Tuesday Morning Quaterback: It Doesn't Pay To Punt
Sept. 26, 2006 — -- Once again this weekend the NFL landscape was littered with Preposterous Punts. Trailing 24-3, San Francisco punted on fourth-and-1 on the Philadelphia 40. Even the great Bill Belichick ordered a punt from the Broncos' 35. As this column repeats ad infinitum (Latin for "by using AutoText"), NFL coaches punt in opposition territory, or on short yardage, in order to avoid blame -- if a team goes for it and fails the coach is blamed, whereas if a coach does the safe thing and kicks and then loses, the players are blamed. But skip the psycho-dynamics and ask: Should a football team ever punt?
A year ago at the Hall of Fame reception in Canton, Ohio I found myself sitting between Bill Walsh and Don Shula. I posed this question: In a day when the Bears line up five-wide and Texas Tech passes 60 times a game, are there any fundamental innovations that have not been tried? Walsh supposed someone might try using trick formations for an entire game. Shula twinkled his eyes and said: "Someday there will be a coach who doesn't punt."
Think about all those punts on fourth-and-1, fourth-and-2, fourth-and-3. The average NFL offensive play gains about five yards. Yet game in, game out, coaches boom the punt away on short yardage, handing the most precious article in football -- possession of the ball -- to the other side. Nearly three-quarters of fourth-and-1 attempts succeed, while around one-third of possessions result in scores. Think about those fractions. Go for it four times on fourth-and-1 -- odds are you will keep the ball three times, and three kept possessions each with a one-third chance of a score results in your team scoring once more than it otherwise would have. Punt the ball on all four fourth-and-1s, and you've given the opponents three additional possessions. (It would have gotten one possession anyway when you missed one of your fourth-and-1s.) Those three extra possessions, divided by the one-third chance to score, give the opponent an extra score.
Bottom line? If you face fourth-and-1 four times and punt all four times, your opponent will score once more than it otherwise would have. If you go for it all four times, you will score once more than you otherwise would have. (These are simplified probabilities that do not take into account that the one-score-in-three figure assumes most teams voluntarily end drives by punting on short yardage; subtract those punts, and a possession becomes more valuable because a score is more likely to result.) Few teams face fourth-and-1 four times in a game, but the numbers for fourth-and-2 and fourth-and-3 work out about the same, and most teams do face fourth-and-short several times per game. Probabilities suggest a team that rarely punts will increase its scoring while decreasing its opponents' point totals.
Think I'm crazy? Let's turn to this 2005 paper by David Romer, a professor of economics at the University of California at Berkeley. Romer's work got attention from the sports media because he contends teams facing fourth-and-goal should almost always try for the touchdown. I'm not so sure, and will address that in a later column. (Short version of my counterargument: Field goals are nothing to sneeze at.) But there is gold, absolute gold, in the overlooked later pages of Romer's study. His numbers say that anytime the situation is fourth-and-4 or less, teams should not punt. Romer thinks teams should try for the first down on any fourth-and-4 or less even when in their own territory. After all, the average play gains almost five yards. On average you will retain possession, and the pluses of that exceeded the minuses of the inevitable failed rth-down try
Romer put the opening quarters of all NFL games from 1998 to 2004 into a database, then analyzed when coaches ordered punts, when they went for it, and how these decisions had an impact on field position on subsequent possessions. Here are Romer's three key conclusions. First, inside the opponent's 45, go for a first down on any fourth-and-7 or less, unless a field goal would decide the game. Second, inside the opponent's 33, go for a first down on fourth-and-10 or less, unless a field goal decides. In Romer's sample years there were 1,068 fourth downs in which the above formulas said go for the first down, yet NFL coaches kicked all but 109 times -- meaning they went for it only about 10 percent as often as they should have. Finally, Romer's numbers say that an NFL team should try for the first down on any fourth-and-4 or less, regardless of where the ball is on the field. Of course some fourth-down tries would go down in flames and even create easy scores for the other side. But over the course of a season of rarely punting, Romer maintains, the team that eschewed the punt would score more than it otherwise would, while its opponents would score less.
Suppose an NFL or major-college coach came into a season determined to go for it any time it was fourth-and-4 or less. I don't think a coach should be doctrinaire about this. I'd punt if it was fourth-and-4 inside my 20, and I'd be inclined to punt in the second half if protecting a lead. But otherwise, the coach commits to going for it instead of punting, even if the first few attempts backfire. Surely a strategy of rarely punting would sometimes boomerang, but on balance it could lead to more scoring for your team while depriving the other team of the ball. The strategy could cause exhaustion and panic on the parts of defenses that thought they had done their jobs by forcing fourth down, only to discover your offense had no intention of passively jogging off the field. Teams that rarely punted might pile up big advantages in points and time of possession. If Don Shula's "coach who doesn't punt" appeared on the NFL scene, that coach, Tuesday Morning Quarterback suspects, would revolutionize football. Player talent being equal, that coach might blow the doors off the National Football League.
Which leaves us with the question of whether the coach conjectured by Shula could ever exist. Such a coach would need to be completely unconcerned with the media and owner backlash that would follow a loss caused by a no-punt policy. Such a coach would need to be fearless, and financially independent. Will there ever be such a coach? Tuesday Morning Quarterback wonders. But next time it's fourth-and-3 and you hear the announcers say "now they have to punt," just remember: No, they don't have to punt.
In other football news, all hail the United States Saints! That's what TMQ called the team last season during its wanderings, and surely they were the United States Saints last night upon their triumphal return to New Orleans. The emotion of the event was powerful, but the Saints played so well it forced one to wonder: Maybe this is actually a top team. How fitting if the football gods repaid the Saints' horrible 2005 with a wonderful 2006. Between Drew Brees and Reggie Bush looking so, so good while Daunte Culpepper and Mario Williams look so, so bad, there must be serious buyer's remorse today in Miami and Houston. The Saints' early blocked-punt touchdown was not only a sweet play but conformed with TMQ's immutable law of punt defense: Send Eight to Make a Punt Go Backward. As this column has pointed out before, NFL coaches rarely send more than five rushers after the punter. It's blame-shifting: if the coach calls an all-out rush and the kicker is roughed then a coaching decision is blamed, whereas if there's a light rush, a return and the offense fails to move the ball, then the players are blamed. On the first Atlanta punt, New Orleans overloaded the line with eight men in tight, and all eight came: Thunk! For the rest of the contest, New Orleans rushed only five. NFL special teams coaches, take heed.
In other news, they were booing in Foxborough as the Patriots left the field at halftime. Sure, New England has won three of the last five Super Bowls. But what have you done for us lately! Meanwhile, Ben Roethlisberger has lost two consecutive games and threw three awful interceptions on Sunday. Now he's only 28-6 as an NFL starter. He's really tailing off! How soon till Steelers fans start booing Roethlisberger?
Stat of the Week No. 1: Baltimore and San Diego have outscored their opponents by a combined 110 points.
Stat of the Week No. 2: At 1:59 p.m. ET on Sunday, almost an hour into their third game of the season, the Buccaneers scored their first touchdown of the 2006 season.
Stat of the Week No. 3: Because the Raiders had a bye, it will be October before Oakland scores its first touchdown of the 2006 season.
Stat of the Week No. 4: Seattle is on a 23-4 streak at home.
Stat of the Week No. 5: At 11:22 p.m. ET on Sunday, the Broncos allowed their first touchdown of the 2006 season. Considering an overtime, the Denver defense played 12 consecutive quarters without allowing a touchdown.
Stats of the Week No. 6: At 8:53 p.m. ET on Monday, the Falcons allowed their first touchdown of the 2006 season.
Stats of the Week No. 7: At the end of the first quarter in Seattle, Eli Manning had two interceptions and minus-12 yards passing.
Stats of the Week No. 8 : The Giants have not won in Seattle in 25 years.
Stats of the Week No. 9 : Stretching back to Jan. 1, when the Buccaneers won the NFC South, Tampa has lost four straight.
Stats of the Week No. 10 : The Saints are 3-0 for the first time since 1842.
Cheerleader of the Week: Vincent Hendricks of Houston nominates Summer of the Texans' pep squad. According to her team bio, Summer "works as an engineer at NASA's Johnson Space Center." A rocket scientist cheerleader! She also has a VFR pilot's license and is working on her instrument rating. If her team photo is an indication, Summer is not only a rocket scientist cheerleader, she's a sultry rocket scientist cheerleader. This really must be the third millennium!
Sweet Series of Plays of the Week: On "series" plays, one action sets up another. Game tied at 7, Indianapolis was on its own 31. Marvin Harrison lined up wide right, with tight end Dallas Clark also on the right as a slot receiver, and ran a middle crossing pattern as Clark ran an out; Harrison caught a pass for 38 yards. Two plays later the Colts lined up the same way. This time Harrison came in motion back toward the formation. At the snap, Harrison headed for the middle cross again, but from the inside of Clark, while Clark ran for the sideline and tailback Dominic Rhodes ran a flare right. Two Jax defenders went with Harrison. Two came up to cover Rhodes. That left no one on Clark; Peyton Manning saw this and motioned Clark up the field, where he caught a 30-yard touchdown pass.
Sweet Play of the Week No. 2: Trailing Pittsburgh 17-14 early in the fourth quarter, Cincinnati faced fourth-and-1 on its own 30. Following professor David Romer's advice, the Bengals went for it and converted. Cincinnati did not score on the possession, but Marvin Lewis' decision to go for it on his own 30 communicated to his charges that he was challenging them to win the game, which they did.
Sweet Play of the Week No. 3: Leading 3-0, Denver had third-and-1 on the New England 32 with 56 seconds remaining in the first half. Normally it's best to rush on short-yardage downs: but if you're going to pass, throw the home run, not some dinky three-yard out. The Broncos play-faked, Javon Walker went deep up the right sideline single-covered and caught the touchdown that made it 10-0 at the half. Flying Elvii defensive backs -- there were only 56 seconds remaining, why were you surprised Denver went to the end zone?
Sweet Play of the Week No. 4: On a 23-yard completion to Muhsin Muhammad, Rex Grossman of the Bears play-faked left, then play-faked right, then threw. You don't often see two play-fakes on the same down. But was this pass-wacky unit really the Chicago Bears? At one point in the game, the Bears' coaches had called 26 passes versus eight rushes.
Related Stories
ABC News Live