By Tej Seth
Since the Moneyball renaissance in baseball was so successful for the teams that leaned into using analytics, it was only a matter of time for teams in the NBA and NFL to apply the same principles to their sport. The NBA jumped on it quickly with two major movements relating to shot selection since the turn of the century. In the year 2000, the NBA league average for a team’s percentage of three-point attempts from the corner was 23.9%. Gregg Popovich’s San Antonio Spurs, who would amass five NBA titles by 2017, attempted 34.5% of their 3-point attempts from the corner. Whether the Spurs were data savvy or figured this out intuitively, corner threes generated an average of 1.16 points per shot while other threes averaged 1.05. The corner three, which boasts the shortest distance between the shooter and the hoop at any point along the three point line, has spread across the NBA game to the point of people advocating for adjusting the three point line. Other teams within the league followed in the Spurs’ footsteps by increasing their corner three frequency.
The other movement gained momentum around 2010 as teams traded in the notably inefficient midrange jumper for a threes and lay-ups approach. Defenses, as well as a 24-second shot clock, work in tandem to prevent a team from shooting efficient shots on every possession. As the clock ticks closer to zero, teams are stuck hurrying into their last, and generally least-efficient, option. Insert superstars with unique talents like Kevin Durant and Kawhi Leonard and you have players that rise above the rest. To be a truly top-tier scorer, one must be more efficient than the average player in conditions where the defense has “won”, statistically speaking, the possession.
Quarterbacks, like the NBA’s crunch time scoring leaders, work to find the best opportunity with a seemingly accelerating (internal) clock. Passes (excluding throwaways) can be grouped into three categories from the offense’s perspective: left, middle and right. Passes over the middle of the field in football have become what corner threes are in basketball. These throws have been shown to be both more efficient and more available in favorable situations. Passes out wide, in contrast, are akin to the midrange shot. These difficult throws represent more difficult situations that requires players performing at their best.
Middle of the field passing, despite being the most efficient area of the field for an offense to pass into, can be difficult to execute as it may require the quarterback to hold onto the ball longer to get through progressions or be an area of the field where a safety is present. Effective play action and RPO’s can help the quarterback get some lay-ups as defenses play with the middle of the field open more often, but defenses tend to push the offense into more difficult throws. There could be a selection bias here, though, as quarterbacks might get sacked more on plays where their progression takes them to wanting a middle of the field throw, but they are not able to get the throw off in time. We can see that this not wavered over time as, regardless of pass depth, the middle of the field seems to be the area where the offense can generate the most Expected Points Added (EPA) per pass attempt.
Looking at this from a season-level, passes over the middle of the field have consistently produced a higher EPA per play than passes thrown to the left or right. Middle of the field passes, like the NBA’s corner three, have also become some of the least frequent pass location despite its high efficiency.
Every season from 2015 to 2020 (with the exception of 2017 partially due to multiple big quarterback injuries) saw middle of the field passing clear 0.25 EPA per pass. A season long passing attack of 0.25 EPA per pass would have been the 2nd most efficient in the league during the 2022 season. Defenses have deployed different ways to deal with middle of the field passing and the EPA per pass of such throws has been on a steady decline.
Weak side rotation 1 cross working out exactly as designed:
-pass crosser to spinning safety
-DB that passed the route off becomes the rat
-new rat finds work off the QB’s eyes pic.twitter.com/go4E2vwzE6
— Shawn (@SyedSchemes) November 8, 2021
Despite the defense’s best efforts, the league’s best play callers still know how to give their offense a boost by hitting the “easy buttons.” Previous research has shown how play-action passes can be particularly explosive given the linebackers natural desire to fit the run fake opening up the middle of the field to both a higher chance of completion and a higher expected yards after catch.
Jefferson on Drift Swirl. Vikings in 21p get the Giants in base on 2nd and 3https://t.co/LnN8x42sJ8 pic.twitter.com/dATZp9UXaa
— Shawn (@SyedSchemes) December 27, 2022
Mike McDaniel schemed up passes over the middle of the field on 28% of Tua Tagovailoa’s pass attempts in 2022 using run-pass options to move linebackers from their spots and getting the ball into the hands of Tyreek Hill and Jaylen Waddle. Kyle Shanahan kept the rate of middle of the field passes consistently high as the offense shifted from Jimmy Garoppolo to Brock Purdy. Lamar Jackson is another interesting quarterback to take note of as his gravity in the run game can open up the middle of the field more than most quarterbacks.
working through some stuff for a paper right now
GCOE is "ground covered over expected" – change in Euclidean distance between RB and LB/SS after two seconds, adjusted for situation. Against these offenses in 2021: pic.twitter.com/GzeGE8Smpk
— Eric Eager 📊🏈 (@ericeager_) May 11, 2022
In 2022, there was a correlation between how often a quarterback threw over the middle of the field and their EPA per pass overall. A team like the Chargers could hope hiring Kellen Moore increases their pass rate over the middle of the field.
Play callers need to pick and choose their spots of when to throw over the middle of the field considering the game situation, matchups, and their own personnel. The Chiefs, often found at the top right of these graphs, do not need to rely on middle of the field passing as Patrick Mahomes has unmatched efficiency. The Bills and Jaguars take advantage of Josh Allen and Trevor Lawrence’s respective arm strength to attack out wide as well.
Quarterbacks that are considered in the “middle tier” such as Ryan Tannehill, Derek Carr, and Matt Ryan hit their lay-ups and are above-average in middle of the field production while being below-average in out wide passes. As shown below, their offenses also don’t require those types of passes as often.
Much like the superstars who succeed in the midrange in the NBA, the NFL has elite quarterbacks who make the tough shots look easy and succeed throwing out wide. The group of Patrick Mahomes, Joe Burrow, Josh Allen, and Trevor Lawrence make up some of the best quarterbacks in the league and are also those who are asked to pass out wide more often than the average quarterback. Their unique skill sets can make their play callers feel more comfortable calling these plays because of how efficient they are at executing them.
Quarterbacks seem to have more control over passes out wide while scheme and weapons seem to play more of a role in passes over the middle of the field.
On a year-to-year basis, there is around twice the amount of stability for how quarterbacks perform passing out wide as opposed to over the middle of the field. This supports the original hypothesis that middle of the field passing can be influenced more by scheme and supporting cast while quarterbacks have more control over their passing out wide. As the surroundings change, middle of the field efficiency can change as scheme and yards after the catch play a role in how effective these throws are. Finding a quarterback that can be effective on passes out wide and scheming passes over the middle of the field might be a plan teams could follow as they look to build on what they already have in the passing game.
We can learn about what a quarterback is asked to do based on the situations quarterbacks are asked to pass in. Based on down, yards to go, yard line, score differential, and time remaining in the half and the game, a chance that a team passes can be given for any play. If there is a 70% chance of a pass and a team passes, they are given a 30% pass rate over expected on that play.
Taken together, we can see the situations elite quarterbacks live in. They are effective passing out wide and when the defense expects the pass. These are the midrange jumpers of the NFL – reserved only for the best players in the most high-leverage situations. Teams with top end quarterback talent can continue to hit their midrange shots while teams without these quarterbacks may have look elsewhere. Manufacturing space in the middle of the field through play action is one way for these teams to generate production. Unfortunately, the shot clock winds down sometimes and situations can arise where the teams that cannot generate middle of the field passes have their limitations exposed.