Points taken: Going for 2 isn’t so risky

Chris Petersen was too busy working on a Boise State game plan to watch the NFL on Sunday.

When Denver Broncos coach Mike Shanahan eschewed a final-minute tying extra point in favor of what turned out to be a winning 2-point conversion, word got to Petersen, perhaps all of football’s most aggressive coach, quickly.

The decision stunned the NFL, where it’s rare to dare. It left Petersen laughing at the reaction.

“I heard some of the announcers were saying, ‘He shouldn’t be going for it. What is he doing? They’re playing at home,’ ” Petersen said Tuesday. “I’m just kind of chuckling that people made such a big deal about it. He obviously knew exactly what he was doing.”

Well, you’d think. Shanahan is in his 16th season as an NFL head coach, has reached the playoffs seven times and won two Super Bowls.

When he decided to become just the seventh NFL coach to attempt to win a game by going for 2 though, you would have thought he was some bumbling rookie. It wasn’t just his coaching acumen that came into question, it was his general sanity.

He either had guts or was nuts.

That the play worked is all that spared him from a week of Ed Hochuli treatment. Which says more about everyone else than the decision itself.

The fact this is rarely attempted (its worked three of the seven times) is a testament to the NFL’s Stepford-like thinking, its constant overreaction to anything different and its forever rewarding of conservative over innovative.

It’s what puzzles Petersen so much.

What Shanahan did, he says, is exactly what a coach should do. He considered all the variables of the game and made a bold decision that he thought best positioned his team for victory.

In this case with a hot quarterback – Jay Cutler – he thought it was a high-percentage play. Not making the call would have been the wrong one.

“You kind of have a gut feeling,” Shanahan explained simply.

That’s nearly taboo in the NFL these days. Just about everything is planned, calculated, predicted, analyzed and reanalyzed. Anything that goes counter to conventional wisdom is ridiculed.

What if the wisdom isn’t wise?

Petersen, who boasts a 25-3 record, certainly doesn’t believe in it. He famously beat Oklahoma in the 2007 Fiesta Bowl on a final play 2-point conversation, a statue of liberty no less. Just last Saturday, in Boise’s victory over Bowling Green, he went for 2 after securing a 6-0 lead in the first quarter.

They used to call guys such as him “riverboat gamblers.” It’s an outdated term since, as Deadspin’s Clay Travis notes, just about every 80-year-old grandma in the South and Midwest has gambled in a riverboat casino these days.

It’s a fitting description for NFL coaches then. They are more about nickel slots than all in on three queens. These days an NFL coach is more likely to rely on an actuary than an ace up his sleeve.

Brian Burke is a former Navy pilot and current math whiz who runs the fascinating website AdvancedNFLstats.com. He says 99 percent of extra points are converted but just 44 percent of 2-point conversions.

Thus, had Shanahan kicked for the tie Denver would’ve had a 50-50 shot in overtime or a 49.5 percent win probability. That’s better than the 44 percent of going for two.

“On average, the 2-point decision was the wrong call in Denver’s situation,” Burke said. “In order for the 2-point attempt to be a rational decision, Shanahan must have thought the chances of success would be at least 50 percent.”

Burke is the first to admit on average doesn’t necessarily mean anything and statistically rational isn’t always the way to make game decisions.

Many commentators miss that. This isn’t even about the risk of losing the overtime coin flip – 17 percent of flip winners score immediately and 76 percent win eventually.

Most times the safe decision is made because risking it and failing means there is a 100 percent chance the coach will get all the blame. For reasons unknown, in the NFL a passive loss is better than an aggressive one.

Petersen can’t understand why. Shouldn’t the decision garner more respect? There is simply no right or wrong call there. Playing it safe can be just as foolish since all this talk about the odds of a 2-point conversion working are based on empty data.

There are so many variables in a football game – momentum, weather, matchups, fatigue, play calling, etc. – that each 2-point conversion should be treated as an isolated action and not part of some statistical trend compiled by other teams against other opponents in other situations often in other seasons.

The most obvious and important variable is talent. The success rate for the current Cowboys is likely to be far greater than the current Rams. So why would they both subscribe to a statistic based on data generated by neither of them?

And why would coaches have to live in fear of media and fans holding them to such a number?

This wasn’t about Shanahan bucking the odds because there were no calculable odds on whether Denver could convert. The odds of Cutler converting a 3-yard pass at that time might have been quite high.

Petersen’s famous 2-point conversion against Oklahoma came in part because he knew he had a high-probability play.

With the Sooners’ spinning from a series of previous trick calls, Petersen lined three receivers right to draw the safeties. Once that happened, the game was all but determined. The fake Statue of Liberty pass just gave the left side of the Boise line time to push forward for a run.

While the play looked daring, it was actually conservative. The ball was safely handed off to Ian Johnson for a simple sweep left. Three OU defenders were sealed off by three Boise blockers. Johnson jogged in without anyone within 5 yards.

So while Petersen was hailed for his courage in “ignoring the odds,” what were the real odds that this specific play was going to work? Maybe 90 to 95 percent? Only a failed block or a superhuman defensive effort could have changed things. You couldn’t run the same play again (OU wouldn’t be fooled) but this one time it was virtually unstoppable.

No wonder Petersen looked so calm on the sideline that night. And no wonder he laughs when people question decisions based on non-relevant statistics.

Denver’s play wasn’t as nearly foolproof. It was essentially the same one it ran to score the touchdown, a fairly high-percentage play. Shanahan’s growing confidence in Cutler was the variable that pushed the decision, he said.

“That’s why you go for it,” Shanahan told our Michael Silver.

With such longevity in Denver, Shanahan also was unlikely to be fired due to one decision, no matter how misunderstood it might have been. The fact his defense was struggling to stop San Diego (as Boise’s was with Oklahoma) might have played a part as well.

It’s why each decision should be treated on its own and not universally condemned. It’s why plays like these and the old riverboat gamblers who once called them shouldn’t be so rare. It’s why the coach who goes safe isn’t always going smart.

It’s why conventional wisdom sometimes isn’t so wise

Dan Wetzel is Yahoo! Sports' national columnist. He is the co-author of the book "Death to the BCS: The Definitive Case Against the Bowl Championship Series," which following five printings of the first edition was re-released in a second, updated edition in October. Follow him on Twitter. Send Dan a question or comment for potential use in a future column or webcast.
Updated Thursday, Sep 18, 2008