 points (3 X (.882-.632)). So by declining the penalty, the Ravens would have “given” the Raiders .75 points in exchange to end their drive. On the other side is the risk of:
–Improving the chance of a FG. A 10-yard gain on 3rd down after the accepted penalty would have reduced the cost of declining to approximately .18 points.
–Risking a first down by pass or penalty which entails additional touchdown risk (let’s just label that as an unknown portion of 4 points).
There are a whole lot of gross assumptions made above that I’d look to improve in an actuarial model, but the decision should not have been close. There are also other outcomes (turnover, another penalty or sack to push the Raiders out of FG range, or a gain short of the sticks which leads to conversion on the following play etc.), but I’ll simply conclude by saying I would really love to inspect the rationale for the decision chosen.
–The Ravens were caught for 1 play with what appears from the broadcast video to be just 9(!) defenders on the field. I searched my database, which extends back to 1999, and while they have played with 10 a handful of times, the Ravens had never previously been caught with just 9. Our seats are on the opposite sideline and it looked like the change got confused and 2 players who were coming on the field suddenly turned around and ran off. I’ll review the all-22 (available Tuesday) and will add a description to this article when I do.
Individual Notes (note all snap counts exclude penalties, kneels, and spikes and as such will be lower than other published totals. The Raiders ran 51 such snaps)
Elvis Dumervil returned for 19 snaps (16 as a pass rusher). He drew a holding call on Alexander and was part of 2 pressures outside the pocket (one after a flush by Weddle and another time when Car left a clean pocket). It was not an inspiring return.
Matt Judon was inactive with the return of Dumervil, but Kamalei Correa (6 snaps) remained active. Kamalei was pancaked by Penn and Olawale on a run play (Q2, 8:09). He held the right edge effectively versus the pulling Jackson (Q2, 2:25), which appeared to help slow the play to allow Urban and Orr to clean up. He was in on the final TD and tried to create a stunt opportunity for Jernigan (Q4, 2:12). He twisted underneath between LG and LT to try to pick up both blocks, but Penn easily passed him off and picked up Timmy.
Jerraud Powers (15 snaps) was active for the first time in 2016. He made my notes only for being soft on 1 coverage and being blocked effectively on the 28-yard run (Q2, 4:38).
Shareece Wright (51 snaps) did not play poorly, which I realize will be an unpopular opinion. Wright was given the unenviable job of playing LCB opposite Michael Crabtree for the bulk of snaps. That was made more difficult by the Ravens’ lack of pass rush success. Here are my notes in racing form:
–(Q1, 13:28): Has good coverage on Crabtree by right sideline, overthrown incomplete denies 3rd and 4.
–(Q1, 10:23): Carr throws PR3 (3 + 0 YAC) for Cooper who is taken down immediately by Wright.
–(Q2, 7:03): Carr throws PM16 (15 + 1 YAC) for Crabtree who is hit by Wright, but holds on. Converts 3rd and 5.
–(Q2, 4:14): Carr throws fade to Crabtree PR6 TD over Wright. No chance on well-thrown ball.
–(Q3, 10:34): Carr overthrows Crabtree approximately 25 yards down right sideline covered by Wright with Weddle over top.
–(Q3, 2:00): Carr for Crabtree 12 yards near right sideline, Wright PD from behind.
–(Q4, 6:54): Carr for Crabtree 6 yards, PD Wright denies 3rd and 14.
–(Q4, 2:59): Carr to Crabtree PR11 by right sideline, Wright very soft.
–(Q4, 2:12): Carr for Crabtree PR23 for TD in back of EZ beats Wright badly on double move. Lewis, in for Webb, takes an awful angle and is also beaten. Perfectly thrown ball makes any coverage difficult.
If you leave Shareece Wright on an island to cover Michael Crabtree you should expect him to have difficulty. In fact, he did a decent job in coverage and was the victim of 2 great passes he couldn’t reasonably be expected to defend. The time given Carr to thread the needle was a more significant factor in those 2 plays. Crabtree’s 3rd touchdown was in a zone where Weddle was closest with Webb, Mosley, and McClellan also surrounding.
While the loss to the Raiders has significant tiebreaker implications, I don’t look at this as a loss that defines the season.