Mike Tomlin iced the kicker twice last night, but with an unfortunate twist: First, he called a timeout on his own kicker, Shaun Suisham, who proceeded to miss a 54-yarder. Then he stopped the Titans' Rob Bironas just as he was about to attempt the game-winning 40-yarder. Bironas waited two minutes and nailed the kick, handing Tomlin and the Steelers a 26-23 loss.
Does interrupting the kicker and making him wait actually do anything? Icing is part theater, burning an otherwise useless timeout so the coach can do some coaching, screaming to the world "I AM A COACH! I AM A PSYCHOLOGICAL MASTERMIND!" It can also draw attention away from coaching mistakes that led to the field goal opportunity in the first place.
The question is whether this bit of drama has any influence over the result of the kick. Academic and amateur researchers have both weighed in, using varying methodologies and arriving at different conclusions. A 2010 University of San Diego study found that icing was effective, while a study in Jon Wertheim and Tobias Moskowitz's 2011 book Scorecasting found no effect. ESPN's Stats and Info blog found that iced kicks actually went in at a significantly greater rate than non-iced ones, though they didn't control for anything. It's an issue that has vaulted statistical research into mainstream coverage, which is always a good thing.
However, none of these studies are of much use for decision-makers: they're all based on results from a world in which the threat of icing is present on all kicks, whether they're iced or not. Even if a study authoritatively concluded that kicks went in at the same rate with or without being interrupted by a timeout, it could still be the right decision to call one.
Imagine two coaches, one who always calls timeouts before end-of-game field goals, and another who never does. Psychologically, a kicker should be comfortable with either one, because he knows exactly what will happen when he lines up. But what about a coach who flips a coin to decide whether or not to stop the play? Now the kicker has no way of knowing whether he'll be called back or not.
Then suppose the coach who always ices the kicker sees the coin-flipping coach and has an inspiration. Once per season, with the game on the line, he breaks his habit and lets the kicker—who's counting on an icing attempt—kick away. If the kicker misses, in surprise, the icing has paid off in its absence.
This is why it's hard to come up with blanket prescriptions about strategy and tactics. They belong to the realm of game theory.
Game theory is an area of formal study with applications across many different areas of science. What happens in dynamic systems where future reaction is informed by knowledge of past interactions? It gets complicated very quickly. A single instance of icing not only affects the kicker's routine—reinforcing or disturbing it—but the coach's reputation for being willing to ice, which in turn affects the expectations and decisions made by all the other coaches and kickers.
All aspects of football are informed by game theory, whether the decision-makers have heard of it or not. If an offensive line is stronger on the right side, most rushes will go towards the right. But with the defense geared up to expect a strong-side play, a weak-side run could be just as effective. As the offense and defense change tactics to take advantage of and prevent inefficiencies, they eventually settle into an equilibrium: There's a certain chance that a running back will go right, and a lesser chance that he will go left, depending on how great the imbalance is. Game theorists study how these equlibria establish themselves and shift over time.
My favorite thing about game theory is that it provides some scientific validation for occasionally being crazy as fuck.
Risky blitzes and trick plays might have a lower average output than their more boring counterparts, but that's partly because the lurking threat of the fake allows the ordinary plays to be more effective. It's part of what makes coaches like Les Miles so tough: he makes his opponents so paranoid about what he might do 1 percent of the time that they start to question their plans for the other 99 percent. It's all about reputation.
That's what icing the kicker is really about: coaches aren't just trying to get in the heads of the opposing kickers. They're also trying to establish their reputations as tacticians who have enough tricks up their sleeves to give opponents headaches when they try to prepare. But last night, Tomlin looked more like a second grader trying to anticipate paper over rock than like a brilliant field general. If he didn't have a problem icing his own kicker (something Suishham probably wasn't expecting), then why did he ice the other team's (something Bironas probably was)? When it comes to this particular tactic, the equilibrium has shifted: a well-placed non-timeout before a kick is undervalued. Sometimes the boldest, most unpredictable choice is to do nothing at all.