Aug 24

TABM 2014 NFL Preview: Five Factors, AFC 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 AFC East.


Buffalo Bills (6-10, 4th place in 2013)
Buffalo BillsIn terms of team success, we can’t seem to get a different story for this franchise. For yet another year, the Bills played poorly enough to stay out of playoff contention down the stretch but well enough to stay out of the league’s basement. As a result, Buffalo remains in this purgatory of not being able to get franchise-changing talent (like Jadeveon Clowney) at their original drafting position while having to go through a carousel of coaches and quarterbacks to find the correct fits. Last year saw the first head coaching trial of Doug Marrone and the rookie campaign of E.J. Manuel, which finished with the typical 6-10 record. This marks the third consecutive season with a such record, as well as the seventh time in eight years the team went either 7-9 or 6-10. Something’s eventually got to give, right? Well, Buffalo is looking for that to happen, and did so by adding more playmaking ability, as the team moved up the draft board to select wide receiver Sammy Watkins.

Down1. Big-play defense: See “The League’s Sack Percentage” for the full details of Buffalo’s pass rush. Perhaps the most frustrating part when looking at the past 14 years of playoff-less football in Buffalo is that Bills are generally really good at something in each season. The team was no worse than 6-10 in 11 of those 14 seasons, thanks to that bright spot on the team from one year to another. Last year, that bright spot came from the team’s ability to make big plays on defense. Buffalo ranked second in the league with 57 sacks and third in the league with a 9.22 sack percentage. This is abnormally high relative to the league’s 6.67 sack percentage average. This rate will likely regress, because the team seemed to have it all come together for the pass rush at the same time. Three players tallied double-digit sacks: Mario Williams (13.0), Kyle Williams (10.5) and Jerry Hughes (10.0). Both Williams and Hughes set new career-highs by at least five sacks. Meanwhile, the Bills ranked second in the league with 23 interceptions and a 4.10 interception rate. Buffalo had only 12 interceptions in 2012, so expect some elasticity-based interception regression in 2014. The Bills have shown over the past decade and a half that they cannot maintain the success of being really good in any given area from one year to the next. We cannot expect the league-leading 12.94 Negative Pass Play Percentage to happen again for the Bills. Neither can expect the team to build on its huge efficiency improvement. The Bills went from 27th (10.6% DVOA) to fourth (-13.8% DVOA) in play-by-play efficiency (per Football Outsiders). There’s too much elasticity at hand. Regression results: lower NPP% and worse Defensive DVOA

DownUp2. Drive-based defense: While there’s been that one thing that the Bills are really good at doing, it gets overshadowed because the team cannot capitalize on those strengths. Just look at the impact of the big-play defense. The team ranked second in yards allowed per drive (24.93-yard average), but 12th in points allowed per drive (1.68-point average). This is explained in part by the average starting line of scrimmage (31.5-yard line, 30th in the league). However, the bigger problem comes with the defense being on the field for a league-leading 214 drives. This points to one issue: the Bills trailed only the Eagles in situation-based pace. Note the importance of the pace being “situation-based,” as the Eagles went 7-1 in the second half of the season and often milked the clock late in the game. The Bills didn’t have such luxuries. The high offensive pace kept sending out the defense on the field, often times in compromising situations. This didn’t maximize the impact of an efficient defense that allowed 4.90 yards per play. If the Bills improve their offensive efficiency and maintain their situation-based pace, it could help the team. But the jump will need to strong than the regression at hand for the Buffalo defense. Buffalo hopes Manuel got the bang for his buck in terms of a learning curve. Regression results: better opponents LOS drive start, but more yards allowed per drive

DownUp3. Run offense: With Manuel still learning the ropes of the NFL, the fast-paced offense put most of the pressure on the rushing attack. As a result, the Bills led the league with 546 rushing attempts and ranked second with 2307 rushing yards. This isn’t normal of a losing team, as only three such teams have totaled 540+ rushing attempts since 1987 (per Pro Football Reference). On a basic quantitative-based level, this looks like a major asset for the offense. However, the heavy workload caused the team’s rushing efficiency to suffer. The 15 rushing touchdowns (10th in the league) and 4.23 yards per attempt average (14th in the league) showed that quality didn’t meet quantity. Perhaps most telling were the league-leading 32 fumbles and the 17th-ranked Run DVOA (-4.0%). The worst of it was experienced by C.J. Spiller. One year after quietly putting together an elite-quality season, Spiller played like one of the least-efficient runners in the league. According to Football Outsiders, he went from a 27.6% rushing DVOA (first among qualifiers) to -17.8% (43rd). Elasticity will certainly help out Spiller’s case in 2014, but the coaching staff must find what best suits their dynamic running back. Regression results: better Spiller rushing DVOA, but fewer rushing yards

Down4. Pass defense: When discussing the big-play defense, the success of the pass defense was briefly mentioned. The 23 interceptions certainly played a big role for a unit that thrived most of the season. However, the Bills thoroughly handled their business against the pass by ranking third with 5.29 Defensive Real Passing Yards per Attempt, third with a 74.91 Defensive Passer Rating and second with a 65.53 Defensive Real Quarterback Rating (per Cold, Hard Football Facts). Knowing that the sacks and interceptions will both regress, that should affect the latter two stats for the worse. Also, consider that Buffalo’s 55.26 defensive completion percentage ranked best in the league and nearly six points better than the league average. Therefore, there should be some regression that impacts all three stats. Finally, look at the personnel changes that will hurt the Bills. Defensive coordinator Mike Pettine became the head coach in Cleveland. Safety Jarius Byrd signed as a prized free agent in New Orleans. Middle linebacker KIko Alonso suffered a torn ACL in July and will likely miss the season. (Update on 8/26: Alonso was placed on the reserve/non-football injury list, which means Buffalo will decide whether to put on season-ending IR or not by Week 6.) Those three changes alone may send this pass defense back towards the league’s middle third of teams. Regression results: worse DRPYPA, DPR and DQBR

Up5. Return game: Two years ago, the Bills saw a historic season of punt returning by Leodis McKelvin bolster the unit. Last year, McKelvin was one of the worst punt returners in the league. As a result, Buffalo dropped from 17.12 yards per punt return to 6.23 yards per punt return. Furthermore, according to Football Outsiders, the punt returning efficiency dropped from second in the league to dead last. As a result, the Special Teams DVOA dropped from 9th to 30th. This went above and beyond the expected regression. Now, there is some elasticity at hand. In general, punt return average varies a lot from year to year. Maybe Buffalo will finally see a normal season of punt returning. Regression results: more yards per punt return


Miami Dolphins (8-8, 3rd place in 2013)
Miami DolphinsThe media’s narrative on the team’s bullying scandal drowned out the actual football narrative presented to the public eye. However, Miami’s football narrative was greatly influenced by the chaos in that locker room. The problem with the bullying focused primarily on the offensive line, and it showed with the team’s league-high 58 sacks allowed. That made life very tough for the running game, which couldn’t provide consistent production from one game to another. Worse yet, it made life very tough for Ryan Tannehill, who’s still learning as a quarterback and is coming to a career crossroads over the next few years. As a result, the offense never really took off to the point of becoming a playoff-quality unit. Eventually, Miami fell apart down the stretch, losing its final two games while scoring a combined seven points. Entering 2014, the defense remains a strength while the offense still has questions. Will the offensive line regression be enough to vault this team into the playoffs?

Up1. Pass protection: The offensive line’s struggles clearly provided the biggest issue for this team. Not only did the 58 sacks allowed lead the league, but it tied the league-high for the most allowed since the 2006 Lions and 2006 Raiders. The list of teams that allowed 56+ sacks in the past decade is very small (per Pro Football Reference). This suggests that any normal progression from 2013 to 2014 will lead to some sort of improvement. Perhaps it also comes with an improvement of the team’s 28th-ranked Offensive Hogs, which was weighed down by the 11.81 Negative Pass Play Percentage (per Cold, Hard Football Facts). And, perhaps it comes because the team commits more penalties to limit the sacks. According to the NFL Penalty Stats Tracker, the Dolphins tied for a league low with 30 accepted offensive penalties, half of which came in pre-snap situations. Miami had only 12 holding penalties in 2013, so maybe an uptick there leads to fewer sacks allowed. Regression results: fewer sacks allowed

DownUp2. Ryan Tannehill: Just by looking at the pass protection issues, it wasn’t a pretty 2013 season for Tannehill. However, it could’ve been much worse. He owned a 24:17 touchdown-to-interception ratio, but that was aided by opponents dropping 11 interceptions (per Football Outsiders). In 2012, a rookie Andrew Luck led the league with 14 dropped interceptions. He followed that up by throwing 14 passes that were charted as “interceptions” or “dropped interceptions,” showing that he made significant improvement in his second season. Will we see that type of progression from Tannehill? Probably not, as he’s not a quarterback built like Luck. What we do know is that the dropped interceptions will decrease. Does that mean it leads to more interceptions thrown or fewer passes that become potential interceptions? The data examined by Football Outsiders says the “adjusted interceptions” correlates twice as well from year-to-year as interceptions alone, but the 0.32 coefficient still shows a weak relationship. Odds are the regression involves a little bit of both. Regression results: fewer Tannehill “dropped” interceptions

Up3. Turnover battle: Last year, the Dolphins had a seemingly innocent minus-2 turnover margin. However, this margin is subject to regression and could be the tiebreaker for Tannehill’s case. This margin acts as the fifth consecutive season in which Miami lost the turnover battle. Since the merger, there have been 10 instances of a franchise going at least six consecutive seasons with a negative turnover margin. However, it last happened with the 2002-07 Browns. It’s happened only thrice in the past two decades, only including the 1991-96 Rams and 1992-98 Raiders. Note that Miami would need just an even turnover margin to break this trend, so if Tannehill throws a few more interceptions, it won’t be the end of the day for this regression. Frankly, the defense could do better than 24 takeaways, given its quality of play. We expect them to make more big plays. Regression results: better turnover margin

DownUp4. Team subsets: This area provided a tale of two units more than anything else in 2013. A bad offense played worse in key subset situations. A good defense played better in key subset situations. Miami finished with an 80.13 Offensive Passer Rating. However, the rating dropped to 63.12 on third and fourth downs, as well as 56.15 in the fourth quarter and overtime (per Pro Football Reference). Miami also finished with a 77.26 Defensive Passer Rating. Better yet, the rating dropped to 63.12 on third and fourth downs, as well as 63.81 in the fourth quarter and overtime (also per PFR). Both of these subsets strayed from the mean by a considerable amount, especially on offense. This dynamic likely had to do with a horrific pass protection and raw starting quarterback. The defensive dynamic isn’t as clear, which is perhaps more worrisome. That might mean it’s a fluke. Therefore, we think regression becomes a double-edged sword. Regression results: better subset OPR, worse subset DPR

Down5. Pass defense: The Dolphins pass defense held up quite well last season, allowing only 17 passing touchdowns on 580 attempts. That makes a 2.93 opponents’ touchdown percentage. It’s tough, especially in this era, to keep opponents under three percent. That greatly affected the team’s fifth-ranked 77.26 Defensive Passer Rating. Miami did get a pass from its 2013 schedule by playing the Browns, the Ravens, the Bills (twice), the Buccaneers and the Jets (twice). They even faced the Patriots twice during a down year. Expect the schedule to be a bit tougher. The AFC East will be better on offense, especially with New England. Meanwhile, the NFC North and AFC West have at least five tough offenses to stop. Miami will not repeat its successful touchdown percentage allowed, which affects passer rating. Regression results: worse Defensive Passer Rating


New England Patriots (12-4, 1st place in 2013)
New England PatriotsAnother year, another AFC Championship Game appearance. New England made its eighth conference championship game in the last 13 seasons. However, the franchise’s quest for its first Super Bowl title since the 2004 season once again came up short. This time, the team was slowly bludgeoned to death by a historically efficient and powerful Denver Broncos offense. What’s remarkable is that this run to the conference title game came during a season in which the passing offense had a total roster overhaul that truly affected Tom Brady. Some could say that the Patriots were lucky to play in the weaker conference and in a weak division. However, by this time, it’s obvious this franchise just knows how to win. New England is just looking to turn it into another championship before Tom Brady’s career comes to an end. What will regression have to say about the 2014 season, and can that lead to another deep playoff run? There are multiple factors at play for regression.

Down1. Clutch play: The New England offense played its worst since at least 2006, but the Patriots made up for it by leading the league with five fourth-quarter comebacks. It very narrowly was six, if not for a controversial no-call in a Monday Night Football loss in Carolina. This was spurred by the team’s league-best 182 points scored in the fourth quarter through the regular season. This includes an insane 126 points over the final eight games. Teams normally don’t see this comeback ability replicated in back-to-back seasons. Only Steve Bartkowski (1978 and 1979) made at least five fourth-quarter comebacks in consecutive seasons (per Pro Football Reference). Don’t expect Brady to be the second. Regression results: fewer fourth-quarter comebacks

Up2. Tom Brady: A big part of the reason the Patriots led the league with fourth-quarter comeback is that the offense didn’t have its normal success. With the said previous success, the team wouldn’t need the comeback to win. Just at Tom Brady’s numbers from 2013 compared to his numbers from 2004-12. Brady first started putting up prime numbers during his third Super Bowl campaign, when his yards per pass attempt average improved from 6.87 to 7.79. That began an eight-year span in which Brady put up a 100.38 passer rating with 7.83 yards per pass attempt and a 6.00 touchdown percentage. Last year, Brady posted a 87.29 passer rating with 6.92 yards per pass attempt and a 3.98 touchdown percentage. This makes it clear that Brady is due for some elasticity-related regression this upcoming season. Regression results: better Brady passer rating

Up3. Third down offense: With Brady’s numbers dipping, the Patriots took a large hit on third down. The team’s third down efficiency, which featured a 39.84 percent conversion rate, dragged down an offense that actually put up great play-by-play metrics. New England actually ranked fourth in Offense DVOA (16.4%), according to Pro Football Reference. In comparison, the Offense DVOA drop to 23rd (-11.7%) on third down. For an offense that routinely dominated for years, it’s just unfathomable to see them playing like a below average offense in one of the game’s key subsets. However, it happened. We just don’t see it happening again, considering Brady’s rebound. Regression results: better third down Offense DVOA and better third down conversion rate

Down4. Placekicking: With this somehow-still-efficient offense struggling on third down, it made many potential touchdown drives settle for field goals. Luckily, placekicker Stephen Gostkowski picked up nearly all of the slack. The kicker converted 38 of 41 field goal attempts, including 16 of 19 from 40+ yards. Along with Justin Tucker, the reliable Gostkowski became one of just six kickers to convert 38+ field goals in a single season (per Pro Football Reference). No kicker achieved this in back-to-back, as much of this field goal success has to do with fluid context from one year to another. Many of these field goals can turn to touchdowns, and some could turn to punts. Either way, that hurts Gostkowski’s fantasy value. Regression results: fewer field goals made

Down5. Overall W-L record: If not for the timely plays and nearly a win’s worth of turnovers, the Patriots likely would’ve been a team that had to settle for the AFC’s fourth seed. According to our Expected Win Differential, the Patriots were one of the primary examples of teams that outplayed their value. Their negative-2.21 EWD was the second-worst total in the league. In other words, this 12-4 team played more like a 10-6 team. In the end, the EWD surpasses the threshold for regression, so expect New England to decline in 2014. The question is “how much,” as Brady’s elasticity-based regression should help to ease the decline. With a normal amount of luck, we’re looking at potential “improvement” from 10-6. Therefore, if this team is to improve from 10-6 and decline from 12-4, we get… well, let’s save that for the season predictions. Regression results: worse W-L record


New York Jets (8-8, 2nd place in 2013)
New York JetsIn the first two years of the Rex Ryan era, the Jets made miraculous runs to the AFC Championship Game. These runs can be considered “miraculous” because of the poor play at quarterback. However, that quarterback play arguably cost at least one of those two teams a chance at the Super Bowl. The 2009 Jets actually had a better relative Defensive Passer Rating than the 2013 Seahawks, only for their offense to have a relative Offensive Passer Rating that was among the worst in the past decade. The 2010 team regressed in both areas as expected, but the offense still wasn’t even close to being at playoff quality. It’s been a hallmark problem for Ryan, which was none more evident than last season. With a new rookie quarterback under center, the Jets languished with the worst Offensive Passer Rating in the league. Luckily, in typical Rex Ryan fashion, the team somehow put together enough wins to finish the season with a respectable outcome. The Jets went 5-1 in one-possession games before the bye week en route to the 8-8 record on the season. What’s now in store for Geno Smith’s second season under center?

Up1. Turnover battle: New York finished with the league’s worst Offensive Passer Rating in large part because the rookie Geno Smith struggled to consistently take care of the football. Smith threw 21 of the team’s 22 interceptions, which is definitely not grounds for having a winning season or making the playoffs these days. The 2010 Saints were the last playoff team to throw at least that many interceptions. Still, that wasn’t the only problem for the Jets. New York totaled only 15 takeaways, as opposed to 29 giveaways. That gave the team a minus-14 turnover margin. The issue with causing turnovers is a problem the Ryan twins have struggled with over the years. Last year’s battle was one grossly hampered by luck. According to Pro Football Perspective, the Jets recovered 6.5 fewer fumbles than expected on the season. We’re talking about a 26-point swing here alone. No wonder the Jets defense recovered an NFL-record low two fumbles on 18 attempts. We’re talking about recovering only 11.11 percent of fumbles on defense! Change that luck alone, and the Jets handle the turnover battle better. Regression results: better turnover margin

Up2. Receiver production: Another reason Smith struggled during his rookie season was that he didn’t have a standout receiving option to bail him out of the tough situations. Generally speaking, a singular receiver gets overrated by the Mainstream Media and the masses, but the receiving corps gets underrated. The exception to this rule generally involves the young developing quarterback. Over the first 30-40 starts in the NFL, a quarterback may tend to lock onto targets until he feels a rhythm and presence when going through his receiving options. Great receivers can help the young quarterback by giving him options that most opponents can’t defend. In no way did Smith enjoy this luxury. New York’s leading receiver totaled 523 yards, making the Jets one of only two teams since 2005 that didn’t have a single receiver with 550+ yards. Once again, the leading receiver for the Jets was Jeremy Kurley. He has 99 receptions for 1350 yards over the past two season as the team’s leading receiver. Note that these numbers were topped seven times by a receiver in a single season in 2012 or 2013 (per Pro Football Reference). This includes Andre Johnson surpassing those numbers in both seasons! Smith may finally get some help from Eric Decker, who gained a habit of getting open when working with Peyton Manning the past two seasons. He has 172 receptions for 2352 yards in those two seasons. This will allow Kurley to become a better fit as the team’s complimentary receiver as well. Regression results: more than 550 yards for the leading receiver

Down3. Run defense: Rex Ryan does a consistent job of helping the front seven to excel. While the pass rush was surprisingly lacking, which we’ll talk about in the next section, the run defense was arguably the league’s best. On 421 rushing attempts, New York’s opponents totaled only 1412 yards to give them a 3.35 yards per attempt average. That made the team the only one to keep opponents under 3.5 yards per attempt. Furthermore, New York allowed a first down on only 16.2 percent of opponents’ rushing attempts. That led the league. The Jets were one of only two teams to hold that stat under 17.5 percent. It certainly helped the run defense when the team had four blowout losses (by 17+ points), as the unit could stack the box in garbage time without any fear of repercussion. With the likes of Muhammad Wilkerson and Sheldon Richardson, this front seven will be still one of the best at defending the run. Just don’t expect it to get any better in 2014. Regression results: more yards allowed per rushing attempt

DownUp4. Third-down play: Consider this to be the opposite case of what happened with the Dolphins. The Jets had a bad offense that actually performed pretty well on third down. Who would’ve guessed that the Jets finished with a better third down Offense DVOA (6.4%) than the Patriots (per Football Outsiders)? The unit finished 27th in that play-by-play metric overall. Meanwhile, who would’ve guessed that the Jets defense did worse on third down (11.5% DVOA by opponents) than the offense? That unit finished 12th overall. As this relates to conversion rate, it’s surprising to see the league’s worst passing offense convert 38.96 percent of third down opportunities. It’s also not likely to happen again if the Jets don’t show some grass roots improvements on offense. Meanwhile, the defense only allowed third down conversions on 36.80 percent of opportunities, but it also allowed some big plays. That’s because the team got pressure on the quarterback in a league-low 20.6 percent of dropbacks (per Football Outsiders). Worse yet, the team created a turnover on only 2.16 percent of third down plays, as compared to allowing 25+ yards on 6.03 percent of third plays. That’s equal to the team’s “big play” percentage on third down (re: sacks and takeaways). Expect a two-pronged batch of regression. Regression results: worse third down offense DVOA; better defensive third down big play percentage

Down5. Overall W-L record: Looking at the team’s point differential, it’s a miracle the Jets finished 8-8. No team since the merger with a non-losing record finished with a worse point differential than New York’s minus-97 total. As a result, the team finished with a 5.34 Pythagorean wins. Luckily for New York, the Turnover Win Impact regression helps out the situation some, as the Jets would get back 1.14 wins in an average situation. This makes for an interesting dynamic, because Geno Smith enters his second season. The question now involves “how much” will this team regress in record due to its 2013 Expected Win Differential. New York owned a negative-1.52 EWD, which slightly surpasses the threshold for regression. We say this team declines in record, but the decline will be limited if Smith better protects the ball. Regression results: worse W-L record

Leave a Reply

%d bloggers like this: