Graphing Clemson vs. Notre Dame: Brian Kelly was right — it’s a lot closer than it seemed

Hey look it’s two not-Alabama teams in these here charts!

We’re trying something new: looking at a notable game from elsewhere in the football-verse. Given that it was between the two other teams in the College Football Playoff — one of whom the Tide will face in the national title game next week — let’s just call it opponent research, eh?

If you read Graphing the Tide vs. Oklahoma, you already had to listen to me soapbox about the value of advanced metrics vs. the old “total yards and points” box score. That is, that these efficiency and explosiveness metrics give us, at worst, a great second perspective; and at best, a more transparent view of what went on in a football game. For the Bama-Oklahoma Orange Bowl, I explained (i.e., beat a dead horse) that the game wasn’t as close as it looked in the box score.

Well, how about that... the other CFP game this weekend fell prey to another misleading box score. Except this time the conclusion is the other way around. Let’s get into it.

Metric definitions

A ”successful” play, as defined by Football Outsiders, is when a play gains enough yardage to keep the offense on track, i.e., 50% of needed yardage on 1st down, 70% on 2nd, or 100% on 3rd/4th. A ”big play” (aka “explosive play”) is any play that gains ≥15 yards (run OR pass).

Success by Quarters

Can’t see charts? Tap here to load the full article.

Hey, wait a minute, I thought that the invincible Clemson Tiger-bots wiped the field with Notre Dame and that the Irish didn’t deserve to be there and that OF COURSE a real Tiger would eat a real Irishman alive!?

Well that Tiger vs. Irish guy thing is probably correct... but this game was much closer than it appeared to be. For a game that ended 30-3 Clemson, it’s pretty shocking how close the efficiencies were! Sure, there are considerations — Clemson putting in backups in the 4th, Clemson running 9 plays more than ND, Dexter Lawrence being out, and Trevor Lawrence’s solid 16.2% big play rate (XR) passing — but still, that’s only a 6-point difference in the overall Success Rate (SR) between these two teams.

I’m not saying that the Irish should’ve won this thing by any means, but it’s safe to call this game much closer than the 10x score differential indicates. Really, the biggest difference here is in the overall explosiveness rate (XR): ND held Clemson to a not-that-great 12% XR, but the Irish offense only squeaked out a 7% XR of their own. That’s not good. And they lost! But not without putting up a league-average 40% SR of their own.

UPDATE: since this posting, Bill Connelly of SBNation and the S&P+ analytics system published his Five Factors box scores for the bowl games, including the Cotton Bowl. His numbers favor Clemson a bit more, with a 45% SR to Notre Dame’s 36%, but the general storylines still stand. Really, it doubles down on the idea that explosiveness (and taking advantage of opportunities with explosive plays) was really Clemson’s key in this one — not really the “play by play” dominance that many are suggesting.

Play Map

The play map tells much of the same tale: Notre Dame really hung on for a quarter and a half: Clemson had zero explosive plays (15+ yard gains) in the first quarter and put up a below-average 35% SR.

But boy did the Tigers break out in the 2nd quarter: that dark orange cluster of long explosive plays in the 2nd quarter is basically the story of the game. Notre Dame’s defense is good, but after they uncharacteristically broke a few times, they didn’t have a worthy answer when they got the ball.

Sure, the Irish put in a solid 3rd quarter (including a few explosive passes and several successful runs) and in general had a good distribution of successful plays throughout the game. But, unlike Oklahoma’s performance against the Tide, the Irish didn’t get many points out of that broad distribution.

Success by downs

Both teams were pretty even across downs, in terms of efficiency, but those 3rd and 4th downs are where the differences are more obvious (and more critical).

In the Orange Bowl, Alabama was 7 of 10 on 3rd down and Oklahoma was 6 of 13. But we saw a lot more 3rd downs in the Cotton Bowl: Notre Dame converted 5 of 17 while Clemson converted 9 of 18 (!).

Per this chart, 3rd down was the most explosive down for the Tigers, too, with almost all of those (excepting Travis Etienne’s long run in the second half) being long passes. In short, this game was won largely on those (many) 3rd downs, and the Tigers were on the winning side of a few (riskier, perhaps) big 3rd down plays. It’s not a slight to Clemson, but a few inches here and there could have dramatically altered these outcomes.

4th down was another story of opportunity: the Irish did convert 1 of 2, but the one they did not convert was in FG range on Clemson’s side of the field, while the one they did convert was on their own 21 yard line in the 4th quarter. For Clemson, they converted their one 4th down attempt, a 4th-and-1, on Notre Dame’s 14 yard line. Inches, people.

Running and Passing, Clemson

Clemson passed a lot, which makes sense given Notre Dame’s notably strong run defense. But... this was a lot a lot, with <33% run rates during long stretches of the game. Trevor Lawrence passed the ball 39 times before his backup came in to toss a few.

And it worked pretty well most of the time: that 45%-ish success rate through the 2nd and 3rd quarters is solid for a team that was passing it so often.

Interestingly, the success rates from passing and running switched during the 4th quarter... usually you see running SR’s slip when a team is trying to burn clock, but after that 4th quarter, running became technically the most successful part of Clemson’s offense.

(Uh, just ignore the majority of the game where Clemson didn’t run the ball well. Yes, Notre Dame has a good run defense, but Clemson has Travis Etienne. Weird. Hopefully Bama can take advantage of this next weekend).

Running and Passing, Notre Dame

As for Notre Dame ... apparently they wanted to run the ball. They really did want to! But it wasn’t in the cards: after mixed-to-poor rushing success early on, the Irish quickly went towards the pass and rode a slight-majority-pass offense the rest of the way. I suppose that’s the effect of the basically-NFL defensive line that Clemson trotted out to meet ND’s run game (with Dexter Lawrence or not).

(I also suppose that this is why the game took so long: both of these teams were passing most of the time, and neither was very explosive doing so.)

The switch to passing worked for a bit for the Irish: they kept things interesting for a quarter and a half before the Clemson defense clamped down on that, too. Interestingly, Notre Dame’s running game started finding more wiggle room late, but that’s likely in part due to Clemson’s large lead and some subbing (Clemson had a lot of tacklers in this game).

Final thoughts, from a Tide POV

Look, Clemson won fair and square, and looked darn good against a darn good ND defense. They’ve looked good all year long and deserve to play for the title.

But anyone perusing the semifinal scoreboards is at risk of coming to the wrong conclusion: the Alabama-Oklahoma game was not that close, and this Clemson-ND game wasn’t nearly the domination that it appears to be. (Convenient conclusion for an Alabama fan, certainly, but I’ll refer you again to the numbers above and for the Orange Bowl. Also, Alabama has benefitted from similar games in the past, including vs. Clemson last season).

Surely the respective game plans will be different for the title game — Clemson is much more well-rounded team than Oklahoma, and Alabama is much more explosive than Notre Dame — but there are some general notes that your Tide Gump can take away from the Clemson vs. Notre Dame semifinal:

• Clemson trusts Trevor Lawrence, a lot, and may want to pass the ball, a lot. They’re good at that. Bama has given up explosive plays this year, so that’s potentially scary, but otherwise you almost want the Tigers to rely on these riskier, long 3rd down passes... I mean, they won’t land all of those close ones, and Alabama’s DBs might be better equipped to snatch a few, too.
• Clemson has started off relatively slowly in a few recent games, including this one. The Tide won’t be able to do to the Tigers what it did to Oklahoma in the first half, but if they can start fast then maybe Alabama can work from an early lead.
• Trevor Lawrence doesn’t run much. Seeing that Kyler Murray was the only guy to really run on the Bama defense in the Orange bowl, that’s probably a good thing for the Tide.
• However, Clemson had eight runners and 13 receivers in this game, so beware all of the relative no-names that may come out of the backfield. This team is deep.
• Hunter Renfrow must be on his, like, 4th college degree by now.
• As someone here on RBR mentioned a few weeks ago: Trevor Lawrence looks like Jar Jar Binks (Update: It was UtahBammer!). It’s yet another angle we’ll need to analyze going into this game. How does that impact Alabama’s young defense?

Roll Tide and enjoy the week. Come Monday, we’ll see our 4th set of Tigers this year!