Swallowing the Whistle: A Guest Post by Tobias Moskowitz

The following is a guest post by Tobias J. Moskowitz, co-author of Scorecasting: The Hidden Influences Behind How Sports Are Played and Games Are Won, now out in paperback. Moskowitz is a University of Chicago financial economist — you may remember Steve Levitt mentioning the book or the Q&A with the authors on the blog. They are also regular contributors to the “Football Freakonomics” project


Swallowing the Whistle
By Tobias J. Moskowitz

With the upcoming Super Bowl this Sunday pitting the Giants against the Patriots again (they last faced off in 2008), who could forget the most infamous play in Super Bowl history?  And in case you did forget, the image of David Tyree reaching back until he was nearly parallel to the field and snatching the ball with one hand and pinning it to his helmet has been either shown or referred to at least 150 times on ESPN and the NFL Network in the last week — and we’re still a week away from the game!

The play was extraordinary, no doubt about it, but perhaps the most interesting aspect of that play was something rather ordinary that happened well before Tyree made his remarkable grab (it was the last catch of his career by the way—one hell of curtain call!), something that is much more likely to be a factor in the upcoming game. 

If you rewind the play to a few frames earlier, you will find Eli Manning all but bear-hugged by a cluster of Patriots defenders — Richard Seymour and Adalius Thomas, in particular — who had grasped a fistful of the right side of his number 10 jersey.  Manning’s progress appeared to be stopped. Quarterbacks in far less peril have been determined to be “in the grasp,” a call made to protect quarterbacks that awards the defense with a sack.  However, next time you view the footage, pay close attention to what happened in the backfield before Manning made his great escape.  Mike Carey, the head official, started on the left side of the field but then backpedaled and found an unobstructed view behind Manning. A few feet away from the play, alert and well-positioned as usual, eyes lasering on the players, Carey appeared poised to declare Manning sacked.  And then . . . nothing. It was a judgment call, and Carey’s judgment was not to judge.

“Half a second longer and I would’ve had to [call him in the grasp],” says Carey. “If I stayed in my original position, I would have whistled it. Fortunately, I was mobile enough to see that he wasn’t completely in the grasp. Yeah, I had a sense of ‘Oh boy, I hope I made the right call.’ And I think I did. . . . I’m glad I didn’t blow it dead. I’d make the same call again, whether it was the last [drive] of the Super Bowl or the first play of the preseason.”

Others aren’t so sure. Reconsidering the play several years later, Tony Dungy, the former Indianapolis Colts coach and now an NBC commentator, remarked: “It should’ve been a sack. And, I’d never noticed this before, but if you watch Mike Carey, he almost blows the whistle. . . . With the game on the line, Mike gives the QB a chance to make a play in a Super Bowl. . . . I think in a regular season game he probably makes the call.”  In other words, at least according to Dungy, the most famous play in Super Bowl history might never have happened if the official had followed the rule book to the letter and made the call he would have made during the regular season.

It might have been a correct call. It might have been an incorrect call. But was it the wrong call? It sure didn’t come off that way. Carey was not chided for “situational ethics” or “selective officiating.” Quite the contrary. He was widely hailed for his restraint, so much so that he was given a grade of A+ by his superiors. In the aftermath of the game, he appeared on talk shows and was even permitted by the NFL to grant interviews — including one to us, as well as one to Playboy — about the play, a rarity for officials in most major sports leagues. It’s hard to recall the NFL reacting more favorably to a single piece of officiating.

(Photo: Peter McCarthy)

And why not?  The men in the striped uniforms and white caps did what they usually do at a crucial juncture: They declined to make what, to some, seemed like an obvious call.

If this is surprising, it shouldn’t be. It conforms to a sort of default mode of human behavior. People view acts of omission — the absence of an act — as far less intrusive or harmful than acts of commission — the committing of an act — even if the outcomes are the same or worse. Psychologists call this omission bias, and it expresses itself in a broad range of contexts.

Most of us probably would contend that telling a direct lie is worse than withholding the truth.  You might feel a small stab of regret over not raising your hand in class to give the correct answer; but raise your hand and provide the wrong answer, and you feel much worse. The first principle imparted to all medical students is “Do no harm.” It’s not, pointedly, “Do some good.” Our legal system draws a similar distinction, seldom assigning an affirmative duty to rescue. Submerge someone in water, and you’re in trouble. Stand idly by while someone flails in the pool before drowning, and— unless you’re the lifeguard — you won’t be charged with failing to rescue.  In most large companies, managers are obsessed with avoiding failure rather than missing opportunities. Errors of commission are often assigned responsibility, but people are rarely held accountable for failing to act, even though those errors can be just as, if not more, costly.

This same thinking extends to sports officials. In our book, Scorecasting: The Hidden Influences Behind How Sports Are Played and Games Are Won, we examine how omission bias affects refereeing.  We find evidence in every sport — baseball, basketball, hockey, soccer, and football — that officials are systematically more worried about making wrong calls than wrong non-calls.  As a consequence, they “swallow the whistle” when the game is on the line.  Especially during crucial intervals, officials often take pains not to insinuate themselves into the game. In the NBA, there’s even an unwritten directive: “When the game steps up, you step down.”

It’s a noble objective, but it expresses an unmistakable bias, and one could argue that it is worse tha
n the normal, random mistakes officials make during a game. Random referee errors, though annoying, can’t be predicted or gamed and tend to balance out over time, not favoring one team over the other.  A systematic bias is different, conferring a clear advantage (or disadvantage) to one type of player or team over another and enabling us — to say nothing of savvy teams, players, coaches, executives, and, yes, gamblers — to predict who will benefit from the officiating in which circumstances. 

One clear example we show in baseball is that the umpire’s strike zone shrinks considerably when there are two strikes on the batter and widens considerably when there are three balls in the count. Many pitches that are technically within the strike zone are not called strikes when that would result in a called third strike. Conversely, the umpire’s strike zone expands significantly when there are three balls on the batter, going so far as to include pitches that are more than several inches outside the strike zone. To give an extreme example, the strike zone on 3–0 counts is 188 square inches larger than it is on 0–2 counts in MLB. That’s an astonishing difference, and it can’t be a random error.

Plot of the empirical strike zone (defined as any pitch called a strike at least 50% of the time by MLB umpires) on 3-0 vs. 0-2 counts. The black box represents the actual rules-mandated strike zone.

In basketball, hockey, soccer, and the NFL, we also find that the rate of officials’ calls, especially judgment calls, goes way down near the end of tight games and the bigger the stage.  How do we know this is referee bias and not just the flow of the game?  Because non-subjective calls that require little judgment — such as delay of game and shot clock violations, where everyone in the stadium can see a giant clock indicate that time has expired or, in the case of the NBA, the entire goal lighting up red — do not decline as the game gets tight or nears its end.  Officials seem to systematically swallow the whistle on judgment calls and let the players determine the outcome.  As fans, we may want that, too, but keep in mind it means the rules of the game aren’t being applied uniformly.

Since the Super Bowl is arguably the biggest stage and since pundits are predicting a tight game, don’t expect too many calls near its conclusion, and who knows, we might end up with another infamous Super Bowl play that perhaps should not have happened.


Do you think holding should have been called on that play?


I've noticed Romo, Rothelisberger, and Brees using every possible second before hitting the ground to unload the ball and avoid a sack. Does the perceived ability of the quarterback to make a play factor in to the non-call?

Big Joe B.

I'm interested in the strike zone graphic. From where does that data come? It sure doesn't look like the strike zone I see: the one that favors low pitches and for the most part, slightly outside pitches. Umpires tend to set up over the catcher's shoulder and their view of the plate, an angle on one side or another, influences how they call 'em. I don't doubt the contention that the strike zone is enlarged on a 3-0 count, however.


Assuming this takes into account lefties and righties over the course of the entire season I bet the guys over at Baseball Prospectus could confirm this information for you. I wonder if there is similar variance in close games vs. blowouts, "underdog" vs. "juggernaut", allstar vs. bench, or even allstar pitcher vs. long reliever.

Caleb b

Read the book, loved it. Especially the part about loss aversion in PGA putting (the concept being that PGA players put better when putting to save par than when going for birdie bc missing the par put is "losing a stroke")

Mike B

In all sports leagues you are forgetting about Rule 0, always do what is best for the game. In the first example given we have a rule meant to protect QB's from getting injured, however in this case QB injury is not a factor because if they were to be sacked the season is immediately over and any injury would have no consequences. On the other hand if you were to enforce the rule aggressively the Referee would effectively be deciding the game instead of the players. Having major games decided by the officials will damage the integrity of the sport as well as make the sport's customers very upset. In any business you should never piss off your customers. The thing is that you seem to forget that if the Pats wanted to sack Manning they have the opportunity to simply bring him to the ground in a completely non-controversial way. A big part of Rule 0 is that teams should win through their own deeds, not due to the rulings of the officials. If you can't slam a QB into the ground they have nobody to blame but themselves.

In the baseball example we have the case of Rule 0 being that players come to the plate to hit the ball, not walk to first base. If they weren't there to hit then why bother to bring a bat? Fans want to see players hit, not walk, especially not walk after 4 pitches so expanding the strike zone makes the game more exciting and a better product for the customers.



I would argue that rule 0 is best followed by ruthless consistency. When an umpire's strike zone changes then pitchers don't know where to place the ball and batters don't know whether to swing. There's nothing unsporting about walking in 4 pitches, or striking out someone in 3.

There are 2 things it would be interesting to know:
1) What is the normally called strike zone?
2) How often did the altered strike zone change the outcome of the plate appearance?

Not calling the pitch is injecting the umpire into the game just as much as any other bad call, we just don't notice it as much (the point of the article).

t pappas

In the Pats-Ravens AFC championship, a TD was not called on the pass from Flacco to Evans at the end. Had a TD been called on the field it would have been reviewed in the booth and possibly/likely overruled. As it was ruled incomplete and occurring in the last two minutes, there was no ability for anyone on the field to request the replay review. The officials in the booth had an option to review and swallowed their whistles. A swallowed whistle or two may have changed the teams in this Super Bowl.


Quarterbacks know that in-the-grasp is NEVER called. Combine that with the hesitation that comes on defenders parts to not rough the passer (it's a quarterback league, don't ya know) because you couldn't fit a piece of paper in between the time a qb gets rid of the ball to the time a defender's hit is deemed "late," or the defender's take-down is flagged as "too rough." The league is getting sissified. I'm sure it's a mandate from the league, and Carey is just following orders, but the quarterback protection racket has watered-down the league considerably. There was no chance he calls Manning as being in-the-grasp.

Last year a hit on Michael Vick, in bounds, was flagged for 15 yards because the umpire felt that Vick was on his way out of bounds. So, now they're officiating intent. I was wondering what happens next time a qb acts like he's going out of bounds, freezes the defender, then cuts back and runs for more yards. I have a sneaky feeling the ball wouldn't be marked back to the point where he originally would have went out of bounds.



On the strikezone data, I would argue that an ump's expectations of the what both actors will do on the next pitch impacts the call more so then him not wanting to issue a walk or strikeout. With 3-0 count, people assume that if the pitcher doesn't want to walk the batter, he will throw a pitch in the center of the strikezone while the batter should take any close pitches. These expectations lead the ump to give the benefit of the doubt to the pitcher and call a strike. On a 0-2 count, batter should be in protect mode and pitcher more willing to throw ball or nibble on the corners trying to get the batter to swing at a "pitchers" pitch. Thus if pitch is taken the ump would give the benefit of the doubt to the batter.


As a rugby referee the law of advantage overrules all others, it gives the non offending side the opportunity to try something (a free play in effect) knowning that the can come back for the original offence. This of course is made all the more daring with good communication from the match official. Non offending team Players also know if the offence is obvious that they will not hear the whistle unless they are disadvantaged again. This makes for a more fluid and dynamic game. At the elite end of the game advantages can take up a considerable amount of the playing time.