Our eyes told us it was ugly, and the advanced stats confirm that and then some. The Ball State defense put up a havoc rate and run stuff rate against Notre Dame that would have been best in the nation last year! The Irish also struggled to capitalize on scoring opportunities, and were extremely inefficient on passing downs for a second straight week. The saving grace was the ND defense holding the Cardinals to 3.65 yards per play, allowing the Irish to hold an almost +2 YPP edge in that category. Teams that win the YPP battle by between 2-3 YPP typically outscore their opponents by 24 points on average, but Notre Dame scraped by with an eight-point win.

Boilerplate: Check out this advanced stats glossary if something is unclear or give me a shout in the comments. Last week I found out leverage rate wasn’t in there!

The Basics

While the yards per play margin wasn’t close, Ball State managed to extend drives just enough to make Notre Dame defend a record 96 plays (plus an end of half kneel down). For those new to this space, I tend to define garbage time fairly subjectively. It’s obviously not possible to do this for advanced stats systems tasked with breaking down every single FBS game, but watching there are usually clear inflection points where teams change strategies / personnel and garbage time is on. Technically this game started the 4th quarter in garbage time with Ball State down 18 points, but then would have exited it once they scored a touchdown on that drive. So in my books zero plays of garbage time, so only end of half / game kneel-downs were excluded, which was an unpleasant surprise.

Efficiency

After a first drive that was a cakewalk, the Irish offense lurched their way to just 17 more points over the ensuing 13 drives. The gameplan seemed off, Brandon Wimbush was frustrated for much of the day, and the Irish failed to convert scoring opportunities. Success rates were extremely average across both running and passing, which isn’t what Chip Long was expecting against Ball State. For context, against Temple and Miami (Ohio) last year the ND team success rate was close to 50%, with run success rates nearing 60%.

The primary concern, as it was entering the season, is Wimbush and the pass game. The play-calling and scheme for this game did him no favors – it seemed like the coaching staff wanted to tape him in bubble wrap and work on success as a pure passer. There were zero designed run calls for the QB in the first half, and zero scrambles (perhaps at the instructions of the staff). Wimbush’s running is arguably the most dangerous part of the offense. It was disheartening to see that without it the Irish couldn’t consistently score on Ball State, but also, why fight with one hand tied behind your back?

There’s enough of sample at this point to know that asking Wimbush to succeed solely as a pure passer isn’t viable. I still believe his running ability is so valuable that he’s still the best guy to run the offense. But if you aren’t going to let him use his legs as a threat, making the passing game easier as a result, why bother?

Still, Wimbush played poorly. Poor efficiency on passing downs against Michigan is to be expected – not so much against a MAC team. The encouraging signs of growth seen last weekend against the Wolverines – strong decision-making, not letting a bad play snowball into frustration and multiple misses – dissipated against a defense that could charitably be characterized as untalented. It seemed like the caution against running – whether designed runs or scrambling – threw Wimbush off. There were more than a handful of occasions to throw the ball away, and instead throws were forced or unnecessary yardage lost.

It wasn’t all awful – Wimbush completed 55% of his passes and averaged 17.5 yards per completion. Of the three interceptions, one was definitely not on the quarterback and a second was (like the Michigan pick) much in part due to pressure and getting hit while he threw. But against Ball State – playing fairly aggressively on defense – could Book have hit those open receivers? Or even Jurkovec? The intermediate throws were often wide open, so who knows if with a better passer under center if those are tighter windows but potentially the running lanes are a little wider.

How do you average 8+ yards on first down and still have an average 3rd down distance to gain of 9 yards plus? When you’re relying heavily on explosive plays, and not gaining many yards or losing them on other downs. The Irish offense was very boom or bust (as we will examine more closely in the explosiveness section), which history would tell us isn’t a sustainable way to perform at a high level.

Now for positive things  – the Irish defense put on a very solid effort despite getting little help from the offense and special teams. The first Ball State drive produced a little anxiety, as the ND secondary played off Cardinal receivers and allowed Riley Neal to pick up some easy short completions and first downs. After that, it was pretty much a shut-down performance, with two picks and eight pass break-ups (four for Julian Love).

The two areas for improvement are closely linked – passing down efficiency and sack rate. Despite heavy pressure all game (9 QB hurries) the Irish tallied just one sack, although the guys in black and white were somewhat complicit. The old “if we hold every play, they can’t call it every play” was extremely effect for Mike Neu’s offense. Still, the Notre Dame defense should have come up with more sacks with better rush discipline, and that issue translated into Neal (did you hear he’s 6’6?) scrambling for multiple first downs.

Still, the Irish continue to limit opponent efficiency while forcing long drives and giving up very few big plays. A few more stops could have translated into better field position and maybe a few more points for the offense, but it’s hard to find much fault with the unit that’s avoided major mistakes and played fundamentally sound.

While the stars are largely performing like stars and getting the expected credit, a couple of unsung heroes of this game were Jonathan Bonner and Nick Coleman. Bonner hasn’t put up big numbers but had four QB hurries, and has made a fairly seamless transition to holding his own at the 1-tech as a fifth-year senior. Coleman had a couple of great plays in the nickel, including the deflection that led to Jalen Elliott’s first pick.

Explosiveness

The saving grace for the Irish offense was that despite so-so efficiency, when they were successful they continue to be fairly explosive. There was just one gain of 40+ yards, and it came on the first series when Jafar Armstrong scampered for 42 and it looked like this game would be a stroll in the park. But there were also 13 gains of between 17 and 31 yards, with 11 of those coming through the air.

In a day without a lot of positives, Armstrong’s contribution in the passing game (3 catches for 61 yards) and a very productive day from Miles Boykin (6 for 119, despite a drop leading to a pick) stood out. The running back rotation continues to be a split between Armstrong (66 yards on 5.1 YPC) and Tony Jones Jr. (61 on 4.7, with a nice long TD). For an offense that’s dependent on big plays so far in this young season, Dexter Williams return from shadow suspension will be welcomed. If you’re going to need big plays to score, having one of your best big play threats taking some more carries helps double down on that effort.

Somehow we’ve come this far without mentioning the offensive line. If you attempted to use the transitive property to apply how poorly the line looked against Ball State retroactively to Michigan, I would have expected approximately zero total yards in the opener. Some of this can be filed in the folder titled “consequences of a poor gameplan / Wimbush refusing to throw it away”, but the offensive line warning lights are officially on:

  • Ball State posted a havoc rate of 27.5%, fueled by 10 tackles for a loss, three interceptions, and four pass break ups. Michigan led FBS last year with a team havoc rate of 23.7%.
  • The Wolverines also led the nation last year in run stuff rate, at 28.9%. The Cardinals, who finished 128th in Rushing S&P+ defense in 2017, stuffed the Irish for negative yards or no gain on 29.4% of their carries.

It’s not all on the line, and Jeff Quinn, etc. – there were poor decisions involved too, from Wimbush to Jones of all runners deciding to reverse fields. But it’s certainly a disturbing start, especially given the lack of success on passing downs, which the Irish would very much like to avoid until the passing game looks better.

Advanced stat profiles aren’t up yet for the 2018 season, but the Irish defense has to again be close to the nation’s best in limiting explosiveness so far. Notre Dame is allowing just 3.95 yards per play this season, a mark that would have been best in FBS last year. There’s room for growth, and the competition will get tougher, starting next week (Kyle Shurmur is probably one of the better QB’s the Irish will see this season, although he suffers the disadvantage of only being 6’4 instead of 6’6).

Finishing Drives, Field Position, & Turnovers

After a perfect effort in the red zone against Michigan, finishing scoring opportunities was a struggle for the Irish offense on Saturday. A red zone visit was snuffed out by a Wimbush sack followed by a missed Justin Yoon field goal. Another drive started with a first down on the BSU 28, then a ended with a two-yard rush and three straight Wimbush incompletions.

The saving grace was that Ball State had fewer scoring opportunities and were somehow worse converting them. The Cardinals, despite being massive underdogs, settled for four field goal attempts and punted from the Irish 40. Thanks Mike Neu!

The Irish held a significant edge in the field position battle despite a so-so special teams effort. Jonathan Doerer putting another kickoff out of bounds was instant rage-fuel, but other than that his kick-offs were solid. I’d still take a touchback every time, but that’s just me. Tyler Newsome didn’t have his best performance either, averaging 36.4 yards per punt. Chris Finke fielding a bunch of punt returns live on the bounce (after signaling for everyone to watch out and move away from a short kick) seems non-ideal.

Fortunately the defense was steady, the offense picked up more yardage than Ball State, and better field position followed. Each of the three Irish interceptions were far enough downfield that they didn’t gift instant scoring opportunities. Starting field position now seems like a fairly trivial area of concern relatively to other parts of the team and offense, but I’m still worried it will pop up and bite Notre Dame in a close game.

On the subject of turnovers, Notre Dame lost that battle 3-2. While two of Wimbush’s picks weren’t completely his fault, there were also lots of deep passes thrown into double coverage and tight windows. An offense designed around Wimbush’s running ability with deep shots sprinkled in seems ideal for the ND offense. But so far the deep passing game has been inaccurate and often asking Irish receivers to make plays with a couple of defenders in the close vicinity (continual blessings for Chris Finke converting one such one-on-two opportunity in spectacular fashion in the opener).

Big Picture

This game was a gross slog, and I was extremely glad when it ended, except for that part where I kept hoping for a late TD or two to make the final score more respectable. Ultimately, the Irish are 2-0, and this poor performance can be easily rationalized away with more wins. Some combination of a terrible gameplan plus a letdown game and too-easy start led to a gross game that ultimately didn’t matter.

If we want to sip on some Kool Aid for a minute, and be receptive to the idea that these types of performances can still turned in by talented and good teams, even national championship caliber ones, allow me to present you two scores from Week 2, 2016. Clemson was coming off of a big win over Auburn, and UGA a win over a ranked UNC team:

Finish September 5-0 and this will be a funny memory, except for the usual haters who will point to a win over Ball State by eight points as definitive evidence this team is no good. But it also could be the moment where we started to realize Wimbush 2.0 was really more Wimbush 1.01, or that the offensive line might be a liability for the season. The schedule will be the only true litmus test – one that annoyingly takes weeks before seeing what color our strip will turn.

Still, at the very least we know this is a missed opportunity, because most previews had key questions like “how many freshmen will play?” and “will Phil Jurkovec get to throw it when he’s in?”. This was the only obvious lay-up in a minefield schedule of teams expecting to make bowl games, and first-team starters had to play the entire game. It felt like a huge developmental win for the program last season to have so many big wins that allowed young players to gain experience, and now it’s hard to feel good about many of those opportunities coming up in any contest.

Notre Dame fell nine spots, to 17th, in S&P+ as a result. The rankings are extremely bouncy right now, as preseason projections are still included but slowly being phased out in favor of 2018 results. But 2018 results are still very messy, especially with opponent adjustments – it’s just hard to know through two games whose wins or losses or close games are good performances or not. Still, when you under-perform expectations but nearly 30 points against a team that was one of the worst in FBS last season, the computers aren’t going to smile favorable on you.

The defense is ranked 4th in Defensive S&P+, with the offense checking in at 77th. And that feels about right at the moment – Clark Lea’s crew is good enough to keep the Irish in and allow them to win every remaining game, but the offense so inconsistent and potentially one-dimensional that lots of close games also seem pre-destined if nothing changes. If anything maybe this struggle is a silver lining in its timing – the staff could have attempted to roll this conservative gameplan out against Vandy or Wake or Navy and suffered a loss as a consequence instead of a win that was ugly but never in doubt.

After the game, BK claimed that not running Wimbush was 1) not intentional and 2) part of “defining who we are as an offense. That’s what we’re trying to figure out. We have some new players that we’re putting in the mix. We’re in that process of kind of figuring out our identity on the offensive side of the ball.” That’s the concerning thing – there seems to be both an acknowledgement that they need to play to Wimbush’s strengths and avoid his weaknesses, yet a slow coming to grips with what exactly the offense is or should be. If a defense of Ball State’s caliber can be incredibly disruptive and through their looks force Wimbush to throw for the Irish to score, then we’ll see many repeats of that in the coming weeks.

A glance ahead at Vandy

The Commodores entered the season projected 79th by S&P+ in the preseason, coming off a 2017 that saw Derek Mason’s team finish 5-7 and 91st in F+. Two games in, and they’ve leapt to 23rd, beating the mighty powerhouses of Middle Tennessee State and Nevada. The Dores shut down MTSU 35-7 in the opener, spurred on by a 2-0 turnover deficit, and then beat the brakes off a bad Wolfpack team 41-10.

These performances have bumped Vanderbilt to 21st in Defensive S&P+, a rise which seems a little hasty for a team that finished 67th in the same category last fall. Still, the Commodore defense returned their best linebackers, and held MTSU to 3.2 yards per rush and Nevada to 1.4. It’s hard to argue with that early performance, even if the degree of difficulty is about to go up substantially.

The offense checking in at 43rd at this early juncture is a little more believable. Kyle Shurmur took a nice leap forward as a junior, throwing 26 TDs to 10 INTs and averaging 7.4 YPA (dragged down by expected struggles against Alabama and UGA). Junior WR Kalija Lipscomb is a big-play threat that had already tallied three touchdowns this season.

It’s worth noting that looking good against bad teams was kind of Vanderbilt’s MO last season. They turned in percentile performances of 88%+ against Tennessee, Middle Tennessee, Alabama A&M, and Western Kentucky. This was complemented by four performances worse than the 15th percentile against Bama, Florida, Georgia, and Ole Miss (those were all in consecutive games, by the way).

A quick aside – Bama might have broken them a little, because the numbers from that game are insane. The Tide won 59-0, with the Vanderbilt offense averaging 2.4 yards per play and 7.6 yards per possession. Vandy’s success rate was eight – 8!- percent. The offense’s percentile performance shows up as “NULL” on the advanced stat profiles. That’s bananas.

That tangent aside, it’s easy to see both 1) what S&P+ likes about Vanderbilt so far and 2) why skepticism that the Commodores will play up to the level of a quality Notre Dame team (presuming it plays at a top 10-20 quality level) is warranted. That said, there are things to dislike about the matchup- a veteran QB with a dangerous deep-threat, an offensive line returning all five starters, and a defense that has performed well against the run. Vandy’s front seven may not be great, but it’s certainly better than Ball State, which could become a problem if the offensive line doesn’t drastically improve this coming weekend.