The common line of thinking in the NFL is that you need a great quarterback to succeed, which has driven the league to the point where half the NFL every year pays over 10% of the cap to about half of the league’s quarterbacks. Every quality of quarterback is in this group, yet only one quarterback still in the playoffs is over 10% of the cap– and that’s Sam Bradford at 10.78% for the Vikings who went down with an injury and now isn’t even starting over the “journeyman” Case Keenum, who is paid 1.14% of the cap. Tom Brady has set an example with his cap hit between eight and nine percent of the cap this year and in 2016, which allowed the Patriots to field the best defense in 2016 and the best scoring defense in the NFL this year from week five on.

The other three teams in the NFL Playoffs Final Four have constructed a run-first, defensive model that’s taken awhile to catch on as teams are now copying the model the 2013 Seahawks used to successfully beat Peyton Manning’s top ranked offense that year. The league is a copycat league, but these trends take a few years to take hold as teams need time to re-construct themselves in the image of the roster they’re copying. While the Vikings are spending over 10% of the cap on Bradford, they had already begun the process of building this run-first, defensive model with Teddy Bridgewater at quarterback before the injury; but because of the low-costs they had on offense, they were able to splurge a little on Bradford. As Andrew Beaton of The Wall Street Journal realized in an article on the Vikings, their 40.1% of the cap spent on offense is less than all but two Super Bowl champions, the 2004 Patriots and the 2012 Ravens, two teams that also had some larger investments on the defensive side of the ball. When a team does have a first tier quarterback (over 10% of the cap) and high spending on defense, the team then has to find some serious values somewhere on the offense to compete for a championship with this model because there isn’t enough money to go around if you’re paying conventional rates for wide receivers and the offensive line.

It has worked out for the Vikings because they have just $9.3 million invested in quarterback Case Keenum, running back Latavius Murray, and wide receivers Adam Thielen and Stefon Diggs. Thielen performed at the level of a first tier wide receiver in 2017 with 91 catches for 1276 receiving and four touchdowns, which is a performance that can typically cost between six and nine percent of the cap, but which the Vikings got for just 2.24% of the cap. Stefon Diggs performed at a level that could cost something like 4% of the cap with his 64 catches for 849 yards and 8 touchdowns, but he cost just 0.50% of the cap as he’s a fifth round draft pick on his rookie contract. Latavius Murray and Jerick McKinnon created 945 and 991 offensive scrimmage yards, while consuming 2.27% of the cap, which is something that could cost four or six percent of the cap for an elite, complete running back.

Keenum is still being talked about in journeyman terms, but his 67.6% completion percentage with 22 touchdowns to just 7 interceptions, 3547 total passing yards for 236.5 yards per game and 7.4 yards per attempt was first-tier production. Most importantly, he was efficient: the Vikings had the ninth highest net yards per attempt passing, the second least interceptions in the NFL and the 11^th most passing yards in the NFL. While he was very inexpensive, he still performed at a level similar to what someone at Bradford’s costs could produce, which reminded me of what Tom Brady did for the 2001 Patriots taking over for Drew Bledsoe who consumed 10.29% of that year’s cap. In fact, Keenum’s offense is more productive and more efficient than the Brady-led offense was that year. The Vikings also spent just 11.60% of the cap on their offensive line, which is very low for the whole group, but according to Football Outsiders they were the 19^th best run blocking line in the league and the sixth best pass blocking line. They were able to piece together an offense that was 10^th in the NFL in points scored and 11^th in yards produced with many inexpensive and unheralded players, which is a big part of why Pat Shurmur will be the New York Giants’ next head coach.

This idea that you need a quarterback to great succeed isn’t unfounded; quarterback is the most important position at every level, but when a team over-invests in the position as some teams do, then the rest of the roster typically begins to have holes that are exposed come playoff time when your team plays a complete team like one that’s in the top 10 in offense in defense, top 10 passing and rushing, and so on. It’s rare that a team finds so much value as the Vikings have, which is why it’s critical to spend intelligently at this position. Even if a team hits on a quarterback at a 10%+ rate, it can typically create roster issues elsewhere and it typically also takes finding value at positions that supplement the quarterback, like wide receiver. Aaron Rodgers is unquestionably one of the best quarterbacks of all-time, but with a cap hit of 12.16% in 2017, plus Randall Cobb at 7.58% and Jordy Nelson at 6.92%, the team was highly likely to have a defense that ranked in the bottom third of the NFL regardless of Rodgers’ health. A large investment in a small handful of “great” players also decreases the number of “good” players a team can have on their roster.

While Bradford was costly over 10%, the Vikings had 27 players over 1% of the cap, which is a good metric for measuring how many “good” veteran players a team has on their roster. Looking at the Super Bowl champions from 1994 through 2009, before the new CBA locked in the low rookie contract rates we see today, the average champion had 27 players over one percent of the cap. The new CBA actually created the unintended consequence of an increase in top end pay typically going to quarterbacks, receivers, cornerbacks, defensive ends and offensive tackles, while most of the “good” veterans in the middle have been priced out of the league by less expensive rookie contract players.

From 2011 to 2016, the average Super Bowl champion has averaged just 24.2 players over one percent of the cap. The 1998 Broncos had 31 players over one percent, the 2003 Patriots had 30, and the 2009 Saints used cap rollover to carry 32 players on their roster over this number. The Vikings have an older school approach that was made possible by their roster construction strategy leading into the Bridgewater Era and it seems almost unaffected by the one big investment in Bradford. They’ve gotten top tier production out of cornerback Xavier Rhodes at 6.24% of the cap, which is the bottom of the first tier for that market, while defensive end Everson Griffen gave them 13.0 sacks at 5.15% of the cap, which is actually a second tier price. Bradford was a good player for them in 2016, he led the NFL with a 71.6% completion percentage and gave them 258.5 passing yards per game with 20 touchdowns to just five interceptions, so he produced at a high level, so it’s not like he’s a wasted cap figure—he just go hurt, and Keenum has played at a similarly efficient level. The cap hits of Bradford and Rhodes consumed 17.02% of the cap for two good players fits right into the 16-18% range we want to see teams cap their top two player costs to provide themselves the opportunity to build out the rest of their roster as Minnesota has done. Their top three with Everson Griffen costs 22.17%, which is in line with what the Patriots have in Tom Brady, left tackle Nate Solder, and safety Devin McCourty at 21.62% of the cap, a good rate for three top of the food chain players.

Looking at the Super Bowl Champions data from my just released book Caponomics: Building Super Bowl Champions, the top paid quarterbacks for the 22 capped champions have been all over the map, which caused me to seriously question the concept of most of the NFL deciding to pay quarterbacks franchise quarterback money over the last few seasons once they hit their second contract as long as they proved they were the team’s starting quarterback.

While much of the NFL was increasing pay in veteran quarterbacks to the point where the top 15 players at the position all make money in the same high cost range, it opened a huge opportunity for teams with low-cost quarterbacks, specifically players on rookie contracts—like Seattle, to use their quarterback’s low-cost years to build a team in the run-first, defensive model that’s worked. The two Steelers champions were built off the rookie contract of Ben Roethlisberger as well. The 2012 Ravens were able to take advantage of Flacco’s last year on his rookie deal. The 2003 and 2004 Patriots were able to take advantage of the low-costs of Brady’s early years. This is a model teams should be following to build up the roster during the low-cost rookie contracts that will allow him to lean on a good roster, then when the quarterback hits his prime and higher earning years, the team can be more reliant on that quarterback as they’ll have to be because of his higher costs.

This is the model being used by Jacksonville, Philadelphia, and the Vikings. Even though Carson Wentz was an MVP candidate averaging 253.5 passing yards per game with 33 touchdowns to 7 interceptions, the team had the NFL’s third best rushing offense, which created the balance that made Wentz a better player and keeps this team competitive with now under Nick Foles under center for the injured Wentz.

Jacksonville has built their model with Blake Bortles at just 3.94%, which allowed them to spend 27.00% of the cap on their defensive line, which is similar to the 28.18% the Seahawks spent on their defensive line in 2013. The Eagles spent 21.80% of the cap on their defensive line and the result was seven players with over 20 pressures produced in 2017 and they were the only team to generate pressure on more than 40% of passing plays. The four defensive lines that are left all have elite talent and depth; the Vikings might actually be the weakest in the depth department with just five defensive linemen with over 38% of snaps played, while the Eagles have seven playing over 40% of defensive snaps. With Shamar Stephens out for their game against the Eagles having played 38.59% of their defensive snaps this season, the Vikings depth might be tested and this lack of depth on the line may show with a weaker pass rush in the fourth quarter, so that’s a storyline to watch.

The 2013 Seahawks set a blueprint with a remarkable eight defensive linemen playing between 46 and 58% of snaps, which may be the start of a trend as coaches understand that the explosive and violent nature of the position makes it vital to have multiple good players who can perform at their highest capabilities, rather than a couple better players who can’t perform at their highest capabilities due to fatigue. The 2016 Patriots had six defensive linemen play between 44 and 65% of snaps. The 2017 Eagles have seven players between 40 and 65% of snaps. The 2017 Jaguars had Calais Campbell, Yannick Ngakoue, and Malik Jackson all playing between 73 and 78% of snaps, Abry Jones and Dante Fowler, Jr. played 46.96% and 44.83% respectively, while Marcell Dareus has played between 43.5% and 66.2% of snaps in his last seven regular season games for the team. It’s critical to create depth across the line to execute this low-cost, potentially lower-performing, quarterback strategy as the goal is to nullify the other team’s quarterback. If you’re playing a team with an elite quarterback, a general vision of your goal is to decrease or slow down his performance to the point where your quarterback can be more productive and/or more efficient. Your quarterback doesn’t necessarily need to out-produce the player in yardage, but the goal is for him to be more efficient with a rushing attack at his disposal that the elite quarterback might not have and a better defense than that quarterback. We saw this strategy succeed for the Seahawks against the Broncos while Manning threw for 280 yards to Wilson’s 206 yards, Wilson had a better completion percentage and 2.5 more yards per attempt than Manning. Wilson had 8.2 yards per passing attempt, while Manning had 5.7 per attempt. The Seahawks defense also gave up just 27 rushing yards, while their own offense created 135 on the ground. While Manning was only sacked once, the Seahawks constantly moved him off his first read, disrupted his processing, and forced two interceptions.

The Eagles executed this style of decreasing the quality of play of the other team’s elite quarterback against the Falcons with Matt Ryan completing 22 of 36 passes (61.1%) for 210 yards (5.8 yds/attempt) and one touchdown, while their back-up Nick Foles completed 23 of 30 (76.7%) for 246 yards (8.2 yds/att). The effect of great coaching like Doug Pederson’s game plan cannot be understated. As Danny Kelly wrote for The Ringer, Foles “leaned on dump-offs, check downs, and run-pass options,” which helped facilitate the win and put Foles in his comfort zone. Continuing, he wrote, Foles’ stats were padded by receivers and running backs picking up yards after the catch and he had an average depth of target of just 5.2 yards per Pro Football Focus, which was almost two full yards short of any other quarterback that weekend. They hope he is able to do a little more against Minnesota this weekend with much more accuracy on deep passes than he showed against Atlanta, while the defense maintains the same kind of pressure they put on Ryan. Minnesota is happy to have Case Keenum and his 55.7% completion percentage under pressure, which was second in the NFL in 2017 behind Jimmy Garoppolo and slightly better than Tom Brady at 55.5%.

In looking at past champions during the research process of writing Caponomics, I found something that should be common sense, but we’ve lost sight of with our acceptance in the notion that you need a great quarterback to succeed at all costs, which has driven the price of the market up to heights that make it hard to compete for many teams. As seen above, Steve Young has a record cap hit of 13.08% and only seven of the 22 salary capped champions have had a quarterback over 10% of the cap, yet half the league seems to do it every year. What the research has taught me is that Joe Flacco at 14.70%, Kirk Cousins at 14.34%, Matt Ryan at 14.22%, and Carson Palmer at 14.45%–and if you include any other large cap expenditures with those players–creates a situation where these teams can only compete for a championship if they get some kind extreme, unlikely value out of other parts of their roster.

The 2016 Falcons almost pulled off a Super Bowl win with Matt Ryan and Julio Jones combining for what would have been a record-setting 25.54% of the cap, with the previous record being Steve Young and Jerry Rice at 21.64%, because they had a defense almost entirely filled with rookie contract players. The probability of hitting on as many rookie contract defensive players as they did is very low and it was the lack of depth on defense that ended up doing them in with the Patriots running 99 plays during that game and wearing them out by the time the final whistle blew.

It was very helpful that Dan Quinn and the Falcons organization was able to look at the 2013 Seahawks as their defensive prototype as well because they had player prototypes in their head that allowed them to build out their defense with a lot of success. Along with Kyle Shanahan’s elite offensive mind, they’re an example of the extreme value produced by great coaches. Their young defense improved as the 2016 season went along, but they were still ranked 27^th in points allowed, 25^th in yards allowed, 28^th in passing yards allowed, and 17^th in rushing yards, which was exploited by a Patriots offense that was third in points scored, fourth in yards gained, fourth in passing yards, and seventh in rushing yards.

The four remaining teams this year have all been competent passing the football, but–outside of the Patriots at second in passing yards– the rest of this group is outside of the top 10. The Patriots were 10^th in the NFL in rushing yards, while the Jaguars, Eagles and Vikings ranked first, third, and seventh with passing offenses that ranked 17^th, 13^th, and 11^th in yards produced. The Vikings ranked first in both points and yards allowed, while their defense was second in yards allowed both passing and rushing. The Jaguars were ranked second in both points and yards allowed, while ranked first in passing and 21^st in rushing, an issue that may have been remedied in a big way with the acquisition of Marcell Dareus mid-season. The Eagles were fourth in both points and yards allowed, while they were 17^th in passing defense and first in rushing yards allowed giving up just 79.2 yards per game on the ground. The Patriots were fifth in points allowed, but may have some defensive weaknesses being ranked 29^th in yards allowed with the 30^th ranked passing defense and the league’s 20^th ranked rushing defense. The Patriots did seem to right the ship a bit from a yardage perspective later in the season. Having the league’s fourth best redzone defense as well helped them.

Point being, all four of the teams left have complete rosters. Yes, each has some issues: the Patriots have some issues in yards allowed, the Jaguars have issues with Blake Bortles passing the ball, while Nick Foles and Case Keenum can perform, but leave us with some question marks heading into Championship Weekend. All four of these teams have coaches who have created strategies for success that can overcome the issues they do have and we’ll see on Sunday who can execute those strategies best. But none of them are really bad at any phase of the game.

My take is that the Patriots will move past the Jaguars as they will be able to make the Jags’ offense one-dimensional. While the Patriots don’t rank well from a yardage standpoint, they do have elite defensive backs who can cover a wide variety of match-ups, while the Jaguars don’t present match-ups in the passing game that should scare a backfield with McCourty, Patrick Chung, Duron Harmon, Stephon Gilmore, and Malcolm Butler. I think the Patriots offense uses their running backs extensively in the passing game as a means to beat the Jaguars pass rushers and avoid throwing at their elite cornerbacks. The match-up of Danny Amendola versus slot cornerback Aaron Colvin will be an x-factor in this game, while it would surprise me if Jacksonville or anyone else figures out how to stop a healthy Rob Gronkowski. We’ve seen the Patriots execute this quick passing game to success in past playoff match-ups and we’ll probably see it again on Sunday. I think the Jaguars keep it close for the first half, but the Patriots find some advantages they can take over the course of the whole game, as they usually do.

Both the Eagles and Vikings will be facing better defenses than the ones they faced last week, and that’s not to say the Falcons and Saints don’t have good defenses, but both the Eagles and Vikings have elite defenses. My take is that the Eagles will be able to produce enough pressure with their pass rush to decrease Keenum’s efficiency as, while he’s performed well under pressure, he hasn’t seen a defense that produces as much pressure as the Eagles do. With the passing attack slowed down, the Eagles will also be able to stop a rushing attack that, while it ranked well over the course of the season, doesn’t scare me with neither Murray or McKinnon averaging over four yards per carry and ranking 23^rd in the NFL in yards per attempt at just 3.9.

The Eagles on the other hand weren’t just third in the NFL in rushing, but they were also fourth in the NFL averaging 4.5 yards per carry. While the Vikings lost their best running back, Dalvin Cook, early in the season, the Eagles added their best running back, Jay Ajayi, midway through the year. Ajayi had 499 offensive yards in seven games and averaged 5.8 yards per carry. LeGarrette Blount had 766 rushing yards and a light workload for him with just 173 carries, which has likely kept him a little fresher as the season has gone into January and provided the team a great, power back to close the game out in the fourth quarter, a very valuable tool. Undrafted rookie Corey Clement is the final piece with 444 offensive yards this season as he really came on in the second half the year and showed considerable explosive quickness. He and Ajayi combined for 8 catches for 75 yards against the Falcons, which will likely be a key against the Vikings to give Foles high percentage completions. They’ll look to execute screens, swing routes, flat routes, and other quick passes. They may even look to hit one of these running backs on a wheel for a big play after Minnesota spent the week watching Philly’s backs catch short balls, which could be a big play Pederson has circled on his call sheet.

I imagine it will be a low-scoring game that’s won in the trenches; I think the Eagles have the better offensive and defensive lines. While Jason Peters is out, which has affected their performance, the rest of their offensive line has been elite all season with Jason Kelce, Brandon Brooks, and Lane Johnson being three of the best in the league at their positions. The Eagles defensive line will dictate the way the Vikings are able to play offense, while the Eagles offensive line will allow their offense the time to throw the quick passing game and the push running the football to control the clock. I don’t see the Eagles receivers and tight ends having a quiet night either, they have too much talent across the offense for Minnesota to stop everyone–even with Foles at quarterback. That said, the Vikings are the NFL’s best defense and Case Keenum has had a very efficient season. This game is more of a toss up than the AFC match-up, but I’ll take the Eagles.

Zack Moore is a writer for OverTheCap.com, author of the recently released book titled, “Caponomics: Building Super Bowl Champions,” and an NFLPA Certified Agent. You can follow him on Twitter @ZackMooreNFL.