Walkthrough: Playing it Safe
by Mike Tanier
You just gained nine yards on first down. It's second-and-1. Run it up the middle for a safe first down? The heck with that! Fake the hand-off, send the receivers deep, and take a crack at a big play. If it fails, you can run on third down. Of course, the defense knows you want to do that, so they keep some safeties deep. Maybe it's better to just concede that inside run than give up a game-changing bomb.
Second-and-short is thought of as a "house money" situation, but most teams are surprisingly conservative with a yard or two to go and three downs to get there. The play-action bomb on second-and-1 is a very uncommon play. Using our Game Charting Data, we can learn just how often it's used and how successful (or unsuccessful) it is.
First, the data set: I am including both second-and-1 and second-and-2 plays in the sample. From three yards out, a conversion on third down can no longer be considered "automatic." (It's not automatic in any case, but third-and-3 is overwhelmingly a passing down, so coordinators think of it as a different situation than third-and-1 or -2.) I also took out the red zone plays, because you cannot throw a bomb from the opponent's four-yard line.
There were 754 second-and-short plays that met our criteria last year. Of those plays, 501 were runs, not counting scrambles. That means teams ran 67 percent of the time, and that's without factoring in the more conservative strategies you find in the red zone. Forget gambling -- teams play it safer on second-and-short than they do in almost any other down-distance situation.
Of the passes thrown on second-and-short, only 51 traveled 15 or more yards in the air and only 40 of the total passes involved play action. The play-action percentage (5.3 percent) is amazingly small in a situation where a run fake seems beneficial. Remember, opponents are expecting the run, because that's what happens two-thirds of the time. Play action appears to be a little underused.
On deep passes, teams were 20-of-51 for 505 yards, one touchdown, and six interceptions. Ouch. That high interception total may explain why the short-yardage bomb isn't used often: Safeties are looking for it, cutting the reward-risk ratio way down. On all passes, teams were 148-of- 239 for 1,480 yards, three touchdowns, eight interceptions, and eight sacks, with a few scrambles and an intentional grounding foul by Rex Grossman rounding out the data. The completion percentage on second-and-short passes is 61.2 percent, fairly close to the league average. All but 10 of the 148 completions netted first downs: there were three fumbles, two one-yard catches on second-and-2, a few tunnel screens for losses, and a zero-yard pass from Drew Brees to ... not Reggie Bush! David Thomas was the culprit this time!
|Figure 1: Eli Manning Play-Action Pass|
As for rushes, those 501 running plays netted 2,215 yards for 4.43 yards per carry, a significantly higher rate than the league average. Running plays yielded 364 first downs, a 73 percent Success Rate. So the offensive coordinator could call a play that works 73 percent of the time, or he could call a pass that yields a first down about 60 percent of the time, or he can make the "bold" call to throw deep and deal with an interception rate just less than 12 percent. No wonder the play-action short-yardage bomb is definitely a pick-your-spots call in modern football.
The longest second-and-short bomb of last season was an Eli Manning pass to Kevin Boss against the Titans. It wasn't a play-action pass, but as Figure 1 shows, the Giants were in an I-formation, and the two backs released at the snap as if the play were an off-tackle run. The Titans blitzed a safety on the play, and when they telegraphed the blitz by putting the safety on the line, Manning called an audible. It's pretty clear from the diagram that Kevin Boss will be isolated in man coverage against a linebacker, so Manning figured he could hit Boss on an angle route, and he was right. The catch-and-run yielded 54-yards, but the pass itself traveled just 17 yards, as Manning hit Boss in the window before the deep safety could arrive for help. The fact that the huddle call was probably a run tells us a lot about how the bomb strategy is used. It is held in reserve until a coordinator or quarterback sees a clear mismatch that can be exploited.
A few other notes:
- The only touchdown scored on a second-and-short bomb was a 23-yarder from Shaun Hill to Calvin Johnson. Note that we are using the term "bomb" loosely on a lot of these plays. Here's why:
- On passes that traveled 30 yards or more in the air, quarterbacks were 1-for-12 for 53 yards and three interceptions. The one successful play was a Michael Vick-DeSean Jackson 53-yarder. Vick also threw two incomplete long bombs in this situation; no other quarterback threw more than one of these passes. The play-fake and long, arching mega-bomb is rarely called and has a low success rate.
- Vick and Matt Ryan each threw five deep passes (we are back to 15+ yards now) in this situation. Ryan was 2-of-5 with two 19-yard completions to Roddy White. Four of his five attempts were to White. Paging Julio Jones!
- Joe Flacco completed just one 27-yard pass in his only attempt in this situation. Play-action bombs on second-and-short sound like the ultimate Ravens strategy, but it rarely happened last year.
- The Game Charters listed 18 "dump-offs" in this situation for 89 yards. Only three of these passes involved play action, but several of them no doubt occurred when the quarterback didn't like his deep option and bailed on the play. Even if you classify these dump-offs among the deep passes to represent plays in which the quarterback could have gone deep, the strategy is still not very prevalent or popular.
Quarterback Top Fives: AFC South
The Oilers/Titans and Colts both have a lot of Hall of Famers to sort through, plus some near-HOFers and cool 1970's guys who give us a lot to talk about. So let's split them up into two weeks and pair each with one of the newbie franchises in the division. The Colts come next week. This week, the Oilers-Titans have a Jaguars undercard.
1. Mark Brunell. If Blaine Gabbert starts on opening day and has productive seasons immediately, he will need at least five years to threaten Brunell; realistically, Brunell will be safe at the top of this list for the rest of the decade. Brunell had a five-year run of excellence from 1996 to 2000 that any quarterback would be proud of, plus a couple of hang-around years in Jacksonville in which he was still very good. It's nowhere near a Hall of Fame resume, but there are guys who ranked third or fourth for historic franchises who would kill for it.
2. David Garrard. He is a consistently ordinary quarterback who fooled a lot of people with his three-interception performance in 2007. Exceptionally low interception rates are not a sign of greatness, but of Neil O'Donnell-ness. Jason Campbell is also near the bottom of the all-time interception percentage rates, and Garrard is like a Campbell who runs a little better and plays for an organization that doesn't blame all the world's problems on its quarterback.
3. Byron Leftwich. For the first three years of his career, no one talked about Leftwich's funky delivery except Mike Mayock-types. By 2006, it was all anyone could talk about. It is one of the oddest deliveries in modern football history, and it was immediately recognizable as a problem, but there are lots of people who talk and write about football who won't notice something like that until it is pointed out many, many times -- and will then fail to notice anything else.
4. Quinn Gray. Brunell, Garrard, and Leftwich combined to throw 90 percent of the passes in Jaguars history. Gray threw 12 touchdowns and five interceptions in a handful of appearances, looking like an up-and-coming prospect at times. Garrard beat him out for the right to replace Leftwich, and Gray proved that a two-game run at the end of the 2007 season was actually his peak.
5. Blaine Gabbert. Gabbert needs two good games to crack the list and half a productive season to blow past Gray.
Houston Oilers/Tennessee Titans
1. Warren Moon. The Oilers were a balanced team in Moon's early seasons. In fact, they were almost run-oriented. Earl Campbell was still staggering around during Moon's "rookie" year, and the Oilers spent the next few seasons drafting some of the most decorated runners in college football in search of the next rushing superstar. They took Heisman Trophy winner and South Jersey tough guy Mike Rozier in the first round of the 1984 Supplemental Draft (which was really a USFL rights draft). Allen Pinkett, a Notre Dame scat back, arrived in 1986. The team drafted fullback Alonzo Highsmith (really an all-purpose power back) in the first round in 1987 and Lorenzo White in the first round in 1988. All four of these guys could play: Rozier and White both had 1,000-yard seasons, Pinkett was productive, and Highsmith rushed for 466 and 531 yards in an era when fullbacks were going the way of the passenger pigeon. Moon rose to Pro Bowl status as the trigger man for a relatively old-fashioned offense. The Oilers threw 428 passes and ran 558 times in his first Pro Bowl season, though their splits in Moon's other seasons weren't as run-heavy.
And then the run 'n' shoot arrived. Adopting the offense was like discovering that the Tupperware meeting your spouse went to was really a cult indoctrination. She comes home wearing a white robe, throws away the television, and declares that from now on your lives will be devoted to planting beans and discovering the downside of a communal marriage. One minute, the Oilers had a fullback who carried the ball regularly and some tight ends on the roster. The next, everyone involved in the offense was a 172-pound wide receiver.
Moon put up some staggering numbers while throwing to these shrimps. He also fumbled 18 times in 1990, his first year in the offense. Moon's quick release, pocket presence, and leftover running ability kept his sack totals low, and his numbers from 1990 and 1991 are staggering. Still, it's hard to believe anyone thought an offense system this doctrinarian was a good idea. This was the era of Buddy Ball and 46-defenses, a time when sack totals were just starting to dip after a long peak in the mid-1980s. When Jack Pardee suited up his seventh 175-pound wide receiver, didn't he think for a moment that it was overkill? That the Oilers might be better off with a 250-pound tight end to block in short-yardage situations, or against the Eagles and Bears?
One thing I noticed when looking through the Oilers run 'n' shoot statistics is how little they got from their return game in those years. McNeill and Pinkett were very good return men, and a team full of Smurfs should be able to find someone who can break some long returns. The Oilers didn't return a punt or a kickoff in the era, and their per-return averages in that era are pretty terrible. It's almost as if they had no one who could block for returns. It all added to the perception that run 'n' shoot teams piled up yardage but could not score touchdowns: Their average drive started a few yards back, forcing them to go farther for every point.
When you look at Moon's stats from 1992 and 1993, you can see how quickly diminishing returns set in for the run 'n' shoot. Moon and the Oilers were a good team despite their gadget offense in those seasons, not because of it. All of the smart coaches grabbed bits and pieces of the run 'n' shoot, adapting it to their needs without sending all of the fullbacks and tight ends to the unemployment line. The cultists were left throwing passes to Gary Wellman. Moon's numbers rebounded sharply when he left Houston for a more conventional Vikings offense. His numbers and reputation were both helped and hurt by the run 'n' shoot: The totals are terrific, but they come with a grain of salt, and the interception percentages and other rates would have been better in a less experimental system.
2. Steve McNair. I was tempted to rank McNair ahead of Moon. Moon is a Hall of Famer, but his Canton credentials are bolstered by a) his years with the Vikings, b) the numbers he put up in the run 'n' shoot, c) his historic significance as a trailblazing black quarterback, and d) the fact that he played forever and sportswriters came to love him as an elder statesman of the game. McNair will never sniff the Hall of Fame, but things might have been different if Kevin Dyson broke a tackle or if he chose his paramours more prudently.
McNair had a lot of near-Pro Bowl seasons, like 2002, when he threw for 3,387 yards and 22 touchdowns on an 11-5 team (rushing for another 440 yards) but lost a berth behind Peyton Manning and Rich Gannon to Drew Bledsoe, who threw a zillion passes for the Bills. It was odd; Warren Moon couldn't get out of the Pro Bowl if he tried, reaching Honolulu in mediocre years like 1988 and 1992. Vince Young got to the Pro Bowl every time he made it few some starts without getting benched. McNair made it just three times but could well have made it another three times.
I get the feeling that McNair will be like Roman Gabriel or Frank Ryan in a few decades, a player who is nearly forgotten. A few of you ranked McNair ahead of Moon when I posed the question on Twitter, so I know it is not foolish to think McNair was better. Ultimately, McNair didn't do enough to justify first place, particularly in the playoffs, when he had a knack for 112-yard passing days. But it was close in the same way that picking Gabriel over Hall of Famers for the Rams was tough.
3. George Blanda. Blanda and Moon are alike in many ways. Both waited for years for a starting opportunity. Both put up huge numbers that require an asterisk for a few years. Both hung around forever as lovable backups. Both are in the Hall of Fame partly (in Blanda's case, largely) due to extenuating circumstances. Moon paved the way toward acceptance for black quarterbacks. Blanda captured the imagination of a generation by playing into his late-40s.
Blanda threw 35 touchdowns for the AFL Oilers in 1961, then spent four seasons leading the league in interceptions, throwing 124 of them in a 52-game stretch. The Oilers were great in 1961 and 1962 but pretty bad afterward, and Blanda wound up in Oakland as a kicker by 1967. This sounds like an example of a veteran journeyman dominating weaker competition for a year, fading rapidly as the league around him improved, and being relegated to a specialist role once a common draft began rapidly leveling the NFL and AFL. Seventy percent of Blanda's passing touchdowns and 76 percent of his yards came from his pre-merger, pre-common-draft Oilers career. It's a dubious centerpiece for a Hall of Fame resume.
Arguing against Blanda makes me sound like Scrooge, and I don't want Raiderjoe to come and kill me. I remember exactly how famous he was in the mid-1970s, and his Namath-like status as a transcendent character and nugget of NFL history certainly count for something when ranking him as a Hall of Famer (as opposed to a great quarterback). Doug Flutie has a lot in common with Blanda, in that his every move was fawned over after a while; Flutie, Moon, and Blanda constitute their own sub-class of "late-career hero" quarterback. (Jeff Garcia also qualifies; perhaps you can name a few more). To argue backward from Blanda's Hall of Fame induction to prove that he really was a great quarterback is to overlook a lot of pretty obvious facts.
4. Dan Pastorini. Pastorini had a two-year stretch in which he was 2-21 as a starting quarterback, throwing 12 touchdowns and 29 interceptions. Can you imagine a quarterback holding onto his starting job after a stretch like that? Pastorini didn't; Lynn Dickey replaced him in 1974 but was awful, and Pastorini got slightly hot when he won his job back.
The Oilers were a terribly run organization from 1970 to 1973. Bud Adams and Sid Gillman were in charge, and while both of them know a thing or two about football, they flailed around a bit after the merger. They signed coach Bill Peterson to what they called a "lifetime contract" in 1972; Peterson was an innovator at Florida State in the 1960s, but he was running an undistinguished program at Rice University when Adams pegged him. The "lifetime contract" lasted a littler more than one year before Gillman came down from the front office to replace Peterson. A guy named Ed Hughes also coached the Oilers briefly. He was apparently a well-respected coordinator, but Adams and Gillman fought with him over assistant coaches and quarterbacks, so he only lasted a year. It's a miracle that Pastorini survived the turmoil, but that was common in football before free agency, when teams could keep a player around for years and years without worrying about something silly like an expired contract or a competitor with a better offer.
Pastorini punted full-time until 1976. Punting quarterbacks took a long time to go totally extinct. Danny White was still doing both jobs in the early 1980s, Randall Cunningham punted eight times for the Vikings in 1997, and Tom Tupa started 13 games for the Cardinals from 1989-91 and still took preseason snaps at quarterback for Bill Parcells' Jets. By Pastorini's time, it was becoming uncommon, and most teams employed either a full-time punter, a kicker-punter (already becoming rare), or a guy like Bob Parsons who was listed at tight end but made a career of punting. It is fascinating how long it took full-scale specialization to catch on, though, especially in 1970s football, when field position mattered so much. Punting was an incredibly important part of a quarterback's job in early football, and teams had no problem expecting a starting or backup quarterback to punt, even in a very modern era. (Blanda, or course, was expected to back up Ken Stabler while kicking, even when he was 45.)
Pastorini was not a great quarterback or punter, but he excelled at handing footballs to Earl Campbell and watching the ensuing carnage. That gave him a long career in Houston.
5. Vince Young. Some might argue for Ken Stabler here. In 1980, Campbell rushed for 1,900 yards, the Oilers finished 11-5, and Stabler threw for just 12 touchdowns and 28 interceptions. In 1981, he was just as interception-prone and even more sack-prone, and Campbell began wearing down, taking the Oilers with him. Stabler's seasons were a lot like Vince Young's seasons, and Stabler was about as unreliable as a person as Young has been, though it was considered charming when he did it. Young deserves this ranking. Jake Locker can rocket into fifth place with a few seasons of dependability; after that, the climb gets tougher.