Aug 25

TABM 2014 NFL Preview: Five Factors, NFC East

From year to year, the NFL experiences rapid and dynamic changes. As we explained upon the creation of this website, many of these changes have to do with regression to the mean. Because of this regression, a majority of these changes can actually be predicted. With proper research of the NFL stat books and a keen understanding of the game’s context, you can successfully predict the changes.

This is exactly what during last year’s Five Factors feature, when we published the 2013 NFL Preview. The feature had greater than 90 percent success rate, making this feature one of the best in the world at correctly predicting key factors for any given season. Therefore, we will once again spend eight August days discussing five factors of regression for each team in the NFL. This will help to paint the picture for the 2014 NFL season predictions, which will be made at the end of the month after this feature is complete.

Today, we look at the NFC East.


Dallas Cowboys (8-8, 2nd place in 2013)
Dallas CowboysUnder the guidance of Head Coach Jason Garrett, the Cowboys know nothing but mediocrity. If that wasn’t bad enough, the mediocrity comes despite has the team had going into each year. Dallas has a quarterback that can play with the best of them in the final three quarters. And better yet, the Cowboys had some very strong regression factors in their favor heading into each season. However, each year, the boys in Big D found a way to mess it up. In 2011, the defense blew five fourth-quarter leads. In 2012, the team struggled to force turnovers. Last year, the defense allowed an insane amount of yards. Each year ended with a loss in the unofficial NFC East Championship Game. Each year ended with an 8-8 record. Can the string of mediocrity come to an end? No franchise has ever played .500 football for four consecutive seasons. The defensive regression at hand would suggest better times in Dallas, but this team seem to have regression-proof problems.

Up1. Defensive efficiency: No unit was as maligned in the off-season as the Cowboys defense. Luckily, for all the talk from the media that things could be worse, they are just dead wrong. Regression says that the defense can’t really do any worse than it did in 2013. Teams moved up and down the field with regularity against the Dallas, racking up 6645 yards on 1094 plays (6.07-yard average). The 2013 Cowboys joins the list of defenses that allowed at least six yards per play, a list that has only four repeat teams (per Pro Football Reference). Every team on this list since 1960 improved its defensive efficiency the following season. It should certainly help if this Cowboys defense regresses after allowing 24.2 first downs per game (at least one more per game than every other team) and 71 pass completions of 20+ yards (at least six more than every other team), while forcing a three-and-out on a league-low 16.3 percent of drives (per Football Outsiders Almanac 2014). Regression results: fewer yards allowed per play

Down2. Red zone offense: For those ready to yet again jump on the Romo-hating bandwagon for another playoff-less year, note that the often-criticized quarterback led an offense that did abnormally well in the red zone. The Cowboys ranked third in the league by scoring a touchdown on 68.63 percent of red zone drives (per Team Rankings). Just by comparison, Dallas would’ve led the league in 2012 with this percentage. No team finished with a percentage this high in consecutive seasons since the 2004-06 Chargers. While DeMarco Murray might be one of the best running backs in the league, he’s not a scoring machine like LaDainian Tomlinson. The 2013 red zone efficiency simply won’t be matched. Regression results: worse red zone touchdown percentage

Up3. Third down offense: There was one area that Romo and passing offense did truly struggle, and it involved a key subset. The Dallas offense fizzled on the third down, converting on only 35 percent of opportunities. According to Football Outsiders, the dip in efficiency was quite noticeable. An offense that ranked 11th overall in DVOA (7.5%) ended up faring only 22nd on third down (-21.7%). One of the culprits was pass protection, as Romo was knocked down on 9.0 percent of his dropbacks, third in the league (per FOA14). However, the biggest culprit was Romo. Just look at his 2013 situational stats, according to Yahoo Sports. His third down production was a huge problem, and an abnormal issue at that. A 50.8 completion percentage and 5.8 yards per attempt is just downright unlike Romo. Those numbers are close to 60 percent and 7.8 YPA for his career on third down. Expect that subset to see improvement in 2014. Regression results: better third down conversion rate and better third down OPR

Down4. Fumble luck: According to Pro Football Perspective, the Cowboys are the luckiest team when it came to recovering fumbles in 2013. Between offense and defense, the Cowboys recovered six more fumbles than expected. They led the league by recovering 67.6 percent of fumbles, including recovering an astounding 13 of 16 on defense. The overall fumble recovery rate results in a 24-point swing in favor of the Cowboys, which marks the difference between them outscoring their opponents and them being outscored in 2013. Quite simply, there is no way this luck continues. Fumble recovery percentages are highly fluid from one year to the next, as the context often determines who gets the recovery. Regression results: worse fumble recovery percentage

Down5. Special teams play: This unit shined, especially at the skill positions. Dwayne Harris succeeded on both kick and punt returns. He averaged 30.61 yards per kick return and 12.8 yards per return. Both averages will be hard to maintain, given a generally normal set of circumstances. The kickoff rule changes in 2011 have helped to increase the kick return average, but a 30-yard average is still impressive and rare. Only the record-setting Cordarrelle Patterson posted a better kick return average. Meanwhile, the punt return average did act as regression to Harris’ outstanding average in 2012, but it may still be subject to decline. Take away Harris’ 86-yard punt return touchdown from last year, and he averages 8.95 yards per punt return. If he doesn’t return a touchdown this year, expect an average similar to that. Meanwhile, placekicker Dan Bailey is off to an amazing start to his career, converting over 90 percent of his field goal attempts through three seasons. This includes making 57-of-61 attempts over the past two seasons. What made last year’s 28-of-30 performance so impressive was that he converted 16-of-17 attempts from 40+ yards. We can’t say he’s due for certain regression, because he may be that good, but we can say he won’t have such success again from beyond 40 yards. Regression results: worse Harris kick return average or punt return average; worse field goal percentage from 40+ yards


New York Giants (7-9, 3rd place in 2013)
New York GiantsBy the end of 2013, the Kevin Gilbride offense officially wore out its welcome. For a franchise that won two Super Bowls in unbelievable fashion in the offense, it took one of the most mistake-prone seasons in recent memory for a coaching change to come. Gilbride announced his retirement in January, perhaps because the writing was on the wall. He wasn’t about to reinvent himself somewhere else after serving as an NFL coach for nearly 25 seasons. Enter the new Offensive Coordinator in Ben McAdoo, who will bring a West Coast-style of offense with him from Green Bay. Nobody has ever confused Eli Manning for Aaron Rodgers when it comes to consistency and accuracy, but this should cut out many of the errors and miscommunications of the receiver-option style of offense that led to so many turnovers last year. While all that looms as the “big issue” in East Rutherford, the defense hopes to build upon a strong second half to make this team an NFC East contender once again.

Up1. Ball protection: Turnovers were a problem from the get-go for the Giants. In the season opener against Dallas, New York turned over the ball six times. This was the first of six consecutive games in which the Giants had at least three giveaways. With no surprise, that led to the team starting 0-6. For the season, the team had 11 multi-turnover games, which led to 44 giveaways for the season. Detroit and Washington tied for second with 34 takeaways. Note that 20.1% of drives ended in a turnover. The 2001-02 Rams were the last to have 40+ giveaways in consecutive seasons (per Pro Football Reference). Given the improved passing efficiency since then, and the West Coast offense being implemented this season, we can’t imagine this trend to continue for the Giants. Don’t expect Eli Manning and crew to lead the league in interception rate (5.11 percent) once again. No quarterback has thrown 20+ interceptions in consecutive seasons since Jake Plummer in 1999 and 2000. Regression results: better turnover margin

Up2. Scoring efficiency: Thanks in large part to the turnover problems, the Giants took a huge step back in terms of scoring efficiency. After averaging 2.31 points per drive in 2012, New York only averaged 1.33 points per drive in 2013. This makes a case for elasticity-based regression. Furthermore, the Giants struggled immensely on special teams. New York allowed on average 13.65 yards per punt return. This included three punt returns allowed of 80+ yards, which marks a first since 1970. New York’s opponents started their drives on average at the 31.7 yard line (second-worst in the NFL). Finally, opponents scored nine return touchdowns. Somehow, the Giants ended the season with a better-than-expected negative-2.85 Yards per Point Differential (per Cold, Hard Football Facts). Once special teams and turnover regression takes place, this differential should actually end up going in the Giants’ favor. Regression results: more points per offensive drive and better Yards per Point Differential

Up3. Team health: While Eli Manning’s turnovers were a constant for the Giants, it might have been the only constant. That’s because the team was ravaged by injuries. In fact, according to Football Outsiders Almanac 2014, the Giants were the most-injured team found on the publication’s database. New York reportedly were victimized by 144.6 Adjusted Games Lost due to injury. This includes 82.1 AGL on offense, despite Manning starting all 16 games. Specifically the 36.3 AGL on the offensive line helps to explain why the offense went through so much turmoil in 2013. Manning was sacked a career-high 39 times just one season after leading all qualifying quarterbacks in sack percentage allowed. It seems like injury regression will be tied into elasticity-based sack regression, as Manning usually does a great job getting the ball out quickly. Finally, the defense ranked 31st in the league with 62.5 AGL, which means this unit too will benefit from a healthier 2014. Regression results: fewer Adjusted Games Lost

DownUp4. Defensive subsets: All said and done, the Giants defense ended 2013 looking in pretty good shape. Football Outsiders ranked the team sixth in Defense DVOA despite the team facing the toughest schedule on offense, according to the play-by-play metric. New York played 10 games against teams in the Top 11 in Offense DVOA, which should make it extra impressive that the Giants allowed only 4.88 yards per play. However, there are defensive subsets that cloud the picture as to how good this unit really was in 2013 and how it will relate to 2014. On the positive side, the defense should benefit from subset regression on third down. The team ranked 17th in Defense DVOA on third down and allowed a pedestrian 85.20 Defensive Passer Rating in the process. On the negative side, the defense got many breaks in terms of who it faced under center in the final 10 games. New York faced the Aaron Rodgers-less Packers, the Nick Foles-less Eagles, the Josh Freeman-led Vikings and the Kirk Cousins-led Redskins. As a result, the Giants allowed just 4.51 yards per play in that span. The schedule was actually a lot easier for the defense than the metrics say, so New York’s defensive efficiency was deceivingly good. Expect these two dynamics to converge somewhere in the middle. Regression results: better third down Defense DVOA, but more yards allowed per play

Down5. Pythagorean Win Differential: See “The Turnover Impact on the Pythagorean Win Differential” for more details on the Giants’ turnover issues affecting their Pythagorean win total. All you need here, though, is that the Giants were expected to win 5.54 games based on their points scored and points allowed. This suggests that the team was notably worse than their 7-9 record. This would make them a prime candidate for regression — if not for the regression from the Turnover Win Impact. Between the team’s PWD and TWIr, the Giants own a negative-0.23 Expected Win Differential. Therefore, we can’t say the Giants will decline in 2014. However, what we can say is that New York won’t outplay its expected win total by so much again. If Big Blue’s points scored and points allowed says it will leads X wins, expect the true record to follow that total by either rounding up or down to the nearest game (e.g.: 8.6 Pyth wins –> eight or nine true wins). Regression results: Pythagorean Win Differential closer to zero


Philadelphia Eagles (10-6, 1st place in 2013)
Philadelphia EaglesChip Kelly’s first campaign in Philly came down to a tale of two half-seasons. Over the first eight games, the Eagles offense staggered as they couldn’t find a consistent quarterback to handle the starting role. Michael Vick played mediocre ball under center and eventually gave way to Nick Foles after injury. Foles would get injured as well, which then forced Matt Barkley into spot time before Vick could come back. The Eagles started 3-5. Then came a Nick Foles start in which he threw a league-record seven touchdown passes in Oakland. Kelly’s offense would then take off, and the Eagles went 7-1 down the regular season stretch. Foles had inhuman efficiency during the span, and it made Philly one of the most offensive dangerous teams. However, they lacked the balance and lost at home to New Orleans on Wild Card Weekend. Now as Kelly enters his second season, he must find a way to establish the team’s balance to overcome the impending offensive regression.

Down1. Nick Foles: See “The Hyper-Efficient Quarterbacks: What’s Next for Manning and Foles?” for more details about the success of Foles. Here’s a quick recap. Foles threw 19 touchdown passes before his first interception, falling short of only Peyton Manning (in 2013) for best start in TD:INT ratio. (Manning started last season with 20 touchdown passes before an interception.) For the year, he held a 27:2 TD:INT ratio, which set a new NFL record. Furthermore, Foles finished with a 119.21, which led the league. His 9.12 yards per pass attempt also topped the NFL. We save further discussion for that feature. None of these feats will be repeated in 2014 by Foles. Regression results: total dip in production for Foles (TD:INT ratio, passer rating, yards per attempt)

Down2. Offensive efficiency: All said and done, the first year of Chip Kelly’s offense was a rousing success. The Eagles practiced great offensive efficiency, averaging 6.33 yards per play and 1.05 points per minute of possession. Only the 1993 49ers, 2000 Rams and 2011 Eagles improved their offensive efficiency in a year following at least a six yards per play average (per Pro Football Reference). The Eagles achieved this feat in large part due to their big-play ability. Philadelphia led the league with 80 pass plays of 20+ yards, with the next-best team totaling 68 such plays. In total, the Eagles averaged 14.2 yards per reception, with the next-best team averaging 13.2 yards per reception. Chip Kelly’s may lend for these big plays, but can we expect the Eagles to have this level of success again in 2014? It’s not likely. Teams will figure out parts of Kelly’s offense. Regression results: fewer yards per play

Down3. Run offense: Along with the big-play pass offense, the running game contributed heavily to Philadelphia’s offensive success. The Eagles totaled 2566 rush yards on 500 attempts, averaging 5.13 yards per attempt. This crew joins a select list of teams that totaled 2500+ rush yards (per Pro Football Reference). The 2004-06 Falcons are the only teams to achieve this in consecutive seasons over the past quarter century. Much of this had to do with the zone-read looks in the Kelly offense. According to ESPN Stats and Info, the Eagles averaged 5.63 yards per rush attempt on zone-road plays. If teams figure out the Kelly offense enough to slow it down like we expect, then the rush efficiency will suffer. Regression results: fewer yards per rush attempt

Up4. Red zone offense: For all the success of the offense, it could’ve been better in one key area. The red zone offense performed questionably. According to Football Outsiders, the Eagles dropped significantly from a 22.9% DVOA on offense overall (third in the NFL) down to a 1.7% DVOA on offense in the red zone (18th in the NFL). Furthermore, for all the team’s high ranking in efficiency, the Eagles were just 11th in scoring efficiency per drive (36.8 percent) and 13th in red zone touchdown efficiency (53.23 percent). Comparing Philadelphia’s overall production to its red zone subsets, it seems like the Eagles had some abnormal issues in the tight zones. We think this good of an offense should be able to produce better in the red zone. While there will be overall regression, it won’t be seen in the red zone. Regression results: closer ranking differential in red zone and overall Offense DVOA

Down5. Expected Win Differential: See “The Turnover Impact on the Pythagorean Win Differential” for full details on the regression at play for the Eagles. Certainly, turnover margin had its impact on the team’s success and unfavorable EWD in 2013. Philly’s plus-12 turnover margin accounts for a negative-0.98 Turnover Win Impact regression. This regression is certainly expected, based on two factors: (1) elasticity after a negative-24 turnover margin in 2012, and (2) the 15 forced fumbles against opposing running backs, which is four more than the next-best team. Add in the Pythagorean Win Differential, and the Eagles finished with a negative-1.59 EWI that falls past the threshold for regression. Have this applies to the team’s W-L record regression is interesting. We know this team won’t be as good as 7-1 over a half season, but a 3-5 record over a half season seems like a considerable undersell. To conclude, we can’t guarantee W-L record regression based on the bad first half last year. Regression results: EWD closer to zero


Washington Redskins (3-13, 4th place in 2013)
Washington RedskinsAfter a stellar rookie season for Robert Griffin III ended in painful defeat on Wild Card Weekend, regression was expected to force Washington to take a step back in 2013. The season ended much worse than one stung by regression. Like in 2012, Washington began the 2013 season with a 3-6 record. However, as the former regular season ended with seven consecutive wins, the latter one ended with seven consecutive losses. The final few weeks made for a particularly unnerving disaster, as now former head coach Mike Shanahan shut down Griffin for the last three games. Many put the team’s struggles on Griffin’s back, but his regression was coupled by a defensive disaster. After a Week 5 bye, Washington allowed 24+ points for 11 consecutive games.  With new head coach Jay Gruden, the media hounds are looking for a quarterback controversy between Griffin and Kirk Cousins. We’ll see if that make year three an issue for Griffin.

Up1. Scoring defense: As we already mentioned, the Washington defensed looked particularly awful last season. Besides the 11 consecutive, the defense allowed 24+ points in all but two games. The 14 times allowing at least that many points ties an NFL record previously set three different times (per Pro Football Reference). As a result, Washington allowed 478 points for the season, including 126 allowed in the first quarter. The latter figure ranked dead last in the NFL. One big factor in this was the schedule. Washington played Chip Kelly’s offense twice, Tony Romo twice, Peyton Manning, Philip Rivers, Josh McCown (in relief) and Matt Ryan. The 2014 games should collectively be easier for the defense, especially with the AFC South on deck. Furthermore, the regression points should be an overall help as well. Regression results: fewer points allowed

Up2. Special teams play: In a surprising twist, the defense wasn’t the worst unit for Washington. The special teams play was somehow worse. Washington ranked last in three key areas. First, the team posted only a 33.8-yard net punt average. A slight part of it had to do with no touchbacks from punter Sav Rocca, but much more of it had to do with the poor punt coverage. Second, the Redskins allowed 16.79 yards per punt return, when the league on average allowed 9.42 yards per punt return. Third, they totaled only 19.96 yards per kick return, finishing as the only team with a sub-20 average. Football Outsiders claimed the 2013 Redskins had one of the worst special teams units ever, finishing with a negative-12.0% DVOA. All of these issues should be able to get at least a bit better, perhaps even considerably better. Regression results: better return averages and better special teams DVOA

Down3. Pierre Garcon: Readers got a glimpse of Garcon’s special season when we discussed the great performance of Antonio Brown. Garcon and Brown join Jimmy Smith (in 2001) as the only receivers to catch at least five passes in all 16 regular season games. His consistency was amazing, tallying four games with six catches and five games with seven catches. In fact, Garcon’s two best games were a nine-reception performance against the Giants and an 11-reception performance against the Cowboys. Garcon averaged only 4.4 receptions per game in his first season in Washington. Therefore, we except regression based both on the sheer numbers and on an elasticity basis. Regression results: fewer Garcon receptions

Up4. Scoring efficiency: As we already discussed the team’s defensive issues, it should be no surprised that the team ranked last with 11.85 Yards per Point Allowed. However, those issues coupled with troubles converting offensive yards to points, as the team needed 17.71 Yards per Point Scored. That total ranked 30th in the league. In total, Washington ranked 31st with negative-5.86 Net Yards per Point Scored. Only Jacksonville and Houston were other teams to hold a margin worse than negative-five. We already know with the scoring defense regression and the special teams regression, the former scoring efficiency figure will improve. However, we think with Robert Griffin III set to return after his regression-influenced sophomore season, the latter scoring efficiency figure will improve as well. It’ll be an all-around improved affair. Regression results: better Net Yards per Point Scored

Up5. Expected Win Differential: See “The ‘Minus-7’ Win Differential and Its Elasticity” and “The Turnover Impact on the Pythagorean Win Differential” for more details on the regression at play for Washington. Here’s what you need to know. First, Washington fell from 10-6 to 3-13. Most teams who dropped at least seven games in the standings will improve their record the following season. Second, the Redskins combined a plus-1.76 Pythagorean Win Differential with a plus-0.65 Turnover Win Impact regression. This gives the team a plus-2.41 EWD that falls well past the threshold for regression. Much of this has to do with a 2-7 record in one-possession games last season, including five one-possession losses during the season-ending eight game losing streak. All said and done, we can expect a bounce back in the nation’s capital. Regression results: better W-L record

Leave a Reply

%d bloggers like this: