Super Bowl LIII Preview: An Analytical Guide to Sunday's Big Game
Maybe you feel like this shouldn't be this year's Super Bowl matchup. After all, the Rams are in thanks to a blatantly bad no-call at the end of the NFC Championship game, and the Patriots are still dancing due to an offsides penalty that didn't need to happen.
Me? I'm an unbiased analyst who's just looking forward to a good game this weekend. The past is the past. I'm onto Super Bowl LIII.
Sunday should be fun. And interesting. And probably dramatic, because that's what we get with New England Patriot Super Bowls.
Whether you're watching the game with nothing on the line or playing a contest (or 10) over on FanDuel, this preview -- this is the one -- is for you. We'll be digging into the numbers to uncover unique storylines, matchups to exploit, and ways for each team to lift the Lombardi on Sunday night. You may even find out something new about Tom Brady's eating habits. (You won't.)
Let's get nerdy.
Here at numberFire, instead of simply using win-loss records to determine how strong a team is, we use a metric called nERD. Fitting, I know.
nERD looks at things like point differential, opponent, and on-field play through an expected points model to determine by how many points a team would be estimated to win against an average one on a neutral field. If a squad has a nERD of 3.02, then we'd anticipate that team beating an ordinary one by about a field goal on a neutral field. If the nERD score was -6.89, then that team would be expected to lose by close to a touchdown and an extra point.
Entering the playoffs, nERD pegged Los Angeles as the best team in football with a rating of 8.90. All the while, New England ranked fifth-best with a nERD of 7.25. Really, the two teams weren't drastically different performance-wise.
Their journey to this spot wasn't nearly the same, though.
According to nERD, the Rams were the better team throughout 2018, but they also peaked early in the season. While New England's only gotten better since their embarrassing loss to the Lions in Week 3, Los Angeles had a mediocre three-game stretch from Weeks 13 through 15, and the team hasn't fully gotten back to where they were earlier in the season.
And that serves as a good starting point for us. What changed for LA during that set of post-bye games, and what does it mean for New England's approach in Super Bowl LIII?
Getting to Goff
Teams lose multiple games in a row all the time. But since the Rams were rolling for much of 2018, it was a little surprising to see them drop two straight contests, especially late in the season.
In Week 14, the Rams were smoked by the Bears. Maybe the final score didn't show that, but LA posted just six points and turned the ball over four times during their Sunday night game in Chicago. It wasn't pretty.
Then, the following week, the Eagles came to town and forced three turnovers, beating the Rams by a touchdown.
In those two contests, Jared Goff's numbers were more cringe-worthy than a Tom Brady dance move. (We've got to take jabs at the greatest of all time when we can, right?)
|Out of Split||14||8.98||32||7|
Say what you want about the opponents in those two games -- Chicago had arguably the best defense in football this year -- but a 5.35 yards per attempt rate isn't going to cut it. That rate's actually lower than the average allowed by the Bears this season.
How'd this happen? Why was Goff so abysmal for this two-game stretch? Well, there's no single answer to this, but one piece of it is certainly pressure faced. Per Pro Football Focus (PFF), Goff was pressured on over 52% of his drop backs against the Bears. That was the highest pressure rate he'd seen all season. Against Philly, that number fell to a little over 29%, roughly his season average. The Eagles were able to get in his face, though, as Goff was hit seven times in that game (the second-highest mark of the season, behind only the Chicago game), finishing with a 3.47 yards per attempt rate when pressured. For reference, across the entire regular season, Goff's yards per attempt rate while pressured was 5.82, according to PFF.
This issue isn't unique to these two games, either. Goff's been fantastic when kept clean: in a clean pocket this season, he had the fourth-highest quarterback rating among relevant passers. Under pressure, his quarterback rating ranked 28th. And of the quarterbacks this season with 200 or more drop backs, Goff ranks first in the differential between his yards per attempt in a clean pocket versus his rate when pressured.
So far in the playoffs, New England's been able to get to opposing quarterbacks, despite finishing the regular season with the second-lowest sack rate in football.
|Quarterback||Regular Season Pressure Rate||Against New England|
What could help Goff in this contest is the play-action pass. No 300-plus-drop-back quarterback ran a higher rate of play-action passes this year, and on those tosses, Goff had a top-five yards per attempt rate. Keeping the Pats' defense on their toes could help mitigate pressure thrown Goff's way. And what's interesting is that, against New England, both Rivers and Mahomes threw play-action passes at a lower rate than in the regular season.
The Kansas City Blueprint
Almost all quarterbacks will be worse under pressure, but this year's Super Bowl features two passers who see pretty large discrepancies in efficiency when they are and aren't under duress. Goff may rank first in yards-per-attempt differential when kept clean compared to when pressured, but Brady was eighth within the statistic this year.
Generating natural pressure on Brady is nothing new -- it's how teams have tried to beat him for years. And this is definitely a matchup where the Rams could win this weekend, because on the season, no team had a higher pressure rate than Los Angeles.
Does that mean we're bound to see a more balanced approach from New England's offense?
So far in these playoffs, New England's had a much more run-heavy attack when in neutral game script situations, or when the game is within a six-point margin.
In the regular season, their 1.47 pass-to-rush attempt ratio when games were close ranked 11th-highest in the league. Their 1.08 ratio in the playoffs would've ranked fourth-lowest in the regular season, ahead of only Tennessee, Baltimore, Buffalo, and Seattle.
Now, if you look at season-long numbers, you may notice that the Rams finished dead last in yards per carry allowed this season. Funny enough, Kansas City -- the team New England ran all over in the AFC Championship -- finished 31st.
But there's a big difference between the Rams' ability to stop the run and what we saw with the Chiefs a couple of weekends ago.
numberFire has an expected points model aptly called Net Expected Points, or NEP, which tells us the number of points a team or player adds or loses against expectation. Each time there's a run or a pass that provides positive expected points, it's deemed a success. When something bad happens -- like a rush for two yards on third-and-three -- it's a failure. The percentage of positive plays made, then, is called Success Rate.
During the regular season, the Patriots were a fairly good running team, but they weren't unstoppable. The chart above shows that. On the x-axis, you've got Rushing NEP per rush, or the number of points added by a team's group of running backs on a per-rush basis. The y-axis is the Success Rate among those players, which ranges from about 33% (the Jets) to 50% (the Rams). New England's data point is staring at you in red.
Like I said, they weren't unstoppable.
New England has found crazy efficiency running the football during their two playoff games this year.
Naturally, with such a small sample, strength of opponent is going to play a larger role when looking at these types of numbers. And throughout the 2018 season, the Chiefs allowed the highest Success Rate in the league to opposing running backs. They were effectively numberFire's worst-ranked rush defense. So the Patriots weren't just running the football to run it. They were exploiting a weakness.
As I said above, the Rams appear to be as bad as the Chiefs at stopping the run. But they're not. They're not like the Chiefs. According to numberFire's expected points model, since Week 12, Los Angeles has surrendered a sub-35% Success Rate to opposing runners, which is a top-five number in the NFL. In the playoffs, despite facing two top-10 rushing offenses in the Cowboys and Saints, they've allowed a 38.9% Success Rate. That would've ranked in the top-10 had they kept it up throughout the regular season.
The return of Aqib Talib has helped with this. Yes, he's a cornerback, but he also allows more resources to be allocated to other areas of the field. This season, the Rams have allowed a 4.57 yards per carry average with him healthy an active, and without him, that figure jumps to 5.18.
What this means is that New England probably can't just replicate the offensive game plan they had against Kansas City. They may not be as successful on the ground.
But here's the thing: to alleviate some of the pressure that Brady will see, it'd be logical to think that we may see the classic Patriots passing attack that's littered with quick, short attempts. It would make sense to see a lot of looks go to James White and Julian Edelman.
That may not be so easy against the Rams.
The chart above is from AirYards.com, and it shows how well the Rams defended passes by depth in 2018. As you can see, LA was much stronger compared to the league's average on shorter throws, and once passes started traveling 15 or more yards through the air, they began to struggle.
That could be an issue for New England. Without Josh Gordon, Brady and the Patriots' approach to deep-ball throwing has changed a bit.
|Att||Deep Ball Att||Deep Ball Rate||Deep Ball Comp %||Deep Ball Yd %|
With Gordon in the mix, about 17% of Brady's passes were traveling 15 or more air yards. That's dropped by 2% since Gordon left the team. Perhaps more importantly, Brady's seen his percentage of yards coming from those throws drop by almost 8%. So not only is he throwing it deep less frequently, but he also hasn't been as effective on those throws.
But those tosses may need to convert for the Patriots to find success on Sunday.
A Gronkowski Sendoff
Just like most of the Patriots' offense this postseason, there's been a new type of Rob Gronkowski. Against the Chiefs, Gronk saw 23.9% of Brady’s overall attempts, when that prorated number during the regular season was 15.6%. And this happened while New England has become more focused on running the football, at least from a pass-to-rush ratio standpoint.
The Super Bowl presents an interesting matchup for him, too. Through Week 16 of the regular season -- so, relevant games for all 32 teams -- the Rams ranked sixth in yards allowed to tight ends. A hefty 25.02% of their yards allowed were going to the position.
Tying back to the notion that the Patriots may have to chuck it deeper than normal, the Rams also gave up 14 deep-ball plays to the middle of the field this season -- where Gronk will frequently line up -- which tied them for the fourth-most in the NFL. Gronkowski led all Patriot pass-catchers in deep-ball targets this season, making him the player to watch when Brady doesn't get rid of the ball a second after it's snapped.
A Slot Matchup to Exploit
We know Julian Edelman can do serious work in the slot, but it should be noted that, among the more highly-used slot cornerbacks this season, LA's Nickell Robey-Coleman ranked best in yards allowed per snap according to Pro Football Focus. New England's approach in this game stands to benefit Edelman (he's seen more deep-ball targets than Gronk in the playoffs), but Robey-Coleman does present a difficult matchup in the slot.
Edelman's not the only slot receiver to watch in this game. In 2018, about 33% of the receiving yards allowed by the Patriots went to the slot. Only the Lions and Falcons allowed a higher percentage. And since Cooper Kupp's season ended in Week 10, over 68% of Robert Woods' snaps have come from that area of the field, per PFF. Brandin Cooks has played 26.0% of his snaps from the slot, while Josh Reynolds is at 31.1%.
Without Kupp in the lineup, Robert Woods has seen a rise in targets per game, while Cooks has seen a dip.
Across the first 10 weeks of the season -- and, for the record, I've removed Cooks' game against Seattle where he was knocked out in the first half without a target -- Cooks was outpacing Woods in targets per completed game played. Since, things have shifted in Woods' favor, partially because 56.5% of his targets are now coming in advantageous matchups in the slot, when that number was 36.7% with Kupp active.
The Rams have also just been throwing shorter passes since the Kupp injury. Whether it's because of Kupp's absence or not, since Week 10, the average depth of target for both Cooks and Woods has dropped by 1.2 yards. And Jared Goff has struggled -- and I mean struggled -- with the deep ball.
|Split||Att||Deep Ball Att||Deep Ball Rate||Deep Ball Comp %||Deep Ball Yd %|
To be fair, Goff's deep-ball ability during his first 10 games wasn't all that sustainable. He was completing nearly 59% of his passes that traveled 15 or more yards through the air, which was the second-best rate in the league behind only Drew Brees. And it was far ahead of the other qualified passers.
Since then? Whew. We're talking about a bottom-of-the-barrel completion rate on deep balls. Like, Ryan Tannehill-esque.
Fortunately, as long as the Rams don't get crazy with their deep-ball tossing in the Super Bowl, this may actually be OK against the Patriots
When compared to the league average, New England is a lot better at defending deep passes than shorter ones. Given the matchup in the slot and his lower average depth of target compared to Brandin Cooks, Robert Woods has a chance to ball out on Sunday.
Todd Gurley was banged up in the Rams' Week 15 game against the Eagles, and since then, his workload just hasn't been the same.
|Split||Att/G||RB Rush Share||Tar/G||Target Share|
It seems like this is all injury-related, but Gurley was taken off the injury report prior to the NFC Championship game, and he's claimed to be healthy. So let's not speculate -- let's just assume he's 100%.
In the playoffs, C.J. Anderson has seen nearly double the amount of carries per game compared to Gurley. Despite this, Gurley is still out-pacing him in targets per game. And that's where we could see the Rams use Gurley in the Super Bowl -- in the passing game.
In the regular season, New England ranked near the bottom-10 in Success Rate allowed to the running back position through the air. They gave up the ninth-most yards to the position, and almost 20% of the receiving yards surrendered by the Patriots this year went to running backs. That's a top-10 number across the league.
The air yards plot above shows us that New England allows a higher-than-average catch rate until about the 12-yard mark. Considering Gurley's the primary pass-catcher in the Rams' backfield, we should expect LA to exploit that.
Meanwhile, Anderson will more than likely continue to see a decent amount of early-down work. According to numberFire's expected points model, when counting the playoffs, no 50-plus attempt runner has a higher Success Rate than CJ Anderson. He's creating positive expected point plays on over 62% of his rushes, when that number is 49.6% for Gurley this year. So, no, CJ Anderson probably isn't going away, at least as a runner.
If you remove the names and look strictly at the matchup and the talent on both rosters, it's easy to side with the Rams in this game. They actually match up pretty well against New England -- they can pressure the quarterback, they've got the right offensive pass-catchers, and their rush defense has looked much better over the latter parts of the season.
This doesn't mean New England doesn't have advantageous matchups, though. Their ability to create pressure could put Jared Goff in a really rough spot given his splits in and out of a clean pocket, and Rob Gronkowski does have an opportunity to exploit a weakness against this LA defense.
If you were to ask almost any NFL analyst who has the best personnel across these two teams, the majority would answer with the Rams. They're talented -- it's a very talented roster. And that showed throughout the 2018 season, as they finished the year with the highest nERD rating in the league.
But Bill Belichick. It's so much easier to trust that he'll come through with a unique gameplan to exploit these matchups. And that's why -- even if the Rams are better -- I'm picking New England to win this game. And with both teams more than likely taking a methodical approach offensively, I see this contest being lower scoring than where the over/under sits at 56.5. So give me the Patriots, 24-23.
If you want to see what numberFire's algorithm thinks, click here.