NCAAF College Football News, Photos, Stats, Scores, Schedule & Videos...

Special thanks to Cody Stovall of Locked On Oklahoma State for joining me to discuss the Cowboys’ roster on this week’s podcast:

For Oregon’s home-and-home series with Oklahoma State, I’ve charted the Cowboys’ last two seasons, had two great podcast conversations with Cody, and constructed the roster databases for both the Pokes and the North Texas Mean Green — whose coaching staff took over in Stillwater this season — and checked out their 2025 tape as well.

As Cody remarked on the podcast, the fate of the Cowboys deserves its own case study – as recently as 2022 they were in the conference championship game, and nearly every season up until that point the path to the Big-XII’s playoff bid ran through Stillwater. And yet nearly every managerial decision the program has made since then has turned out to be the wrong one, with each attempt to course correct just making the situation worse.

Last Saturday’s loss to Tulsa — the second in as many years for the Pokes — exposed what I believe to be a fundamental misunderstanding that new head coach Morris had about what made his North Texas team successful, what was valuable to bring with him to Stillwater and what needed to be discarded. As such, it represented only the latest in this recent history of calamatously bad decisions for this program.

As Cody and I discussed extensively, UNT’s strong 12-2 record last year did have some elements which could probably be imported, with the starting QB as only the most prominent example. But it didn’t have an overwhelming talent edge over its competitors at any position, its advantages were a senior-laden offensive line who dominated the opponents, a horizontal stretch scheme that American conference defenses lacked the athleticism to handle, and a risk-taking staff that didn’t lose its nerve in multiple overtime situations.

Conversely, despite a losing record in the last two seasons, the Oklahoma State roster had quite a bit of talent on it, but the previous staff, and then the interim staff in 2025 after the head coach was fired early, was frequently playing the wrong guys. But late in the 2025 season, the offensive line started to gel and the staff figured out the right personnel groups, with much closer margins. If they hadn’t lost their starting QB to injury in the first quarter of the opener, they’d likely have finished with several wins under their belts.

So when Coach Morris cleared out nearly the entirety of 2025 roster to replace it with lower talent transfers, most crucially every single offensive lineman, and preferentially treated players with North Texas ties, he was setting the team up for a massive gamble. Those players needed to be, in fact, spectacularly better than their talent ratings and cohere instantly. It’s a gamble that hasn’t paid off.

Offense

Starter #17 QB Mestemaker transferred from North Texas, where he maintained an excellent 168.7 NCAA passer rating despite a volume passing offense at 463 attempts. His delivery is polished and smooth, and he’s a the best part of the offense. Even with the pressure difficulties on Saturday, he was still completing passes at a high rate and for the most part avoiding sacks and keeping the ball out of danger – the only thing that suffered for the pressure was the ability to push the ball downfield.

The offense is heavily predicated on horizontal stretch plays, with more than 19% of all of UNT’s meaningful plays last year being throws behind the line of scrimmage. More than three quarters of UNT’s explosive passing plays started with air yards of 8 or fewer and featured the receiver breaking a tackle after the catch.

As Cody and I discussed, this was the primary thing that was just missing on Saturday – the Pokes kept going back to these plays, only to find they couldn’t make them happen like the Mean Green could against American conference opponents. The most puzzling to me is that there’s a big logjam at the short receiver spots — Cody had to correct me on who the starting slot receiver is — and it seems Mestemaker strongly prefers throwing to his right. There are multiple guys vying for the slot position and Z position, and I’m hard-pressed to say who the X-receiver is.

However, there were elements of the passing offense other than horizontal stretch which did work, because I think the elements here show talent and Mestemaker distributes the ball well … if they could reorganize the WR corps (Cody described it as a lack of creativity, it’s not exactly how I’d describe it but I think I know what he meant). Here are some examples of plays which worked and could be expanded on:

Wide Receivers

The receiver with whom it seemed obvious to me that Mestemaker had the strongest chemistry with was the UNT transfer #10 WR W. Young — there are three UNT transfers in the five guys they were throwing the ball to, but Young had by far the best per-target numbers with a 50% success rate and 9.33 adjusted YPT. The other four receivers had miserable per target figures, all below 5 YPT, with a couple under 40% success rate.

Young was also by far the most targeted WR at UNT and had comparable numbers there (a senior had a better success rate but 20 fewer targets, and has run out of eligibility). Here’s a representative sample of how Young was used at UNT:

Here’s how things were going wrong on Saturday, I think it should be obvious what the differences are:

Tight Ends

Curiously, even though this offense used a receiving TE extensively at North Texas, they only employed a blocker on Saturday, sometimes two. I agreed with Cody’s assessment that they were not very effective, and I shared his puzzlement that they beat out others in camp. Cody told me that there’s some hope that #23 TE Hammond (not to be confused with a defensive back of a similar name) who had by coincidence transferred in from North Texas in a previous cycle but hasn’t been healthy enough to play, would be on the mend soon – Cody thinks he’s the best bet for a productive TE.

If Oklahoma State reorganizes its receiving options, I think there’s one standout possibility – Illinois transfer #0 WR Bowick. He wasn’t very well utilized at Illinois, with rather mediocre numbers between the 20s, but he was excellent in the redzone. Cody’s comment was that OK State was in the redzone so rarely that they didn’t get to see him (that’s not strictly speaking true, they Pokes got to the redzone and then lost out on goalline stands because they ran the ball and got stoned in short yardage repeatedly – they elected not to throw to Bowick. But Cody’s point that we don’t really know what Bowick can do in this offense is valid). I also think, if they don’t have a viable pass catching TE, Bowick could be a servicable Y-receiver at 6’5”. Here are some examples of him at Illinois:

Running Backs

Coach Morris and his staff was in for another unfortunate surprise on Saturday when the player through whom the lion’s share of the UNT offense ran, #24 RB Hawkins as a ballcarrier and swing pass receiver, was completely ineffective against Tulsa. In Denton, Hawkins had a staggering 175 carries and 36 passing targets prior to garbage time, with over 6 YPC and 8 YPT. Against Tulsa he got a comparable share of the load in a similarly styled offense without much in the way of halftime adjustments, and got 4.6 YPC and 3.2 YPT, while his success rate fell from 65% last year to 35% on Saturday.

The offensive line is a big culprit for most of the offensive issues – Mestemaker has no time to throw, and none of the backs trying to run between the tackles have a hole to hit, or as Cody noted, the vision or patience to cut out. Hawkins is affected by this of course, but more than anyone else his runs and swing pass targets are to the outside – his production at UNT was more predicated on WR blocks and his own ability to break tackles. Against Tulsa, the factor that kept Hawkins most in check was that those perimeter blocks just weren’t working and he wasn’t able to shake Tulsa tacklers.

Let’s first look at how Hawkins was used last year for background as to why he got 26 meaningful touches against Tulsa, fully a third of the entire team’s:

And now the rushing offense against Tulsa:

Offensive Line

Due to early injuries to a couple of starters, Oklahoma State played eight offensive linemen. One of their starters graduated, but the four other starters plus the sixth man who played most of the season, in addition to the two original starters who were hurt early — for a total of seven experienced offensive linemen — had eligibility remaining at the end of 2025.

Six of them exited through the transfer portal, and the seventh hasn’t played a snap in 2026. It appears that Coach Morris cleared out the entirety of the offensive line room who’d played before with the exception of #75 OG Sanders, and brought in nine transfer linemen to play. Seven of them played on Saturday. Cody told me that an eigthth would have played, #67 OT Nelson, but he’s hurt. They lost another to an ACL tear, #79 LT Sexton (truly unfortunate, the same thing happened to Sexton last year).

Nearly a decade of transfer portal data and modeling still had not encountered this scenario before, I only had the Gaussian curve to predict how bad a line composed of this many transfers would perform. It turned out to be dead on, every lineman from North Texas graded out against Tulsa at exponentially worse than when I charted him the year prior, including #56 C Mercer who was at UNT in 2024, then left for Kansas and reuinted with Mestemaker, only to have multiple bad snaps and communications issues.

There’s a three-man rotation at the two tackle spots but none of them can block at all. #77 RT Lepo lost his starting job at Michigan State after 2024 and transferred out for a reason. #70 OT Torgeson played for Portland State. Somehow Torgeson is more effective than #76 LT Magiya.

Cody and I discussed some ways of fixing this offensive line — putting in Sanders for Mercer, getting Nelson back, benching Magiya and ending the tackle rotations — but they’re hamstrung by the transfer portal effect. They have no viable options because of the elimination of nearly all homegrown lineman. This was a catastrophic error in judgment.

Defense

The scheme has fully switched away from the odd surface it had used for many years and which previous DC Grantham had adapted; now it’s a 4-2-5 with a fist-down and a stand-up end.

There’s a potentially important caveat to the data gathered from Saturday’s game against Tulsa. It seemed to me as though OK State was taking Tulsa’s run game very seriously, and didn’t think much of its passing game at all. I understood that to some extent while reviewing the tape, because Tulsa’s QB kept making unforced errors in high stakes situations. Cody gave me the other piece of the puzzle when we talked, by relaying that the Tulsa head coach had talked up their RB room, and that they were looking forward to taking advantage of what was a poor UNT efficiency rush defense in 2025.

In looking at OK State’s new personnel, I think they have adequate depth and appropriate size and agility in the front to stop the run – good DT depth and size, very big edge-setting DEs (almost too big, Cody and I agreed the pass rush suffered), and a LB trio with good sideline-to-sideline speed. But, as I tried explaining on the podcast, Tulsa’s run game was so unusual that I don’t think OK State got a real shakedown.

By that I mean, it was highly predictable which type of run Tulsa would use in different situations, and were easy pickings. Every single one of their dozen 1^st down runs was to the outside, none between the tackles (OK State had a 50/50 success rate stopping those); meanwhile, every time they ran on 2^nd or 3^rd & long they went straight inside, and got stuffed. There’s just one mixed situation, three rushes on 2^nd & short/medium, which Tulsa won every time but with a variety of things.

In data science, when sample clusters are that compartmentalized we consider it non-robust – not great for making inferences because there are big empty patches. On the podcast I said I’d give the rush defense an incomplete grade, though so far it’s an improvement on UNT’s rush numbers where were abysmally inefficient and frequently gave up super-explosives … but it could be Tulsa wasn’t really trying, Cody implied it might have been a rope-a-dope from the Tulsa’s coach, because they got a lot done with explosive passing on 2^nd downs.

Defensive Tackles

I’ve got no complaints here, they’re big space fillers instead of penetrators, and a couple of them come over from 3-down systems. Cody seemed disappointed that they weren’t wrecking the run but they limited Tulsa to 4 YPC and no explosives except one at the very end. Depth is good.

Defensive Ends

Cody and I agreed that there wasn’t much of a pass rush against Tulsa, but we’re both stumped on the big question – was that because they thought they needed to put in all the edge-setters against the run, or because they don’t actually have, or believe they have, viable pass rushers?

I’ve seen guys like #0 DE Charles be effective before, and on his stat sheet and measurables I think #19 DE Jackson should be a pass rusher they’d go to. I’m inclined to think that #11 DE Huddleston is getting all the reps because he’s a UNT guy, and that’s that with this staff.

I suspect Coach Morris’ staff has drunk their own Kool-Aid on their defensive effectiveness and is unaware of how much turnover luck they had last year, and are betting that Huddleston has magic powers to force and/or fall on as many fumbles as he did in 2025. UNT’s defense got 28 total turnovers last year, nearly the top in the nation, behind just Indiana and Texas Tech. And their recovery luck was astronomical – of the dozen teams which forced more than 20 fumbles, they were the only one who recovered more than half … and significantly more, 59.1%, while the cumulative average for the rest was 94 of 236 or 39.8%.

Linebackers

This group looked pretty athletic to me. I saw both of the transfers’ tape multiple times for other projects and like their lateral agility, usually something that’s lacking in Big Ten backers so I found that to be refreshing.

By far UNT’s biggest issue last year was giving up super-explosive plays. They were 105^th worst nationally in 30+ yard plays surrendered, averaging two a game prior to garbage time, with one of those on average being a 40+ yard play (77^th worst nationally). In addition to letting through an explosive, once a play had gotten behind the defense, it was a virtual guarantee to go to the house – none of UNT’s safeties had the top-end speed to catch a runner in the open field from behind. So UNT remained in the bottom quartile in every category of super-explosive: 98^th in 50+ surrendered, 123^rd in 60+, 106^th in 70+, and so on.

Secondary

So naturally, Coach Morris brought two safeties and a cornerback from UNT for the rotation. While Tulsa only hit the one very big rushing play, they feasted on explosive passing.

This was another area where Cody and I discussed the possibility of making some roster changes. In my opinion, returner #9 CB Fields should get a dedicated starting spot and returner #14 CB Foreman should be allowed to challenge the transfers #8 CB Lewis and #4 CB Horn, both of whom I watched the 2025 tape on.

I think Miami transfer #25 DB M. Williams should get a shot at boundary safety or nickelback, and returner #12 DB Epps should get a shot at field safety.

It was very clear from the Tulsa tape that the coverage only operated from a position of strength when the offense was significantly behind the chains and the pass rush could bring extra bodies and/or the QB got inside his head about it. Some examples:

However, on standard downs and short yardage, Tulsa’s passing offense tore the secondary apart. To avoid starting a riot I have not included all three of the screen passes which gained 20+ yards, just one, and a representative sample of the downfield passing game on standard downs:

More at NCAAF College Football News, Photos, Stats, Scores, Schedule & Videos