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.
The Five Factors feature is a rousing success thus far. We completed the Five Factors for 2013 with a success rate greater than 90 percent. That success rate nearly reached 95 percent when we completed the Five Factors for 2014. This proves that the feature is a necessary read for any season preview. 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 2015 NFL season predictions, which will be made this Saturday.
Today, we look at the AFC North.
—————————————————————————————————————————————————————-
Houston Texans (9-7, 2nd place in 2014)
The past three Houston Texans seasons resemble a roller-coaster. The 2012 season began with an incline towards the top of the mountain, as the Texans began 11-1. However, Houston lost three of its final four regular season games and came up short during a Divisional Round loss. The 2013 season started 2-0, seemingly keeping the Texans at a playoff level, but 14 consecutive losses followed to end the season. This resembled the rapid fall to the bottom. Luckily, a regression-lead path allowed for a smooth incline back to postseason contention. Although the Texans fell short of the NFL playoffs, a 9-7 season in 2014 quickly put the franchise back in position to contend sooner than later. The “Plus-7” regression theory projects that the 2015 Texans will take a step back. How significant will it be? Does this mean Houston falls back under .500 this upcoming season? Or will the Texans still contend for the playoffs despite no franchise player at quarterback?
1. Scoring elasticity: “Plus-7” regression is based on win-loss elasticity, and it will manifest in two ways for the 2015 Texans. Obviously, scoring elasticity is one of those ways. Generally speaking, if you score more points and allow fewer points, you will win more games. And of course, in order to score more points and allow fewer points, a team must tally more yards on offense and allow fewer yards on defense. For the case of the 2013 Texans, they truly struggled to do the former despite adequately doing the latter. Houston won the yardage battle, 5556-5081, but was outscored by a 428-276 margin. The Texans owned an embarrassing negative-8.26 Net Yards per Point Scored. Meanwhile, the 2014 Texans were actually outgained (5571-5013), but owned a 372-307 scoring advantage. They finished with a plus-3.33 Net Yards per Point Scored. The double-digit differential illustrates a prime case of scoring elasticity, so expect the Texans to take a step back. Regression projection: Worse Net Yards per Point Scored
2. J.J. Watt: Watt spent the last few years setting some new standards for regression. Watt finished with 20.5 sacks, 16 passes defensed and 45 defeats (per Football Outsiders) in 2012. We correctly projected those figures would decline. In 2013, Watt tallied 10.5 sacks, 36 hits and 47 knockdowns. We correctly projected the sack-to-hit and sack-to-knockdown ratios to improve. As for 2014, Watt once again tallied 20.5 sacks. He also scored five touchdowns (three receiving, one interception and one fumble). Sack-wise, no player surpassed 20 sacks in consecutive seasons since the sack became official in 1982. Touchdown-wise, Watt is one of only three defenders since the merger to tally at least five touchdowns in a season. One of those guys was Devin Hester, who was scored six touchdowns as a rookie “cornerback.” Hester earned his non-special teams snaps at wide receiver since. The other is Hall of Fame cornerback Ken Houston, who scored all five of his touchdowns on defense in 1971. Note that Watt’s only previous touchdown came during the 2011 Wild Card game. Regression projections: Fewer Watt sacks and touchdowns
3. Run defense: While Watt was busy making plays in the pass rush, he was also helping the run defense to prevent opponents from making big plays of their own. Houston ranked second by only allowing six rushing touchdowns. Those touchdowns came from a combined 17 yards out. The league totaled 60 rushing touchdowns of 17+ yards on the season. Overall, in a very solid season for the unit, Houston allowed 1681 yards on 420 attempts. Combine that with nine forced fumbles against the run, and the Texans ranked second in the NFL (behind only the Seattle Seahawks) with a 77.04 Defensive Rusher Rating (per Ken Crippen’s formula). Expect the run touchdown percentage allowed to regress, especially since Houston allowed six passing touchdowns of the 1- or 2-yard variety. This will affect the unit’s rating as a whole. Regression projection: Worse Defensive Rusher Rating
4. Run offense: No team ran the ball more often than the Texans (551 attempts) last season. However, Houston wasn’t particularly impressive with a 3.93 yards per attempt average. Lead back Arian Foster (1246 yards on 260 attempts over 13 games) did his part with a 4.79-yard average. However, the three back-up running backs (Alfred Blue, Jonathan Grimes and Ronnie Brown) combined to average an awful 3.20-yard average. With Foster now on the shelf with an injury for a few more weeks, Blue and Grimes (and Chris Polk) with have their opportunities to produce better than they did in 2014. The Foster injury prevents the entire unit from improving, but at least the supporting cast will regress in 2015. How come? Only one team since 2008 averaged fewer than 3.20 yards per attempt. Regression projections: More yards per rush attempt from supporting cast
5. Strength of victory: See the Featured Factor of Regression entitled “The Impact from Strength of Victory” for more info. To quickly recap, Houston’s nine victories were aided by a 6-0 record against teams with 12+ losses. Since realignment, only two other teams (2005 Jaguars and 2013 Chiefs) equaled that subset win total. In fact, Houston’s median strength of victory was against a 3-13 opponent. We believe that to be the weakest of any plus-.500 team in NFL history. With a tougher schedule at hand, Houston’s 2-6 record against quality opponents comes more into play. Regression projection: Worse W-L record
—————————————————————————————————————————————————————-
Indianapolis Colts (11-5, AFC South Champions in 2014)
No matter how convincing the case may be for the Colts to take a step back, they seem to keep taking a step forward. Through the first three seasons with Chuck Pagano as head coach and Andrew Luck as starting quarterback, Indianapolis finished with a 11-5 each time. However, in postseason, each season resulted in an advancement one round further in the postseason. The 2012 season ended with Wild Card Weekend loss to the eventual Super Bowl champions in the Baltimore Ravens. Then the 2013 season ended in the Divisional Round at same place of which the Colts lost at in the AFC Championship Game last year. Those losses came in Foxborough, the cite of which Indianapolis lost 45-7 and set off this whole Deflategate circus. The playoff losses underscore what the Colts are missing to become champions: a strong running game and a playoff-caliber front seven. Can they finally fix those problems this year?
1. Pass offense: Led by Andrew Luck’s 40 touchdown passes (and Matt Hasselbeck’s two touchdown passes), the Colts enjoyed their best offensive season since Peyton Manning left Indianapolis. The Colts’ 42 touchdown passes led the NFL, as did their 78 pass completions of 20+ yards (at least 11 more than any other team in the league). Amazingly, 26 of those completions were between 20 and 22 yards. Meanwhile, Luck’s touchdown percentage improved from 4.04 to 6.49, so his play is subject to some elasticity. It helped that Indianapolis played six games against Bottom 8 teams in Defensive Passer Rating. As a result, things should take a step back in 2015. Regression projection: Worse Offense Passer Rating
2. Run offense: The Colts may take a step back in the passing game, but they can make up for it by improving the run game. Indianapolis honestly doesn’t even need to be as explosive as it was in 2014; it needs to show better offensive balance. The 690:415 pass:run ratio isn’t the balance fitting of a Super Bowl champion. Perhaps that’s just what the Colts were thinking when they signed veteran running back Frank Gore. At 32 years old, Gore is at risk of quickly decline. However, he’s been as solid as they come over the past decade. Gore averaged at least four yards per attempt in each season of his career. He replaces Trent Richardson, who was an unmitigated disaster with the Colts. He averaged 3.09 yards per attempt in just under two seasons, while his teammates (running backs only) averaged 4.66 yards per attempt between 2013 and 2014. Signs point to Indianapolis performing better on the ground due to this change at running back. Regression projection: More rush yards
3. Subset defense: While the lack of a run game truly hurt the Colts, their failure to develop a capable front seven perhaps hurt even more. Indianapolis ended their season allowing 42+ points for the fourth time during the season (including the playoffs). This was what happened when going up against elite offenses like Dallas, Pittsburgh or New England’s. These problems were also apparent in red zone situations. Indianapolis allowed 33 touchdowns and 11 field goals in 51 red zone trips. That results in a 4.98-point average per trip. The saving grace for the Colts was their third down defense, as they ranked second in third down conversion percentage allowed (67-of-201, 33.33 percent). Indianapolis also tied the NFL-best with only four touchdowns allowed in the subset. Expect these subsets to converge, which keeps the front seven questions alive in 2015.
Regression projections: Better red zone touchdown percentage allowed and worse third down conversion percentage allowed
4. Team health: The Colts reached the AFC Championship Game despite frequent injury issues. Running back, offensive line and defensive back were the three biggest weak spots for Indianapolis. Overall, per Football Outsiders, the team finished with 104.7 Adjusted Games Lost (30th in the NFL). This now marks back-to-back seasons with triple digits in this metric. Could this be a developing trend? We don’t think so just yet. Regression projections: Fewer Adjusted Games Lost
5. Ball security: In the Colts games, both Indianapolis and the opponent seemed to have trouble hanging on to the ball. Fumbles and drops were two common themes. Indianapolis fumbled 31 times on the season, losing 15 of the fumbles. Meanwhile, the Colts dropped 40 passes on 647 targets, leading to a league-high 6.2 drop percentage (per SportingCharts). On the flip side, their opponents fumbled 29 times and lost 14 of them. They also dropped 36 passes on 533 targets, leading to a league-high 6.8 drop percentage. The change of opponents will certainly affect one aspect of regression. However, the Colts need to find a way to limit drops. New acquisition Andre Johnson should be of help in that regard. The sides will break even in regression. Regression projections: Fewer fumbles and better drop percentage for Colts and their opponents
—————————————————————————————————————————————————————-
Jacksonville Jaguars (3-13, 3rd place in 2014)
The battle for relevance has been an incredibly tough as of late for the Jaguars. Since 2011, Jacksonville owns a pathetic 14-50 record, putting the franchise at the bottom of the rankings in that span. A good part of this can be attributed to quarterback play. The Jaguars unwisely let go of David Garrard to give rookie first-round pick Blaine Gabbert the reins. He flopped out of the position, compiling a 5-22 starting record and a negative-18.53 Relative Passer Rating over three seasons in Jacksonville. His (and Chad Henne’s) failure led to rookie first-round pick Blake Bortles becoming the man under center. His rookie season didn’t show much promise either. Between a 3-10 starting record and a negative-19.36 Relative Passer Rating, Bortles is already in jeopardy of washing himself out. Hopefully for him, it was just a bunch of rookie growing pains that build him mental toughness for the future. Note that he didn’t get much help from his run game or defense. Head coach Gus Bradley was hired to at the least turn things around defensively, so 2015 better show some signs of that.
1. Offensive efficiency: As Bortles struggled in his 13 starts, the Jaguars finished 31st in yards per play. They totaled 4634 yards on 988 plays, ending with an awful 4.69-yard average. This problem boils down to Bortles’ growing pains as a passer. Jacksonville ranked last with 4.78 yards per dropback. This paled in comparison to Bortles’ 7.48-yard rushing average (419 yards on 56 attempts). With experience under his belt, Bortles should be able to open up the offense more. Meanwhile, running back Toby Gerhart also struggled during his first season in Jacksonville, totaling only 326 yards in 101 carries (3.23-yard average). He will be replaced on the depth chart by rookie T.J. Yeldon. The pieces are in place to improve this offense, and this will supplement the regression in place for per-play efficiency. Regression projection: More yards per play
2. Pass protection: Improved pass protection will also help Jacksonville’s per-play offensive efficiency. That was a major problem for Jacksonville, either with Bortles or Henne under center. The Jaguars allowed 71 sacks, the sixth most since the NFL-AFL merger. Only one franchise faced consecutive seasons of 70+ sacks allowed. That was the 1986-87 Eagles, who regressed from an NFL-record 104 sacks to 72 in the second season. The right side of the offensive line is all new, led by the additions of center Stefen Wisniewski and right tackle Jeremy Parnell. They’ll work with young left tackle Luke Joeckel. They also should help give Bortles more time as he progresses in his sophomore season. Regression projection: Fewer sacks allowed
3. Pass rush: Switching to the defensive side, the Jaguars best illustrated their progress under Gus Bradley by accumulating 45 sacks. Unfortunately, that sack total was misleading, in terms of the overall pass rush production from the unit. According to Pro Football Focus, Jacksonville tallied only 38 hits and 125 hurries. These are low totals, especially for a team that averaged nearly three sacks per game. Credit the secondary for this. According to Football Outsiders Almanac 2015, Jacksonville earned a league-high 12 “coverage” sacks. This supplemented the 29th-ranked front four when it came to pressure rate. The normalization of the sack total relative to front four pressure will lead to the Jaguars declining in this area. Regression projection: Fewer sacks
4. Penalties: No team committed fewer penalties than Jacksonville’s 73 last year. Those penalties combined to total a league-low 573 yards. The Jaguars held a seven-penalty and 62-yard advantage on the rest of the league. Jacksonville also adequately forced penalties, so the team finished with 42 net penalties (best in the NFL) for 313 net yards (second in the NFL). This is a stat that often fluctuates from one year to the next. Look no further than the Jaguars going from negative-9 to plus-42 in penalty differential last year. That fluctuation will continue to be the case here in 2015. Regression projection: Worse net penalties
5. Julius Thomas: Like with DeMarco Murray in Philadelphia, this bit of regression involves a standout skill player heading to a new team. In the case of Julius Thomas, he goes from a dynamic offense in Denver to an anemic offense in Jacksonville. Thomas caught 43 passes for 489 yards and 12 touchdowns in 2013. He became the first tight end in NFL history to post back-to-back 12+ touchdown seasons. It’s impressive that Thomas did this without much action overall. There are eight receivers with 48 receptions or fewer to reach the 12-touchdown plateau (per Pro Football Reference). There won’t be Peyton Manning getting the ball to Thomas anymore, so expect a shift in statistical production. Regression projection: Fewer Thomas touchdown receptions
—————————————————————————————————————————————————————-
Tennessee Titans (2-14, 4th place in 2014)
Tennessee began its 2014 season began with a convincing win at Kansas City. It was a quality win that provided an early statement to back up an unexpected 7-9 finish in 2013. However, many of the problems that were outlined in the 2013 NFL Preview and 2014 NFL Preview eventually became evident, as the Titans lost 14 of the final 15 games of the season. Despite that poor record, Tennessee decided to retain Ken Whisenhunt as head coach. The hopes is that he can build up the team while developing new quarterback Marcus Mariota. Drafted second overall, Mariota has a statistical profile that heavily favors him to thrive as the franchise quarterback. However, that statistical profile may be skewed by favorable offensive system used at the University of Oregon. Either way, Mariota brings an interesting style of play that may force defenses to reconsider their approach against Tennessee. Can Whisenhunt ease his rookie’s transition under center?
1. Offensive drive efficiency: Marcus Mariota will get to start from Week 1 in hopes boosting an offense that ranked last in plays per drive. The Titans Tennessee went “three-and-out” in 50 drives while committing a turnover within three plays in 13 other drives. Particularly with Zach Mettenberger at quarterback, the Titans failed to grab a first down in 34 of 64 drives, according to Football Outsiders Almanac 2015. It wasn’t for a lack of early-down running game, either. Tennessee’s 4.24 rush yards per first down attempt (per Pro Football Reference) ranked 17th in the league. It was quarterback play that contributed the most. That’s Mariota comes in. It remains to be seen if Whisenhunt and crew can scheme to Mariota’s advantage and keep opponents off balance a la Robert Griffin III’s rookie season in Washington in 2012. However, if the Titans can just do that in flashes, it would go a long way to help this offense. With more plays per drive normally comes more points per drive. Regression projection: More points per drive
2. Defensive drive efficiency: As bad as the offense was, the defense didn’t fare much better. Tennessee allowed 357 first downs over 191 drives. Only the Atlanta Falcons (358) allowed more first downs. The pass defense had it a bit tougher, allowing 211 of those first downs on just 584 dropbacks. However, the run defense did its fair share of making life miserable. No team allowed more than Tennessee’s 46 rushing first downs and 12 rushing touchdowns on first down. The former shows how truly poor the run defense was. The latter was influenced much by context, as six touchdowns came in the 1-yard variety. We are following a similar line of logic from the offensive drive analysis. A team that allows fewer first downs per drive generally allows fewer points per drive. Tennessee can achieve this simply with better play on first down. Regression projections: Fewer points allowed per drive
3. Takeaways: Tennessee experienced some bad luck on defense as well. The Titans ranked 29th in the league in takeaways, large in part because Tennessee’s opponents lost a league-low four fumbles. Granted, the Titans only forced seven fumbles, but this much is subject to regression as well. Every other team except the Arizona Cardinals at least doubled Tennessee’s forced fumble total in 2014. That’s just simply too low to maintain. Fumble plays are dictated by context, both in the force (occasionally) and the recovery (often). Tennessee simply faced some bad luck in forcing fumbles. Of note, only three of Tennessee’s 39 sacks forced a fumble. Compare that to division rival Jacksonville, who forced 12 over its 45 sacks. Perhaps at a later date, we’ll study the trends of strip-sack rates. For now, though, we will use it as part of our case for takeaway regression. Regression projection: More takeaways
4. Big-play pass defense: For all the went wrong in Tennessee, at least the pass defense performed solidly in some regards. The Titans only allowed 23 passes of 25+ yards, which ranked fifth in the league. It helped to face a slate of opponents with a median 26.5 passes in this subset. It also helped that the Titans so often played from behind, so opposing teams didn’t have to air it out as much. No team faced as many rush attempts as the Titans did last year. Also, theoretically speaking, opponents are not as willing to force passes when nursing leads. Facing more aggressive offenses in 2015 will likely lead to more big plays allowed, even within a similar sample size. Regression projection: More yards allowed per pass attempt
5. Expected Win Differential: See the Featured Factor of Regression entitled “Updating the Expected Win Differential for 2015” for more details. Any time a team goes 2-14 in the NFL, there is an inherent amount of bad luck. Teams are rarely ever that bad, but rather they suffer bad breaks that compound their poor play and manifest prominently over a 16-game season. In the case of the 2014 Titans, they finished with 3.41 Pythagorean Wins. However, they went 1-5 in one-possession games. Add in 0.86 wins worth of turnover regression, and the Titans come away with a plus-2.27 EWD heading into 2015. This suggests that the Titans should at least take one step forward, even if it’s merely from playing with the same quality as before. Regression projection: Better W-L record