clock menu more-arrow no yes mobile

Filed under:

Duck Tape: Film Analysis Update of Washington 2023

A preview of Oregon’s Pac-12 Championship opponent in Las Vegas

COLLEGE FOOTBALL: NOV 25 Washington State at Washington Photo by Jesse Beals/Icon Sportswire via Getty Images

Special thanks to Roman Tomashoff of Locked On Huskies for joining me on the Quack 12 Podcast to discuss Washington’s roster. LISTEN HERE

Thanks as well to ATQ’s Tristan Holmes for some of the preliminary charting work on UW.

Nota bene: This article is a follow-up to my week 7 film study preview of UW which will update how the team has played in the second half of the season. As it happens, the structure and personnel on both sides of the ball are almost entirely unchanged and so this article won’t repeat that background analysis, but rather document how the tendencies, success rates, and some degree of player availability has changed since then. I recommend starting with (re-)reading that article.


Offense

UW remains a pass-dominated offense, however the extreme split between redzone and non-redzone run-pass balanced observed in the first half of the season has eased up considerably in the second half. The Huskies continue to be a basically balanced offense once they get to the redzone, but they’ve dropped from a 73% pass rate outside the redzone to start the season to a 62% pass rate there over the last six games.

That shift away from such a heavy reliance on passing to advance the ball between the 20s has come as UW has experienced a significant falloff in their passing efficiency, yardage, and explosiveness:

Since the pass offense is the driver of UW’s overall team success, Roman and I spent the lion’s share of the podcast exploring possible reasons why this falloff and UW’s run of close games – every game since week 4 has been decided by 10 points or fewer – might have happened.

Both of us think that the primary cause is a drop in accuracy from starter #9 QB Penix, likely caused by some unknown but persistent physical ailment that he’s had for the last six weeks and which is unlikely to suddenly clear up in time for the conference championship game. A cursory glance at Penix’s statistical gamelog this season shows an obvious performance cliff for the passing game – 75% completion rate and 400 yards per outing through the Arizona game, 59% and 271 yards ever since. From watching film it’s clear that he’s routinely missing passes by a wide margin that he reliably made with pinpoint accuracy before.

Despite that, the offensive scheme remains just as potent and has almost as high a rate of explosiveness, with no major changes to the structure of the offense – simply a dip in execution within it, and in perhaps a concession to that reality, more frequent rushing between the 20s. It’s also quite a puzzle as to which passes are going to be wildly off the mark and which are going to stay accurate, something Roman and I discussed for quite some time without resolving. I’ve run a statistical regression on the question and can report no correlations with a high degree of confidence – it’s either random or there’s not enough data to isolate when exactly the issue is triggered. Pac-12 defenses don’t appear to have solved it either, as none have consistently shut down UW’s passing offense, merely made it one of average efficiency but still potentially lethal explosiveness.

Here’s a representative sample of successful passing plays from the last six weeks:

(Reminder – you can use the button in the lower right corner to control playback speed)

  1. :00 – USC tried to mimic Arizona’s defensive success by bailing both backers and one of the DEs into coverage on most snaps. They’re still getting pressure rushing three because of the o-line’s poor grades in short-setting, but as usual with UW’s system it doesn’t matter because the ball is out in under two seconds, on an automatic first down to #0 WR G. Jackson because the defense didn’t leave anyone at the midlevel.
  2. :08 – Another quick throw with the RT on rollerskates; Utah is in zone but Penix has read it properly after the motion, with the backer and nickel in failing to switch off the back leaking and the dig by #4 WR Bernard crossing over them.
  3. :27 – The entire line collapses this time, but even with four guys in the backfield Penix just makes a couple of quick pocket moves with his eyes downfield and finds #37 TE Westover, his favorite target in danger situations and must-convert downs.
  4. :37 – Instant interior pressure means this prayer has to be out immediately, but even with a tiny window against very tight coverage #1 WR Odunze is good for a couple miracles a game.

Another issue that’s cropped up in recent weeks is a performance hit to the receiver corps. Two of their six wideouts, Jackson and #11 WR McMillan, have been in and out of the lineup with injuries, and have looked somewhat shaky when they’ve made their returns. Additionally, the normally sure-handed #2 WR Polk, while available continuously, has had some struggles catching the ball for the last two games – he’s had ten targets but no catches (I tallied three drops, three pass breakups he lost the fight for, three just uncatchable, and an intercepted ball). Roman reported that Jackson and McMillan will be back in the lineup in the title game so they’ll have the full complement, though it’s unknown at what level some of those guys will be playing at.

As an explanatory factor for the dip in performance, in my opinion the receiver issues have had some effect but it’s minor in comparison to whatever’s going on with Penix. Polk’s issues, if they’re anything other than statistical noise, didn’t start until the last two weeks, and the difference between the guys they were missing and the guys that they’ve had is marginal, especially considering they’ve had a Biletnikoff finalist, a highly reliable TE, and at least two other good receivers on the field at all times. Running out the math on my tally sheet, at most all these receiver issues could have possibly added up to is about 35 yards fewer per game, which isn’t nothing but can’t explain a loss of 10 percentage points in efficiency or a yards-per-game gap four to five times that size.

Here’s a representative sample of failed passing plays from recent weeks:

  1. :00 – I’ve spent a lot of time and computing horsepower trying to differentiate the conditions that produce on-target balls with a three-man rush dumping the LT in his lap from the wildly overthrown balls with a three-man rush dumping the LT in his lap like this one, and have come up empty.
  2. :19 – The interesting thing about Penix’s accuracy problem is that it hasn’t affected his confidence at all, he’s still sticking with the more challenging deep and cross-field throws that require high level arm talent, instead of taking the easier crosser with room to work or the TE on the boundary go who’s breaking open against zone.
  3. :27 – Much was made of the weather conditions in this game but watching film closely showed the rain wasn’t to blame for many of the receiver issues.
  4. :53 – The pocket is fine here, or at least as good as it’s going to get with this line, and Penix had time if he chose to slide over and find a better throw. Instead he forces it to Odunze in coverage, despite having a perfect shot to McMillan open or the checkdown to the back for an easy conversion.

The roughly 10-point dip in pass efficiency is persistent in all seven of the major down & distance categories - 1st & 10 and 2nd / 3rd & short / medium / long. Most significantly, this means that over the last six weeks UW is down to 54% pass efficiency on 1st downs, 45% on 2nd & long, and 37% on 3rd & long.

That’s created a real problem operating from behind the chains that they didn’t have in 2022 or the first half of 2023 when they happily operated in long-yardage situations. (As a reference point, the 3rd & long pass efficiency number that Oregon State has been stuck at for the last three seasons with their ongoing QB issues and attendant drive stalling problem has been 32%.) The result has been a tumble in drive efficiency metrics, with drive-based advanced stats systems dropping UW’s offensive rank to the high 30s / low 40s even as play-based systems have kept them higher.

The final major change to the observations in my week 7 write-up regards blitzing. Due to the nature of UW’s passing system and Penix’s abilities within it, explained in greater depth with extensive documentation in that article, I strongly recommended against blitzing at that time. The data showed it didn’t result in any more effective pass pressure because the ball would get out faster than the rush could get home and extra rushers wouldn’t speed that up, just cost personnel who were better off in coverage.

Over the last six weeks, however, the equation has changed – whereas in the first six weeks UW’s pass efficiency when facing a blitz improved by four percentage points over their then-average to 64%, during the last six weeks their pass efficiency against the blitz now drops by five points below their current average to 44%. Blitzing now appears to exacerbate pre-existing lock-on issues and exploits a narrower range of viable options so that reduced coverage personnel is no longer a significant opportunity cost and the increased pressure is well worth the risk.

Here’s a representative sample of blitz approaches in recent weeks:

  1. :00 – The o-line really had problems dealing with ASU and are using an 8-man protection against the blitz, giving the defense a significant numbers advantage against just two in the pattern – that was the real secret to ASU’s defensive success against UW, holding back receiving options to help with protection against the blitz. Penix forces it into double coverage in what’s either a miss or a throwaway.
  2. :18 – The back has to pick up the blitzing safety and so he can’t help with the linebacker who’s splitting the guards; the blitz communication hasn’t indicated what to do about this loop stunt and so the C comes off even with the LG unoccupied and he gets through and affects the throw.
  3. :32 – Utah’s bringing everybody here, which is excessive – they’ve left a big cushion with no underneath coverage and Penix doesn’t have any difficulty hitting it on the quick throw.
  4. :39 – Wazzu adopted a blitz strategy and UW responded by consistently keeping 7-man protections, and therefore no back or TE for quick outlets against pressure, just the big stuff to the receivers because this offense doesn’t have any other components. This play has two of them going to the same spot in the endzone and Penix forces the ball into triple coverage, nearly getting intercepted.

Rushing makes up a considerably higher percentage of overall playcalls during the last six weeks – about 40% now compared to only 34% then, with virtually all of the change happening outside the redzone since they were already a balanced offense once they crossed the 20 yard line. The other significant change has been shift in the ratio of carries between the three main backs, with starter #7 RB D. Johnson now getting almost 79% of them, up about 15 points, and #8 RB Nixon falling back to almost a tie with the third back #20 RB Rogers. Roman said he expects it to almost entirely be Johnson’s ball with just a handful for Nixon and Rogers going forward.

There has been a slight bump in the statistical performance of the run game for the dataset comprising the last six games, up to 53.5% rushing efficiency, 5.0 adjusted YPC, and 13.25% of designed runs gaining 10+ yards. However, the entirety of these gains come from a single game against USC and their horrendous defense, who promptly fired their DC afterwards. If that game is excluded from the dataset, then UW’s numbers return to precisely the same as they were in the first six weeks – 49% efficiency, 4.9 YPC, and 9.75% explosiveness.

Johnson is a powerful runner and has some of the highest “yaco” rates in the conference on my tally sheet, which is my code for muscle runs through contact which flip what would have been a failed run into a successful one by getting that extra yard or two pushing through the tackler. Some examples:

  1. :00 – The one thing the UW offensive line has going for them is that they’re big, and USC’s front isn’t.
  2. :08 – We haven’t seen it much over the last two weeks, but for much of the season I’ve thought run designs that let Johnson use his cutback ability have really shown off a grant plant and change of direction in the backfield.
  3. :28 – This has been more typical in the final two games, running straight into contact since the C whiffs on his assignment, and muscling that extra bit to be credited with a 1st down.
  4. :43 – The offense has a QB read structure built into it but defenses have mostly ignored the possibility of Penix keeping it, so there’s already one guy in the backfield. The second is because the TE is failing to downblock the end, as usual. The third is because the LG misses every one of his blocks, so the LT lets go of his guy and picks up the backer coming in from depth, but then doesn’t control him very well. So that’s a lot of Cougs Johnson has to carry on his back across the LTG, but he gets it done.

There has been an offensive line change, indeed the history of UW’s changes at guard is now quite involved and required a lengthy recounting on the podcast with Roman. The upshot is that the line has had a different right guard, #77 RG Buelow, for the last several weeks. According to Roman they’ll stick with him going forward, instead of #56 RG G. Hatchett as they went with earlier in the year.

On my tally sheet Buelow is a major downgrade in run-blocking compared to Hatchett – Roman attributes it to his 6’9” size and the fact that he really should be playing at tackle, but he said the staff prefers Buelow’s length for pass-blocking purposes and will stick with him even if both are available. (As such Roman said the offensive line change has the least to do with the dropoff in pass performance and may even have helped it; I’m not nearly as sanguine about Buelow, but regardless I agree that it’s a minor factor at most.)

Roman relayed that many in the fanbase blame “weird” playcalling for the downturn in performance, and would like to see more rushing intermixed in drives that have gone three & out (quite the turnaround on a coordinator they were sure was a lock for the Broyles award in September). I was puzzled by this as there has in fact been a significant increase in rushing, which is probably appropriate given the pass game issues and the fact that they’re pretty good at converting short-yardage situations with the run.

It’s also the case that there are more stalled drives that were run-dominated than pass-dominated over the last six weeks, because the real issue is that both aspects are now equally inefficient, but the rushing offense gives them only half the odds of an explosive play. I’m reminded of Gabey Lucas’ observation about an earlier Husky team that those who complain about playcalling are simply advertising that they don’t actually understand what’s bothering them.

Readers of Addicted to Quack can do better, as from long reporting it’s easy to identify rush tendencies in this offense – on the three-quarters of snaps in which the Huskies are in 11-pers shotgun, it’s a pass 77% of the time; on the remaining plays when it’s anything else, be that multiple TEs or RBs or the QB going under center, it flips to a rush play 62% of the time. Splitting out the first and last six games of the season shows those tendencies have each grown more pronounced by about five percentage points in the back half of the year.

Here’s a representative sample of failed rushing plays:

  1. :00 – Stanford wasn’t winning a lot of blocks but stopping this outside doesn’t require it because there’s no sophisticated misdirection involved, it just requires the OLB widening on the TE motion to control sideline access and the safety coming down to kill the play – there’s no one else to occupy those guys since the other TE is backside on this 12-pers run. This was part of three different 3 & outs in this game.
  2. :07 – The pulling LT on this gap scheme goes into his block uncontrolled and Utah’s backup LB knocks him back into Johnson.
  3. :12 – Here’s the first play of the game, no fatigue issues for anyone or any particular effort to shut down the run from the defense since there’s been no strategic preferences revealed yet. But the QB’s alignment gives the play away and OSU wrecks the line – there isn’t a single block the OL is winning. Johnson does an admirable job escaping the mess but can’t turn the corner.
  4. :22 – This power-blocked run in short yardage with 13-pers has no surprises, so they can’t afford for the RG, RT, and TE to each be one man off in their assignments like this. Two TEs hit the ground and two LBs get free shots on the back, preventing the conversion.

Defense

Changes to the defensive personnel and performance have been more subtle than with the offense over the last six weeks. It continues to be the same structure and philosophy, though inevitably some attrition has taken effect and on balance the offenses that UW has faced have been better than those they had when I last wrote about them, particularly on the passing side.

Earlier in the year I’d written that the most significant changes UW enjoyed compared to last season were in the secondary – they’d gotten healthier, moved a couple of guys around to better positions for them, and added a couple of corners who immediately took over. That had led to shoring up their pass defense success rate to 52%, and reducing their adjusted YPA allowed to 6.2 and explosive pass rate to 15% just before Oregon played them the first time.

However, that progress now seems to have plateaued and possibly slipped a bit. Their pass defense numbers over the last six weeks are a 50.4% success rate, 7.0 adjusted YPA allowed, and 15.5% of opponents’ passes gaining 15+ yards (those numbers are buoyed by a single outlier game against ASU in which the Sun Devils chose 34 called passing plays at disastrous success rates and yardage; if that game is excluded from the dataset then UW has uniform pass defense numbers in both phases - 48% success, 7.2 YPA, and 16.5% explosives allowed).

The mystery of what’s happened to UW’s pass rush continues, and Roman told me he’s stumped by it too – there’s no obvious explanation for why #8 OLB Trice or #4 OLB Tupuola-Fetui should be producing so much less havoc than last year. The team’s sack, scramble, or throwaway per dropback rate on my tally sheet is down to a pedestrian 17% from 32%, and in raw stats they’re producing just 1.58 sacks per game (112th nationally) and 4.5 TFLs per game (116th nationally), with only four Power-5 teams producing fewer.

The pass rush is a real asset when it gets home, and Roman and I spent some time discussing the very interesting #52 DL Tunuufi – he’s got a tweener body and plays situationally both inside and outside, and on my tally sheet is actually UW’s most effective pass rusher due to his high motor. The down & distance situational analysis shows the 3rd & short vs 3rd & long differential in pass defense success rates has grown, from about 20 percentage points in the first six games to 29 points in the last six, indicating that the pass rush has grown even more essential to containing opponents through the air.

Here’s a representative sample of successful pass defenses:

  1. :00 – Watch Tunuufi on the twist, the other DT’s getting nothing done of course but he’s slipped out and has much better speed than the other guys who can play on the interior to get to the QB.
  2. :17 – UW blows up this inside screen to the back by bailing out first, then the linebackers keeping up the pressure and not letting the QB scramble out of it.
  3. :25 – Here’s the all-ILB/OLB package with no DTs that UW uses on 3rd & long situations. It’s not often an all-out blitz, but they do usually employ a twist like this which beats the center. The QB makes a poor decision under pressure.
  4. :47 – Standard down, 11-pers, UW has Tunuufi in as a DT next to one of their big backup DTs, which was their typical lineup for most of this game. He blows past the LG on this screen and is too fast for the QB to get away from before he can get it to the back.

Roman brought up some injury news about the safety unit - #13 DB Fabiculanan has missed some time recently with an injury (I’ve seen him on the field at times throughout the last six weeks but he hasn’t recorded any stats), he has been one of their starters since #20 DB Turner and #28 DB Nunley have been absent for a while with injury, as well as the only one who’d rotate with #3 H Powell. Roman said Turner might finally be able to return for the title game, otherwise we’re likely to see backup #24 DB Esteen and not much available relief for them or the other starter, #7 DB Hampton.

The secondary had a pretty easy go of it against Oregon State – they dominated the Beavs’ short receivers and the weather conditions led OSU to get away from passing. Otherwise, the back half of the season has continued the trends evident at the beginning of the year – aggressive play from corners left on islands that generates a lot of pass interference flags, and the 118th worst rate of 35+ yard pass plays surrendered nationally, with between one and three giant pass plays given up in every game except OSU. Roman and I also discussed some hesitancy and tackling issues in the secondary that allows short passes to easily convert with defenders constantly playing off.

Here’s a representative sample of failed pass defenses:

  1. :00 – I’m willing to bet this play caught Oregon’s attention on watching tape, the spacing concept from the TE in motion to the boundary against UW’s zone defensive structure means even with four over three from the OLB bailing he’ll be open on the sideline.
  2. :14 – UW shows blitz but bails out of it like this on about 36% of passing downs, both OLBs are bailing into coverage but not quickly enough to actually cover the hot route right in the middle behind the “blitz.”
  3. :26 – Here’s an actual zone blitz but there’s an assignment error, both the LB and CB pursue the TE into the flat and no one covers OSU’s only viable WR threat on the play, and then the secondary’s problems tackling all night continued.
  4. :39 – Man coverage has been quite an adventure, especially with their safety issues as QBs have found it pretty easy to look off the single-high and then attack the CB on an island.

Rush defense continues to be quite poor, with no real change in any of the overall numbers, for all the same reasons as I wrote about in my previous article. Over the last six weeks they’ve had a 34% rush defense success rate, allowing 5.3 adjusted YPC, with 17% gaining 10+ yards (that explosiveness figure is a big jump since the first six weeks when it was about 12%, however three points of that five comes just from the outlier USC game when the Trojans ran all over them).

There is something of a change in the situational numbers. UW has gotten about eight percentage points worse at defending the run on 1st & 10, from 45% in the first six games to 37% in the last six. However, they’ve gotten almost 20 points better at stopping 3rd & short rush conversions compared to the first six games … but that’s an improvement from 7% to 26.5%, so still pretty bad. Poor rush defense, particularly in short yardage, has translated to UW being 89th nationally in preventing 3rd down conversions and 115th in preventing redzone touchdowns in raw stats.

They’ve had some personnel changes at linebacker, which in my opinion has had a silver lining despite the unfortunate cause. An injury to starter #11 LB Tuputala got #42 LB Bruener a lot more playing time, and I’ve been writing for three straight years that I think he’s the most technically sound backer UW has on the roster and it baffled me why two different staffs had benched him (behind a transfer from USC, #10 LB Goforth, no less). Roman told me that he expects Bruener to get the start again in the title game.

Here’s a representative sample of successful rush defenses:

  1. :00 – Good flow to the play by UW’s best two LBs here, Bruener and #5 ILB Ulofoshio.
  2. :07 – Here’s the speed package with Tunuufi and the true freshman #48 OLB Lane on the inside on a passing down when Utah attempts a surprise run. He’s beating a fellow true freshman LT showing poor technique and Goforth cleans up with proper patience.
  3. :15 – Finally UW put their best couple of front players on the field, Bruener at depth and #91 DT Letuligasenoa on the line, and got immediately superior results.
  4. :32 – Late in the game and we’re finally seeing #68 DT Ale instead of one of the backup big DTs, and again the performance level is significantly better. Tunuufi is fighting off the double team and the back just runs into him because he’s running away from Ale.

I’ve been trying, without much success, to resolve what’s going on with UW’s defensive tackles. I think it’s very clear, and Roman agrees, that their best is Letuligasenoa. He’s been struggling with a calf injury and we haven’t seen him much in recent weeks, coming in only sparingly though playing well when he does. There’s been a similar pattern, though no word if it’s injury-related or not, with #68 DT Ale – I think he’s better than the guys behind him, but I don’t see him very often and usually not until late in games.

Instead who I’ve been primarily seeing are #55 DT Bandes and #99 DT Tuitele, who are both over 300 lbs and former 4-stars, but have always graded out very ineffectively on my tally sheet and that’s continued over the last six weeks. Roman had some polite and reserved things to say about them.

The other players I’ve seen on the interior of UW’s 2-4-5 defense in recent weeks have been the tweener Tunuufi on standard downs, and Lane as part of a speed package on passing downs. I’ve only seen a tiny handful of snaps for #94 DT J. Parker (he’s spent a couple of orders of magnitude less time on my screen than UW commentators taken up talking to me about him over the years).

Roman relayed on the podcast that the coaches stated Letuligasenoa and Ale are fully healthy and will be totally available for the title game, indeed that they were last week against Wazzu as well but they preferred the speed package to help contain their scrambling quarterback and that explains their absence.

I’m sure that Roman reported that faithfully and he’s doing his job properly, but I don’t buy what the coaches are selling. If they didn’t want big guys in they wouldn’t have played Bandes and Tuitele at all; playing them but not the far superior putatively healthy starters doesn’t make any sense. Putting the starters in at the very end of the game when it was all on the line (Letuligasenoa got the game-winning holding flag) indicates that they were more likely to be on the same pitch count that they were the previous week. If that’s true for the last several weeks then I tend to think it will continue to be true, and we’ll keep seeing what we have been: primarily the inferior backups, plus ends thrown in situationally, with a smattering of the starters as they can afford to let them play.

Here’s a representative sample of failed rush defenses:

  1. :00 – Another play from Coach Dillingham that exploits UW’s zone structure, they leave Powell to the field after motioning the H-back in to create an unbalanced to the boundary rather than realign the defense. The toss gets a phalanx of blockers with a WR taking care of the OLB jumping inside.
  2. :14 – One of the few plays I’ve seen Parker in on, though I’m not sure where he’s going. Ulofoshio is gently tapped out of the play, and the safety takes a bad angle on the run into open grass with the defense aligned over trips to the boundary.
  3. :30 – Here are the UW defensive front backups just getting creamed by OSU’s offensive line, and then Powell bounces off the back.
  4. :40 – This was Wazzu’s third 10+ yard run, which was peculiar because they’re the worst rushing team in the league. It’s Bandes and Tuitele in on the interior getting cleared out easily, an ineffective attempt to squeeze down the puller from Goforth, I have no idea what Tuputala thinks he’s doing since the OLB is already covering the backside unblocked, and then Hampton gets run over for ten yards through contact while his teammates just watch.