His offense was the epitome of speed, scoring, and football style. Three years ago, he took the NFL by storm. But it’s been downhill ever since. What happened to one of the game’s greatest minds? And can he recapture the magic in San Francisco?
By Thanksgiving Day 2014, it looked like the NFL had found its coach of the future. Armed with a diabolically clever spread offense and fueled by a sports science program that had somehow made smoothies a topic of national conversation, Chip Kelly’s Philadelphia Eagles smashed the Dallas Cowboys 33–10, all but sealing the 9–3 Eagles’ second straight NFC East title under Kelly. But now, in 2016, the idea that Kelly — whose team made the playoffs just once in three years and who ultimately was fired with a game left in the 2015 season — could personally usher in a football revolution seems almost quaint.
Bill Parcells once said football “is not a game for well-adjusted people,” but Kelly, now the head coach of the San Francisco 49ers, is unusual even by the standards of football coaches. From his puzzling power plays and bizarre roster moves to his odd backstory (it took Kelly 10 years to get his undergraduate degree and most — including Kelly’s biographer — thought he was a lifelong bachelor until The Washington Post discovered last year that Kelly had been married for seven years in the ’90s), Kelly is one of the most enigmatic figures in football. But analysis of Chip Kelly the person and Chip Kelly the general manager has obscured a more straightforward question: What happened to Chip Kelly the offensive guru?
Kelly’s vaunted spread offense incinerated his opponents when he coached at Oregon — including college defenses coached by NFL-pedigreed luminaries like Pete Carroll (613 yards and 47 points), Monte Kiffin (730 yards and 62 points; 599 yards and 53 points), and current Chicago Bears defensive coordinator Vic Fangio (626 yards and 52 points) — and it dazzled the NFL as his 2013 Eagles team finished first in rushing yards, rushing yards per attempt, and yards per play, third in offensive efficiency (per Football Outsiders), and fourth in scoring. But in the two years since Kelly’s offense has gotten progressively worse, bottoming out in 2015 as the Eagles ranked a putrid 26th in efficiency, 23rd in yards per play, and 28th in adjusted yards per pass attempt.
And now Kelly — stripped of any oversight over personnel — is in charge of a 49ers offense that boasts arguably the worst skill-position talent in the NFL and will be led at quarterback by Blaine Gabbert, whose 71.9 career passer rating puts him behind such exalted figures as Geno Smith and Brandon Weeden. While Kelly’s Oregon and early Eagles offenses broke records by weaving together multiple formations, adaptable running schemes, and multifaceted read-options, all powered by an ingenious spread offense philosophy and a frenetic, up-tempo pace, in the past two years those elements have been undermined or simply fallen away, and Kelly’s offense has become, in Evan Mathis’s words, the most “never-evolving, vanilla offense” in the NFL. How did that happen?
The fast-paced no-huddle is fundamental not only to Chip Kelly’s offense, but to Chip Kelly the person. Jon Gruden once remarked that Kelly’s Oregon teams were “as fast as any team that plays football.” Kelly’s Ducks practiced fast, played fast, and were fast. Everything about Kelly was so rapid-fire that he managed to encapsulate his entire coaching philosophy in a single 30-second commercial for UPS, complete with jump cuts and a drum beat.
Chip Kelly and Marcus Mariota (Getty Images)
At least for a season, the story was much the same in the NFL, and Kelly’s methods quickly garnered the NFL’s attention. “They go really fast and try to wear the defense down or force [a] communication issue on defense so … even if you’re aligned right, if you’re not able to get your assignments done quickly [and if] there’s space in there, somebody gets free,” New England Patriots coach Bill Belichick said of Kelly’s offense in December. “The speed that they go at, it’s hard to get much communication in. It forces you to kind of simplify things defensively.”
But defenses adjust, and by the end of 2015, Kelly’s opponents barely seemed affected by his tempo. As Belichick pointed out, the biggest benefit of fast tempo is that it takes play-calling away from defensive coordinators, putting the onus on defensive players to communicate and adjust on the fly. But, as other NFL offenses have increasingly used the no-huddle, defenses have gotten comfortable playing fast themselves, and can now communicate their complex schemes and adjustments with just a word or two. The defenses Kelly’s team faced in 2015 were exponentially more sophisticated than what Kelly faced in 2013, a direct result of defensive coaches and players being better at communicating.
But another element is that while the no-huddle works in the NFL — and Kelly’s 2013 opponents were largely unprepared for Kelly’s pace — it’s not as effective as it is in college football for a very simple reason: The NFL doesn’t permit teams to ever reach the warp speeds Kelly’s Oregon teams typically operated at. While NFL coaches aren’t permitted to openly critique officials or league policy, it’s well understood in coaching circles.
“In the NFL, what they did is the officials stand over the ball until the officials are ready to call the game,” Alabama head coach (and Kelly friend) Nick Saban explained in 2014. “The coach at Philadelphia ran 83 plays a game at Oregon, and runs 65 a game in Philadelphia. … When they went to Philadelphia in the NFL and they were going so fast, the officials said, ‘We control the pace of the game.’ The league said, ‘The officials control the pace of the game, not a coach.’”
So while defenses had to adjust to Kelly’s tempo, they never had to adjust to the tempo Kelly wanted, only what the NFL allowed. But it’s not like the NFL has singled out Kelly, as it applies just as much to Bill Belichick and Tom Brady when they go no-huddle. Good coaching is about adapting. Kelly has failed to adjust.
While Kelly’s Oregon quarterbacks didn’t run as often as people think — Oregon QB Darron Thomas averaged a mere 346 rushing yards per season from 2010 to 2011 — everyone understands that the threat of the QB run is integral to Kelly’s offense. “We run a ton of zone reads,” Kelly said at a coaching clinic in 2011. “[The quarterback] has to read one of the defenders, in effect blocking him. We can block five defenders and read the sixth one.”
Indeed, a major reason Kelly’s offense was so difficult to defend at Oregon was because he would combine a small handful of basic, sound blocking schemes — inside zone, outside zone, and his patented sweep — with a flurry of QB reads of everyone from defensive tackles to linebackers and even safeties. As used by Kelly, the read-option provides an offense with a multitude of advantages: It’s easier to read a defender than it is to block him, the reads become built-in misdirection as the defense doesn’t know who has the ball, and, as Kelly pointed out, a QB who is a threat to run alters the fundamental arithmetic of football.
But in the NFL, the calculus is different. It’s not that different on the field, but it’s the off-field numbers that become more salient, namely the shockingly small number of qualified starting QBs and the exorbitant cap hits the good (and some not-so-good) ones command. In the NFL, repeatedly running your QB may be good X’s and O’s, but it’s bad economics, as losing your franchise QB to injury in exchange for an extra first down is one of the surest ways to lose your coaching job.
Kelly seems to have sensed this. At Oregon he said he wanted “a quarterback who can run and not a running back who can throw”; in the NFL, Kelly seems to have gone out of his way to start immobile QBs, drafting Matt Barkley, signing Mark Sanchez (twice!), and trading for Sam Bradford. Kelly even reportedly refused to offer a tryout to then-free agent (and current Buffalo Bills starter) Tyrod Taylor, who ran for over 2,100 yards at Virginia Tech. And in San Francisco, Colin Kaepernick — at one point the most dangerous dual-threat QB the NFL had ever seen — has been limited this offseason by a variety of injuries. The QB competition between him and Blaine Gabbert never blossomed, with Kelly naming Gabbert his starter after the preseason finale. This isn’t to say Kelly should ask his NFL QBs to tote the ball 15 times a game — the read-option is best used in the NFL to complement a team’s base offense like draws and screen passes — but if Kelly wants to de-emphasize the read-option, then his offense must evolve to counterbalance the loss of a potent tactic.
Instead, Kelly’s answer has been to simply run plays that look like read-options, but without any reads or options. This has not gone well. Defenders who used to stand and watch the QB as the running back ran free now immediately collapse toward the runner to stuff the play.
Kelly once said that the shotgun inside zone “is not a great play if the quarterback hands off to the running back and everyone in the stadium knows who has the ball.” He was right, and his NFL offense is now proof.
The predictability of Kelly’s offense has gone beyond the defense knowing who would get the ball, as defenders frequently now know which play is coming. Kelly, who has long relied on his tempo and the threat of the QB run to keep defenses honest, has done little to hide his offense’s tendencies. Watch Philadelphia’s remarkable 70-yard, four-play (all runs), touchdown drive from 2014, which took a grand total of one minute and 20 seconds off the clock.
A great drive, but the alignment of the tight end and running back gives away the play: If the tight end and running back lined up on opposite sides of the line, Kelly’s team ran a sweep toward the tight end; if they lined up on the same side, it was an inside zone away from the tight end. This giveaway hasn’t always been in Kelly’s offense, but as he phased out read-options he increasingly kept the tight end backside to block the defensive end on inside zone plays. Defensive coaches with experience against spread offenses will tell you that the tight end often gives away the play, and that has certainly become true for Kelly’s offense.
Sam Bradford and Chip Kelly (Getty Images)
The tide truly turned on Kelly’s offense in the Eagles’s 24–14 loss to the Seattle Seahawks in 2014, just one week after Philadelphia’s Thanksgiving Day win over the Cowboys. Seattle stuffed the Eagles offense, holding them to 139 total yards, and after the game Seahawks players were not shy about telling the media they knew what to expect.
“We knew what plays were coming,” Seahawks linebacker Bobby Wagner said after the game. “Their offense is kind of predictable. They have a lot of plays where they can only run one way.”
This wasn’t an isolated incident. After losing to the Cowboys early in the 2015 season — a game in which the Eagles managed only 7 rushing yards — Eagles receiver Josh Huff said Dallas’s players were calling out Kelly’s plays before the snap. Another example came in Week 1 of 2015, as the Atlanta Falcons repeatedly checked into defenses designed to stop whichever play Kelly called. Whenever he called an inside zone — again, with the running back and tight end aligned to the same side — the Falcons, in turn, checked to a defensive stunt designed to blow up that specific play.
Philadelphia’s opponents seemed to know what was coming throughout 2015, even when he tried to mix in other plays. For example, as long as he’s been in the NFL, whenever Kelly’s opponents have geared up to stop his inside zone play, he has typically gone to his counterpunch, a sweep play in which the guard and center both pull to lead the way. But, tipped off by the alignment of the running back and the tight end, defenses were ready for that, too.
It’s one thing for a team to miss a block or for the play caller to guess wrong, but these are abysmal, totally hopeless plays rarely seen in the NFL. Yet Kelly repeatedly deflected criticism that his offense had become predictable by saying that the issue came down to only one thing: “We need to execute.”
Execution was certainly also an issue for Kelly’s offense — what wasn’t? — but it didn’t arise in a vacuum. Kelly’s 2015 opponents were unafraid of his QBs as run threats and could accurately guess his play calls; it’s no surprise they were also able to exploit errors in his team’s execution. “You cannot just fool defenses with tempo,” University of Kentucky offensive line coach John Schlarman said at a coaching clinic, summing up the experience of middling up-tempo spread offenses at every level of football. “There is a difference in a fast playing team playing crisp and a fast playing team playing sloppy.”
Strangely, the predictability and unoriginality of Kelly’s offense is a recent phenomenon. Kelly routinely introduced new wrinkles at Oregon, and, most impressively, he dramatically shifted his offense midway through his first season in Philadelphia. After a 15–7 loss to the Giants in 2013 — a game in which the Eagles mustered a mere 200 total yards and which dropped the Eagles to a disappointing 3–5 record — Kelly marched into the locker room and delivered a message:
“I’ll never forget this in all my years in the NFL,” former Eagles quarterback Michael Vick recalled last year. “He said, ‘We will never look that way on offense the way we looked today, ever again.’”
And, at least for the rest of that 2013 season, Chip was right. The very next week, Kelly’s team bombed the Raiders with 49 points, while QB Nick Foles tied an NFL record with seven touchdown passes. And the offense was off to the races, smashing team records and finishing at or near the top of every major offensive category en route to a 7–1 record to close the season. Kelly did it by adapting, as he increasingly folded in NFL passing concepts brought by his assistants, particularly Pat Shurmur, and found new ways to run the ball from under center. Kelly had created a blend of shotgun spread and pro-style offenses that looked like the future.
Then … nothing. Kelly’s 2015 Eagles offense was essentially unchanged from 2013 (and the 49ers offense this preseason looked identical as well), and what two or three years prior was fresh is now stale and easily defended. If anything, Kelly’s later offenses were more simplistic than his earlier ones, as the creative motions and formations that Kelly once used so well largely vanished.
And it’s not only the running game — Kelly’s pass game has been in stasis since 2013 as well. Though Kelly’s teams have always been run-first affairs — at Oregon he frequently admitted that “we run the ball better than we throw the ball” — to win in the NFL you must be able to throw when the other team gears up to stop the run. And, despite showing the flexibility to experiment in 2013, there has been zero evolution in Kelly’s passing offense since, and, like Kelly’s running game, most defensive coaches can identify what pass play is coming based on how his players align.
One of the most effective plays for Kelly’s offense in 2013 was his “mesh” concept, in which two receivers run quick crossing routes — designed to pick off defenders chasing them — while another receiver curls over the middle and the running back runs a “wheel” route up the sideline. It’s a great play … except when the defense knows it’s coming, something that happened far too often last season.
It’s impossible to win in the NFL if the defense knows the play beforehand. But for Kelly, the problem is amplified because of his tempo: If you stop Kelly’s offense, you also stop his team. While Kelly’s Eagles teams went 24–8 when they rushed for more than 100 yards, they were just 2–13 when they failed to hit the century mark, including 0–7 in 2015. (Kelly’s Oregon teams went 0–3 when rushing for fewer than 100 yards, versus 46–4 when they rushed for more than 100.) In part this is because his passing game cannot carry the load (the 2015 Eagles were fifth worst in the NFL on traditional dropback passes at 5.5 yards per pass), but also because if Kelly’s offense can’t run the ball, his defenses are stuck on the field.
“Chip Kelly is a friend, but I could not run the offense he runs,” Stanford head coach David Shaw said this summer at a coaching clinic. “If you run an up-tempo offense, you better be good at staying on the field. If you cannot get first downs, your defense will play the entire game.” Indeed, the 2015 Eagles defense defended an incredible 1,148 plays, while the team that defended the fewest, the Seahawks, played just 947 snaps. At an NFL average of around 65 plays a game, Kelly’s defense effectively played three more games than Seattle’s.
Albert Einstein once advised his students to “make everything as simple as possible, but no simpler,” and Kelly’s offense, increasingly unable to benefit from either tempo or QB runs, is simply, well, too simple. But he doesn’t need to change his core philosophy and suddenly start using a 700-page playbook. Rather than add a bunch of new schemes, Kelly could better protect the plays he currently runs, by mixing in additional formations, motions, and shifts with his tempo to keep defenses off balance.
Chip Kelly (Getty Images)
Bill Belichick once spoke glowingly about Hall of Fame coach Joe Gibbs’s Washington teams that were, like Kelly’s, built around one-back formations and an elegantly simple running game. “Honestly, they [Gibbs’s Washington teams] only had three plays, running plays,” Belichick explained. “But they had a million different ways to run them: every formation, personnel group, motion, shifting. And it was hard to recognize because it was always different every week. … It’s unbelievable the amount of success they had running, really, running the inside zone, running the outside zone and running the counter [trey]. They won a lot of games doing that.” A little variety would go a long way to helping Kelly’s offense get back on track.
But the question is whether or not Kelly is ready to evolve. As the new coach of the San Francisco 49ers, the man who was at one time football’s leading innovator seeks redemption in the heart of Silicon Valley, America’s current cradle of disruptive innovation, a fitting landing spot given that it appears Kelly is seemingly hurtling toward being the next victim of the “Innovator’s Curse.”
The first idea of the curse is that innovations that can’t be protected frequently don’t benefit the innovator, an issue for Kelly given that one can’t patent football play, and any play that works one week is sure to be used across the league by the next. Indeed, NFL coaches as diverse as Hue Jackson, Pete Carroll, Mike McCarthy, Mike McCoy, Bill O’Brien, Adam Gase, and even Belichick have co-opted Kelly’s ideas, and Kelly’s former quarterbacks coach, current Raiders offensive coordinator Bill Musgrave, said frankly that “the majority of what we’re doing [on offense] is Chip Kelly stuff.” The history of football is in many ways the history of men who watched others win with their ideas.
But the second idea behind the Innovator’s Curse is that, having once innovated, it’s increasingly difficult for the innovator to continue innovating. To use Silicon Valley examples, there are countless IBMs, Xeroxes, and Yahoos: one-time disruptors whose cultures and ideas ossified and who eventually became the disrupted.
If Kelly fails to innovate and evolve, he’ll just be yet another in a long line of football coaches, once considered cutting edge, who themselves were disrupted. But there is some reason for hope. Kelly is a smart coach in a sport where those are in short supply, and, in his first press conference as 49ers head coach, he hinted at introspection when he said he was performing an “autopsy” on what exactly went wrong during his Eagles tenure. But Kelly’s actions since — from his uninspired assistant-coaching hires to his team’s play this preseason — showed nothing that would indicate anything except more of the same, and just Thursday Kelly said the only thing he’s done differently since his time in Philadelphia is “put a lot more sunscreen on.” If Kelly 2.0 fails in San Francisco, it will be a shame for those of us who continue to admire what he did to push the game of football forward, but it certainly won’t be a surprise.