As I detailed last week, Pinnick has led one of the worst offenses in college football by whatever metric you want to use — one that compares to the putrid offenses we watched under Paul Wulff. The best time to bench Pinnick would have been at halftime of Fresno State, after the “tough schedule” excuses had run out; the second best time would have been after Fresno State; the third best time would have been at halftime of Friday’s game at Utah State.

There were always going to be reasons to not do it at those times — most of them legitimate, sensible, and rational. Against Fresno, the game was still in the balance at half, and I’m sure the thought was just to get him playing a little better and then win the game. After Pinnick melted down in that second half, Moore tapped himself as play caller on a short week, so that’s not a typical time to change, either. And then on Friday, the offense was OK in the first half, and Pinnick was OK in the first half, and you had a two-TD lead. So it wasn’t crazy to think they might be able to just ride it out to beat a team such as Utah State.

But also, there isn’t a person — including the coaching staff — who should have watched any of the six quarters leading up to halftime on Friday and thought: The quarterback is good enough. Even with a lead against the Aggies, it was clear the offense still needed better play from the QB, and banking on a 16-3 lead to hold up without making a change was the same flawed logic that led to the loss against the Bulldogs.

As the season was slipping away, boldness was required … and left unrequited.

So now, because of that, we sit at 1-5, without much of a prayer of salvaging even a bowl game after making this change. But hey, making a decision at the fourth best time is way better than making it at the fifth best time, so I guess I’ll take it. At the very least, it adds some intrigue to a season that was quickly becoming a chore to follow.

In announcing the change, Moore tried to not completely hang Pinnick out to dry, but after some prepared remarks that attempted to spread the blame,¹ he really didn’t put a lot of effort into avoiding comments that would lead to some pretty obvious inferences.

Clearly, Moore thinks he did just fine as a play caller and the QB let him down. And if we assume that coaches are largely speaking to their players in these media availabilities (which I do), then Eshelman obviously has his marching orders.

I have no idea if Eshelman will be an improvement over Pinnick. I’ve never seen him throw a ball, and I assume there are legitimate reasons why Pinnick not only started the season as the QB, but also continued to be the QB, even as he struggled mightily. But this change simply had to be made — for Pinnick, for the fan base, and even for the rest of the team.

Consider these comments from safety Jaylen Thomas, who was asked about Eshelman’s elevation to starter: "I'm excited. Obviously, we're due for some change."

Yeeeeeesh.

And then: "(Eshelman) keeps his eyes down the field, he's a good pocket maneuverer. And he's good at completing passes down the field — that's what I'm super excited for and super excited to see."

Clearly, everyone needed to see someone different under center.

Benching Pinnick and starting Eshelman does not seem like it’s a move for only this week (“it’s about what’s best for the team heading into the second half of the season”) and there is no indication that Eshelman will have a short leash. Moore gave Pinnick as much of a chance as he possibly could, they exhausted literally every avenue to improve his play, and he’s out of runway. There’s just nothing else to see.

Which is pretty shocking. WSU was far from the only school after Pinnick following his big season as a redshirt freshman at UC Davis at the FCS level; on the most surface of levels, going after the Big Sky freshman of the year who put up excellent dual threat numbers on a pretty good team seemed like a good bet, at least to fans.

Unfortunately, five games against FBS competition has revealed him to be unqualified for the jump. Check out his throwing matrix:

(I went over EPA last week, but if you missed it: Just think in terms of whether each play results in the team being more likely to score points or less likely to score points, and by how much. In this chart, we’re simply measuring his throws. Also “aDOT” is “average depth of target” in that zone and “YAC” is “yards after catch” by the receiver on completions.)

You simply cannot overcome a quarterback who:

For a point of context, here’s the throwing matrix for Oregon State QB Braden Atkinson (another FCS up transfer) against San Diego State on Saturday, in which the Beavers dropped 66 points — note that he was 6-of-10 for 204 yards on deep passes in just this one game:

The bigger question for WSU, of course, is why Pinnick has failed so spectacularly. I went in search of video from last year to try and see what Moore and his staff saw, and it was harder to find than I thought. But these are a couple that showed some things, if you’re interested:

In watching them, I gotta say — and maybe I'm being biased by how terrible he's been this year — that it really reminds me of watching the high school Hudl highlights of a lightly recruited guy where people are crowing about how good he is, and how he should have more offers, because he’s thrown a bunch of TDs against lesser competition.

There are lots of big plays in these videos, but the QB doesn't show much on them that would make you think "oh yeah, that will definitely translate to FBS." The receivers are typically running wide open, and many of the short passes are thrown almost kind of floating, without any zip. Basically, it's the sort of stuff that should make you wonder how he'll do when the windows are a lot smaller: Some of the tight-window throws he makes here need to come a split-second earlier at the FBS level, and his accuracy on those throws is not good ... which I would take as a major red flag if I had been scouting him.

I’m not trying to shit on the guy and say he looked like a completely incompetent passer last year, that WSU somehow massively failed in its evaluation process. He shows himself as capable of things he has blown this year, which would indicate — even without the armchair psychology that comes from interpreting body language — that he’s suffering from a bit of a crisis of confidence. I think Moore’s quotes alluding to indecisiveness indicate as much.

But it’s a bottom-line business. And the bottom line is that Moore and his staff got the QB wrong, and that has undermined the entire operation in their first year. Every part of the process that led to them getting to this point needs to be scrutinized and adjusted accordingly.

I’d probably start by prioritizing how well a guy can throw before I worry about how well he can run.

The Cougs finished at 140 last season and have turned over basically the entire roster. And while I’ve written about how David Riley’s approach to building the roster this time is an improvement over the approach he took last time (Part 1 and Part 2), I wouldn’t say my expectation is for this year’s team to be significantly better than last year’s team. And for what it’s worth, the other major preseason projection — at barttorvik.com — forecasts the Cougs to be worse than last year, slotting them at 165.

Rarely do Ken and Bart’s computers diverge so much, and I think the simplest explanation is that player movement just makes preseason projections really, really hard.

Bart’s projections are (as far as I know) 100% player based — the computer calculates some likely minutes, usage, and stats for the distribution of players on the roster and spits out a composite projection. It’s fun from a fan perspective because the projections can change daily during the highest periods of player movement (you can even play around with rosters by adding and taking guys away and seeing how it impacts the rankings). But also … it’s super hard to get a computer to figure out how roles will change and players will fit together when everyone is new:

I will be very surprised if Akins (a point guard transfer from Wake Forest) ends up being the team leader in usage and points. I will be absolutely shocked if Houston doesn’t average double digits. But that’s because I’m a human and I’ve thought about how Riley might be approaching roles; when I see usage distributed as evenly as it is here across that many players, it tells me the spreadsheet really has no idea what to do with this roster.

By contrast, player projections are only one part of kenpom’s calculations, which he describes as being made up of “talent, recent program history, and conference strength, mainly.” He also includes coaching history. Over the past three decades, these are the things that he has found to be most correlated with a team’s actual performance once the season tips off.

How do those things add up to WSU looking much better in his system?

I suspect a lot of it is in the “talent” realm, where I would guess that the lofty high school rankings of high major down transfers such as Harrell, Jones, and Phillips are still being factored in. The offense is projected to be about as good as it always is when Riley is coach, but the defense is expected to take a big step forward. Why? Again, just making an educated guess, but I’d think his system really likes the idea of high major athletes performing better on that end, regardless of the coach’s history with defense. And I can’t really disagree with that reasoning — not after watching us try to defend with lesser athletes the past two seasons.

For whatever it’s worth, Bart’s projection has WSU finishing 11-21 and 4-12 in the Pac-12; Ken’s has us 15-16 and 6-10.

I don’t know which projection will prove to be more correct; if you forced me to make my own projection, I’d probably slot them in between — just a little better than last season. But that’s largely influenced by just how underwhelming Riley has been as a coach in his first two seasons. I watched almost all the games over the past two seasons, and I am human, after all.

However, if you’re a person who is inclined to hope … take that kenpom projection and run with it! It wouldn’t take much above that for the team to get on the right side of .500 overall and up to .500 in the league. After last season, I’d consider that a smashing success.