Duck Tape: Film Study of OL Coach Adrian Klemm

New Oregon OL coach Klemm has been coaching offensive linemen since 2009. His first three seasons as an on-field coach were with SMU, then he spent five years at UCLA when Jim Mora took over there in 2012. Due to recruiting violations, Klemm was suspended for the first two games of the 2015 season, his second-to-last in Los Angeles, and in September of 2016 the NCAA issued a two-year show-cause order that would have made it difficult to find college coaching employment. He finished coaching that season, and was dismissed in January of 2017 (Mora lasted one more season before UCLA cleaned out that entire staff for Chip Kelly’s arrival for the 2018 season). As far as I can tell, Klemm didn’t work as a coach for the 2017 and 2018 seasons, then was hired by the Steelers as an assistant OL coach for 2019 and 2020, then promoted to OL coach for 2021.

Oregon fans have become accustomed to – or one might say, taken for granted – the excellent offensive line play the Ducks have enjoyed for the past 20 years under OL coaches Steve Greatwood and Alex Mirabal. I consider those two to be part of a very small fraternity of high quality o-line developers in the eleven-year history of the Pac-12 (the other three being Mike Bloomgren, Stanford 2011-17; Chris Strausser, UW 2014-16; and Jim Michalczik, Cal 2011-13, Arizona 2014-17, OSU 2018-present). Mostly, offensive line play in this conference is well below the standard set in the other Power leagues, and in my opinion that’s the primary reason its teams have struggled in big out-of-conference and bowl games. Klemm has big shoes to fill, figuratively speaking, to continue the o-line tradition in Eugene that’s separated Oregon from the rest of the Pac-12 with the conference’s best winning percentage for the last two decades.

For this article, I reviewed the film of UCLA’s 2015 and 2016 seasons — excluding the two games Klemm didn’t coach — which are his most recent college offensive line performances, entirely with players whom he recruited and developed. Twelve different linemen will appear in the video clips below, reflecting some significant injury problems and personnel turnover between seasons. Here’s how they were recruited, and where they are today:

I’ve never seen a resume like this. It is astonishing that eight of the 12 are still playing in the NFL as of the most recent season (and none for Pittsburgh, in case the reader is concerned about favoritism), since putting 75% of one’s active roster in the league over a two-year span would make Klemm the most successful position coach by that metric I’ve ever studied. And it’s not the blue-chips who’d ostensibly be sure-fire picks, either – it’s mostly 3-stars and even a 2-star who were drafted or signed.

It is also astonishing that two of these linemen – and one more lineman I didn’t see on the field, plus a fourth player – have filed lawsuits alleging injury mistreatment and abusive behavior, including neglect of concussion protocols. (Content warning: the allegations discussed in the following links are upsetting.) The best write-up I’ve found of these cases is here, with an update when Mora was hired last December at UConn, and the most recent news I can find regarding their progress.

I’m not equipped to evaluate the claims in those lawsuits from film study, or attempt to connect on-field performance to any injury issues. I’m also not equipped to say why so many of these players have had multi-year pro careers, since I wouldn’t have guessed it from the film I reviewed – that showed fairly standard Pac-12 offensive line play, some good stuff but a lot of inconsistency and frequently poor technique.

Here are the grades for UCLA’s o-linemen in 2015-16 on my tally sheet. In my charting system, a per-play error rate over a season’s worth of data measured in the single digits is pretty good (Penei Sewell had a 4% career error rate, for example), the low teens is mediocre, and the high teens is bad. Anything over 20% is discouraging.

A few observations just from the numbers: first, run blocking was pretty consistently poor, but pass blocking was actually pretty decent. And while it’s usually the case that run-blocking grades out worse for most o-linemen (actively opening holes is harder than dropping into protection), disparities this wide between run and pass are remarkable. Second, for the players who continued from 2015 to 2016, their grades improve significantly. Third, James is something of a standout, but it should be noted that he came in as a backup tackle in 2016 for the last seven games when Miller was injured, and it was clear to me that he was very young when forced into action and that he’s naturally built as a guard instead (indeed he switched to center in the pros). For a complete, color-coded chart that correlates all the above recruitment, NFL, and performance data plus additional information mapping out when exactly each lineman played, see here.

It seems to me that the best fit for that data is that Klemm has a good eye in the high school recruiting process for future pro potential in frame and aptitude — including quite a few that escaped the scouting services — but I remain skeptical of his ability to teach that talent himself, and concerned that there were so many injury issues surrounding this team.

The last piece of relevant context before we get to video review is that the offensive system changed significantly over the course of these two years, arguably twice. The OC for Mora’s first three seasons, Noel Mazzone, left after 2015 for Texas A&M (some UCLA fans believe Mora fired him for scoring too fast and disadvantaging the defense, something I find inane even by Bruins’ standards). He was replaced by Kennedy Polamalu in 2016, who implemented a staggeringly unproductive pro-style offense which was abandoned halfway through the year to resume a kludged spread offense, and was promptly fired at the end of his first season … and probably not soon enough.

To make matters worse, UCLA lost a generational running back, Paul Perkins, at the end of 2015, and he was the only back in that year capable of salvaging otherwise poor run-blocking into significant gains. So they lost three lineman, their most gifted back, and an acceptable run scheme for 2016 – it should have been expected that their rushing performance suffered, and indeed it did, falling to #127 out of 128 in FBS in both rushing yards per game and yards per carry (Oregon, also 4-8 that season, ranked #27 and #15 respectively). And then after six games their starting QB, Josh Rosen, got hurt and was replaced by a QB I genuinely did not remember at all, whose career NCAA passer rating was a poor 112.6, compared to Rosen’s pretty good 140.1.

All of these reasons are why, on the advice of fellow film reviewer Chris Osgood of Bruin Report Online, I went back and watched UCLA’s 8-5 season in 2015 in addition to just the miserable 2016 season as I’d originally planned, and why this article is a week later than expected. His comments on UCLA’s 2016 coaching staff are not reproducible in a family publication, but they were welcome and useful all the same.

Let’s start with what UCLA was trying to accomplish in the run game. In 2015 under Mazzone, we’re seeing a lot of 11-personnel spread formations with zone blocking, relatively uncomplicated and familiar to West coast football fans for the last couple decades. Then in 2016 under Polamalu, the first half of the year it’s trying to recreate Stanford’s glory days (or perhaps Polamalu’s at USC) with very compact formations and frequently three-TE sets, a fullback, and an attempt at complex power blocking. That goes sideways, predictably (the OL issues aside, the TE and FB blocking were even more problematic), and in the back half of the season they practically dug Mazzone’s playbooks out of a closet and we get back to the spread and a lot of RPOs. To illustrate:

(Reminder - after pressing play, you can use the left button to slow any video to 1⁄4 or 1⁄2 speed)

1. :00 – This was the Bruins’ most effective type of zone run, simply washing down the entire line where the big bodies can just lean on the defense. It minimizes the impact of a few issues, like the RG losing control and the LG getting his hat on the wrong side of the second-level block, and of course the TE misses his block on the backer so the run doesn’t get any more than a 1^st down, but as long as the LT wins the gap is going to be there.
2. :06 – Good zone blocking here from the left side of the line, and the LT wins his by running over the defender, so it’s a big gap to run through. The way the defense is configured, with no safety behind the backer since he’s going after the second back in motion, this would have been a touchdown if the LG got in a better second-level block.
3. :12 – First game of 2016, under-center in 12-personnel. The line is all going one way and winning their blocks, with the H-back slicing under to kick out the DE. The offense gets what it wants from the backer who chases the other TE on the fake RPO (there are a lot of fake RPOs this season). The Z-receiver just needs to pick which DB to block, probably should have taken the CB instead, but in a microcosm of this entire season, runs into the ballcarrier and the back has to do the work himself.
4. :33 – Back to an 11-personnel spread for week 8, and this is some of the best zone blocking all season. Good combo then move up by the center with the LG successfully taking over the block entirely, the backup RT gets his hat placement right so that even with losing his feet he creates enough of a gap initially, and the RG escorts his defender out of the way.

The abysmal run performance at the start of the 2016 season aside, the main issue I was seeing over these two years was what typically plagues Pac-12 offenses – poor development of the offensive line strength, technique, and footwork. The following aren’t representative, of course, since even with an unacceptable error rate of 20% for instance, a lineman is still doing just fine four-fifths of the time. But they are illustrative of the typical issues I’d see over and over again when there were run problems. Some examples:

1. :00 – It might seem like the failed pull by the RG is the biggest problem here, but not really, he makes enough contact with the ILB so that the back can step through. The real problems are the LT’s and C’s techniques. The LT is lunging at the DE with his hands first, before getting his feet in place so that his weight is over his feet and he can block with power. The C simply loses in a “low man wins” contest with the nose – he’s forced high and that lets the NT throw him to the side.
2. :08 – A pin & pull to the boundary just requires way more technically sound linemen than this. Good work by the LT and the C recovers well, but the RG is way too high and his hat is on the wrong side, and he gets worked so far into the backfield that it interrupts the LG’s pull and blows up the play.
3. :16 – This was the most common type of run performance out of these offset-Is – very good blocks by the LG and C, but the RT takes a bad first step and loses inside to a 7-tech (!), and the RG’s footwork at the second level is so unsound he gets knocked to the turf by a Wazzu LB (!!). Also note the basically useless blocks by the TE and FB.
4. :24 – Pretty bad whiff by the RT against the OLB, and the center is lunging into his hit on the overhang backer which lets him both get outside contain and then spin around to pursue inside. Good job by the LT, the LG loses his guy but it doesn’t really matter, and the RG driving back a DT of this size is pretty impressive.

More concerning, I was seeing quite a few assignment errors in the run game – not just failing to block properly, but being in the wrong place or engaging the wrong defender (or no one at all). The complexity of run blocking schemes are why o-linemen are typically the smartest members of the squad, and I’m sure Klemm’s unit was made up of bright guys, but there sure were a lot of boneheaded plays like these:

1. :00 – This play has been an Oregon staple for as long as I can remember, and it was tough for me to watch it blocked incorrectly so often. The defensive strength is aligned to the field (three over two) so this should be a run to the boundary, through the offense’s right A-gap. The right side of the line and the LT have it right, and the LG and C start out with the correct combo on the DT, but it’s the LG who’s supposed to take over the block while the C moves up to hit the ILB on the offense’s right (the back is supposed to accelerate away from the unblocked fieldside ILB). Instead they get it backwards, with the LG going up, and to the wrong ILB, and he still misses him, and the third field defender gets a shot in too.
2. :14 – Again this is a really basic zone run. The defense has three down linemen, so it’s an LT-LG combo and a RG-C combo to start out with, and then the LG and C are supposed to move up to take on the backers, while the LT and RG take over the DL blocks fully. Instead, the LG turns his body and is late to get off his combo so his backer runs past him (and he gets flagged for a hold when he belatedly realizes it). The RG doesn’t combo at all much less take over the block, so that DL gets a free shot at the back, and he trips the C who’s moved up to his backer.
3. :22 – One of many horribly executed I-formation runs. The RG and C don’t know which defender is their man, and given that the FB hits the OLB and so this is presumably outside power to the right, I have no idea what the left side of the line is doing.
4. :29 – The LT gets beat inside, which happened often but isn’t really an assignment error. Same thing with the C losing his guy by getting high and not readjusting his hand placement to maintain control – that’s a technique deficiency. The worrying thing here is how the LG handles the stunt by the ILB. This is zone, not man, so when the ILB follows the unblocked DE to the backside of the play, he’s supposed to let him go and move up to hit a safety or help the right side of the line. Chasing the ILB around is useless and makes the LT’s problem worse.

By contrast, the pass blocking was fairly effective and almost every lineman graded out well on my tally sheet, some very well. There are some technique issues we’ll look at below and I especially dislike the tackles short-setting, but most defenses didn’t have Myles Garrett and Daeshon Hall (like A&M did in their 2016 opener) who could exploit them. In addition to setting up pretty decent pockets for deep shots or getting late in the progression for intermediate routes over the middle, this line was particularly good at RPO blocking and rolling the pocket, which require some special practice. Some examples:

1. :00 – The RT is risking a holding flag for a takedown and the LT is playing with his helmet off, but those aside this is a nice handoff of the twist by the LG and LT. It’s also good re-anchoring by all the pass blockers, keeping hand contact while dropping hips and keeping the feet positioned properly to strike again with power.
2. :10 – Very well executed RPO here, at the most successful general category of playtype UCLA recorded on my tally sheet. The RG’s hat is on the wrong side if this would have been a stretch run, but other than that the LT is bullying his guy, the RT actually has his sealed properly, there’s a nice legal cut, and the center’s pull sells the run while also not getting 3 yards past the LOS to avoid an IDP flag (he even turns and checks that it’s a throw).
3. :19 – Nice job across the board here, against a very good pass-rushing defense. The RT escorts his guy around allowing the QB to step up and the left side of the line is dealing with the T-E stunt perfectly, with the RG handling the 1-tech so well the C is free to help the LG with that stunting end. Rosen airmails the ball for no good reason but that’s not the line’s fault.
4. :34 – Rollout passes, either naked or moving with the line like this one, were more than 15 percentage points more successful on a per-play basis than standard dropback pass-pro across the two years I studied. It takes some agility to move laterally like this while keeping shoulders square to the line of scrimmage to maintain blocks – pay attention to the footwork on the replay.

But when pass protection did break down, my stars it was catastrophic. I saw every goof typical to Pac-12 line play – lunging, imbalanced weight, improper hand placement, playing high, stiff knees, you name it. Some examples:

1. :00 – The tackles are doing well here, but all three of the interior guards are playing way too stiff and their defenders get past them without much effort or advanced d-line moves. The LG needs a better base to keep moving, the C doesn’t have his hands on the DT’s chestplate, and the RG is bent at the waist so he gets rocked back and can’t re-anchor when his guy swipes him.
2. :09 – The LG is lunging at this inside move instead of keeping his feet moving so he can absorb the hit squarely, that turns him and takes the C away from any possible help with the stunt. The LT is doing the best he can with that late shove, but the real problem is that the QB has nowhere to move because of how bad the RT is beat. Even before contact, the RT is turned 90 degrees and is trying to drop while perpendicular to the LOS, which I see constantly from these tackles in pass-pro.
3. :30 – The RT here is showing all three kinds of common poor technique: poor footwork in not establishing a proper base, lunging into contact to lose weight balance, and not maintaining hand contact (I saw this wild hand flailing all the time from these tackles). Again the LT is handling the stunt as well as he can, but the LG needs to take over this block so he can do that but he’s got his left leg planted incorrectly and so that guy just runs past him.
4. :37 – The C is leaning to the side instead of dropping quickly and hitting the rusher squarely, the RG has three different guys he could hit and gets none of them, and – I can’t reiterate this enough – one should not be put on roller skates by a Wazzu d-lineman.

The really problematic thing in pass protection was handling blitzes and stunts, which requires a lot of communication between linemen that I just wasn’t seeing much of. It was particularly painful to watch both years’ games against Arizona St, in which Todd Graham’s Sun Devils just blitzed UCLA into submission on nearly every play and the Bruins never developed an effective response. I could have filled this section five times over with just clips from those ASU games, but it certainly wasn’t limited to them:

1. :00 – When the DE makes an inside move, the RT shouldn’t be turning and chasing him like this. He should be dropping and swiveling his head, looking for that OLB possibly coming off the H-back, or, as actually happens, the DT looping around to his side. (The LT isn’t getting a full drop here and instead turning 90 degrees as I mentioned earlier, but that’s technique and he doesn’t pay a price for it here anyway.)
2. :07 – It was pretty clear by this point in the game that ASU was going to bring all six here, and this pickup strategy is baffling. The defense has two to the right of the nose and three to the left (from the offense’s perspective), so why is the o-line slanting to its right? The LT taking the DE means they’ve got four blocking three in the middle while the back somehow has to block two blitzers and he can’t even handle one. Meanwhile both the RG and RT get embarrassed.
3. :13 – The LG not being able to decide which blitzer to hit is a problem, but the real issue here is the C lacking awareness of the blitz entirely. There’s only one guy going to the offense’s right, leave him for the RG and help the LG with the inside twist.
4. :21 – Here’s an unbalanced formation with all four potential targets going out. I don’t understand the strategy here for the LT. The LG doesn’t need help since no one could be coming up the middle (except possibly the ILB but if so then the crosser as hot route is built into the play). But with the back running to the sideline there’s no one to protect the edge – who is supposed to pick up that backer who’s clearly showing blitz?