What are the odds that Super Bowl XXXIX between the New England Patriots and the Philadelphia Eagles will be a dud? History says they’re pretty good: Eleven Super Bowls have been won by 21 points or more; 18 have been decided by 17 points or more.

To look at it another way, only 11 of the 38 Super Bowls since 1967 have come down to a touchdown or less. If the Super Bowl is, per its billing, the “ultimate football game,” why are so many of the games so one-sided?

First of all, let’s dispose of the myth that the Super Bowl matches up the two best teams. That’s certainly not true this year, when the NFL’s three or four best teams were in the AFC. Nor has it been the case in the past.

For instance, from the 1981 season, which ended with the San Francisco 49ers beating the Cincinnati Bengalis 26-21, through the 1996 season, which closed with the Green Bay Packers beating the New England Patriots 35-21, the NFC won 15 of 16 Super Bowls. In most of those years, fans recognized that the NFC Championship Game was, for all intents and purposes, the Super Bowl.

But that doesn’t fully explain why some Super Bowls are so bad. Even when the best teams are matched, the game often fails to deliver. Two years ago, for instance, the Oakland Raiders and Tampa Bay Bucs should have provided the most competitive big game in recent years – the teams were rated just about dead even in most major statistics. But after the opening kickoff, Oakland was never in it, succumbing 48-21.

Oakland’s embarrassing showing in Super Bowl XXXVII gives us a clue as to how good teams go bad. Tampa Bay head coach John Gruden had been head coach at Oakland and was succeeded by his offensive coordinator, Bill Callahan. When Gruden left for Tampa Bay, he brought the Oakland playbook with him. Whereas Gruden modified the playbook slightly, adding some sprint-out passes that suited the talents of quarterback Brad Johnson, Callahan maintained the same offensive scheme in Oakland.

By Super Bowl time, the Bucs were as familiar with the Raiders’ offensive schemes as was Raider quarterback Rich Gannon, who had five passes intercepted, three of them returned for touchdowns. After one pickoff, a Bucs linebacker was overheard on national TV telling Gruden, “They’re doing everything we worked on in practice.”

It seems incredible that a coach capable of taking his team to the Super Bowl would not think to change a couple of plays when facing his former boss. How can what seems so obvious to all other observers not be obvious to the head coach?

“I’ll tell you what goes wrong sometimes,” Joe Gibbs explained to me in an interview for the Sporting News in 1992. “You fall in love with your game plan. I’ll give you an example. Back in 1983, we beat the L.A. Raiders during the regular season. In that game, we ran a screen pass deep in our own territory for a big touchdown. I was absolutely convinced that given their aggressive tendencies, we could run the same play again.

“Well, in the Super Bowl, we ran it from about our 5-yard line, they intercepted, and ran it into the end zone. It set the tone for the whole game. At halftime, my coaches were all saying our read on their tendency isn’t right, they’ve adjusted. And I kept saying, no, we’ve got to have faith in the plan we came in with.”

The result? The Redskins lost, 38-9.

“I made up my mind then and there,” Gibbs said, “that that was never going to happen again to me, that we’d have a Plan A based on what we thought our opponents’ expectations would be. But I’d be prepared with a Plan B and even a Plan C.”

Gibbs’s Redskins won their next two Super Bowls, 46-10 over the Atlanta Falcons in 1988 and 37-24 over the Buffalo Bills in 1992. Nor was Gibbs the only winning Super Bowl coach to learn the value of a flexible game plan.

“I think you used to see more competitive Super Bowls back in the early years when the quarterbacks had more freedom,” Bill Walsh told me a few years ago. “Starting around 1982 or ’83, we all started hiring more coaches and the entire system became much more complicated. Now, you have to get the entire brain trust together to work out a change on the sidelines.”

Walsh may have something. If one discounts the first two Super Bowls, in which superior Green Bay Packers teams routed their AFL foes by 25 and 19 points respectively, only one game between 1969 and 1982 – the Cowboys’ 24-3 dismantling of the Dolphins in 1972 – was won by more than 21 points. For the 14 games played during that span, the average margin of victory was 11.4 points. In the 22 games since, the margin has increased to 18.1 points.

If NFL history has proven anything, it’s that you don’t have to be imaginative to make it to the Super Bowl, but that you do if you intend to win it. Don Shula is the winningest coach in the NFL history – in terms of consistency he was a model of what a great coach should be – until he got to the championship games.

Counting the 1964 NFL championship game between Shula’s Baltimore Colts and the Cleveland Browns, his record was just 2-5 in 7 championship games, More to the point, his teams were outscored in the second half of those games by a combined score of 91-14. Here’s a real eye-opener: In six Su per Bowls, Shula’s teams scored just two touchdowns in the second half. Their opponents scored 64 points in the second half of those games.

In Bubba Smith’s 1983 autobiography, “Kill, Bubba, Kill”, the former Colts defensive lineman offered an insight into the chink in Shula’s mental armor. During the halftime of the famous 1969 upset loss to the Jets, Smith begged Shula to “let me line up over the center so I change their blocking scheme.” Shula’s reply was, “Just play your position.”

When it comes to improvisation, Dan Reeves makes Don Shula seem like Robin Williams. In the 1980s, Reeves took the Denver Broncos to three Super Bowls and got swamped in all of them. Was it a case of John Elway wilting under pressure? Elway didn’t think so and constantly sparred with his coach in public over Denver’s rigid game plans. The evidence supports Elway in a big way. In 1999, playing for Walsh disciple Mike Shanahan, he riddled Reeves’s Atlanta defense for 336 yards in a 34-19 win.

Most football coaches are a combination of tyrant and ideologue, but the handful that have been successful in Super Bowls have also been pragmatists, men capable of suppressing their own egos long enough to admit their judgment has been flawed. That’s part of it; the other part is being able to think fast enough to do something about it.

What does all this suggest for Sunday? Just that the game doesn’t always go to the best puncher, but sometimes to the best counterpuncher.