[Hint: It’s because the ones that last are no longer seen as "gimmicks."]
We’ve all heard the skeptics. "That read-option stuff is just the latest flavor of the month." "With a whole off-season to plan, defenses will figure it out." "That sort of thing may work in college, but NFL players are too strong and too fast for it to work." "It’s like the wildcat formation ... and gimmicks like that never last."
In this post, I examine that last-quoted claim. Is it true that "gimmicks" never last? Or, to state the question more precisely, do gimmick plays tend (on average) to have a shorter life in the NFL than non-gimmick plays do?
At first, this sounds like an empirical question, which can only be answered by empirical data. If we could list all the gimmicky innovations of the last thirty years or so, we could (if our data were good enough) find out how long the average gimmick lasted. If we then followed the same steps for all the non-gimmicky strategies introduced in the last 30 years, we might then be able to tell whether gimmicks did or did not last as long as non-gimmick innovations did.
As most of you have already realized, though, to perform such a test we would need to be able to define a "gimmick", in order to separate the gimmicky innovations from the non-gimmicky ones. And while it might seem as though gimmicks should be easy to define ("I know a gimmick when I see one!"), in fact the problem is more subtle.
To put this issue in perspective, consider the following objections that might have been (and often were) raised against various strategies of the past. Then decide how many of these strategies are properly described as gimmicks:
- "The wildcat formation is just a gimmick. Mark my words: in a few more years, no one will even remember it."
- "The naked bootleg is just a gimmick. Once defenses expect it, they’ll find a way to shut it down. Besides, it exposes the quarterback to a heightened risk of injury."
- "Play-action passes are just a gimmick. Once a defense knows that a team uses them, their linebackers and safeties will be less agressive and won’t be sucked in by the appearence of a run."
- "The counter play is just a gimmick. Once defenses know to expect it, they won’t be fooled by the misdirection."
- "The shotgun formation is just a gimmick. Once defenses have some time to figure out how to stop it, it’ll end up in the scrap heap of all the other gimmicks that failed."
Of course, I could have listed lots of other examples (The T-formation is a gimmick! The forward pass is a gimmick!), but by now you get the idea. My guess is that many of you were perfectly willing to accept the characterization of the wildcat formation as a gimmick. At the same time, few if any fans today think of the shotgun formation as a gimmick, or the naked bootleg, or any of the other plays and formations listed above. "That’s not a gimmick," we say; "that’s just smart football."
What is it, then, that makes the wildcat different from (or more "gimmicky" than) the others? Well, for one thing, the the wildcat formation fell out of use (in the NFL) after just a few seasons, while the shotgun has been used for decades, and its use has yet to even begin to decline. To many people, the decades that it has remained in use make the shotgun seem clearly less of a gimmick than the short-lived wildcat formation.
In addition, the wildcat formation did not even begin to be used in the NFL until just a few seasons ago, so most fans today can easily remember a time when the wildcat had no place in professional football. By contrast, the shotgun was introduced into the NFL several decades ago, which is longer than many football fans have been alive. As a result, many current fans have never known a world in which the shotgun wasn’t used at least occasionally – and this, too, makes the shotgun seem less of a gimmick.
Generalizing from these examples, I think most of us have an implicit idea of a "gimmick" that runs along the following lines. Gimmicks are plays or strategies that were introduced relatively recently (preferably within the lifetimes of most living fans), and which fell out of use after a relatively short time. Non-gimmicks, by contrast, are those that were introduced farther in the past (preferably before most living fans began following the game), and which remained widely used for a relatively large number of years.
Now -- finally! -- we can see the fundamental problem with the claim that "gimmicks never last." If gimmicks are defined as those plays or formations that (among other things) are used for only a short while; while non-gimmicks are defined as plays or formations that are used for a longer period, then of course gimmicks will disappear more rapidly than non-gimmicks, because that’s how they’ve been defined. Under these definitions, the claim that "gimmicks never last" will always be true -- but it is an empty, formal truth, which tells us nothing except how certain terms have been defined.
To put it another way, therer is a fundamental difference between empirical claims and analytic claims. Empirical claims, like "unmarried men tend to have long hair," might conceivably be false, and these claims can be tested only by gathering data about the hair length of lots of unmarried men. Analytical claims, though, are claims that are true by definition, like the claim that "unmarried men tend to be bachelors." There is no point in trying to gather empirical data about an analytic claim, because no matter how many unmarried men we survey, we know in advance that every single one of them will turn out to be a bachelor.
Unfortunately, though, analytic claims are sometimes the most difficult ones to persuade people to abandon. Indeed, the empirical evidence will often seem to support such claims. To someone who is skeptical about football gimmicks, every gimmick that he can think of will indeed have been used for only a short while. (Otherwise, if it had lasted longer, it wouldn't be classified as "a gimmick"). In a similar way, someone who believes that all unmarried males are bachelors might think that his belief, too, has strong empirical support, because every unmarried male he can think of turns out to be, yes, a bachelor. (Otherwise, if someone is not a bachelor, he wouldn't have been classified as an unmarried male.) Obviously, though, this kind of empirical "support" tells us nothing we didn't already know: nothing that wasn't already implied by the key definitions.
Oh well. The preceding paragraph may sound pessimistic, for it suggests that changing peoples' minds will be a long and slow process. In other words, commentators who last year knew that gimmicks always fail, will probably continue to spout that same belief this year. The more things change, the more they remain the same.
Ultimately, the only way to really change the minds of traditional commentators and skeptics is probably through the results on the field, if read-option teams continue to use it with success. I'm a fan of the Seattle Seahawks, so it's hard for me to even think of myself as a Redskins fan, or a Panthers fan, or (worst of all!) a 49ers fan. But that's where I increasingly find myself: hoping that all four of those teams do well.