December 30, 2015

Best of BP 2015

Stealing Signs

by R.J. Anderson

With the year winding to a close, Baseball Prospectus is revisiting some of our favorite articles of the year. This was originally published on May 28, 2015.

Stealing signs is ubiquitous, according to those inside baseball—you just have to know what to look for, and when and where to look to see it.

But many observers don't know the whats, whens, and wheres of tactical looting, meaning a player in plain sight can tip his teammates on the upcoming pitch without anyone located outside of a dugout noticing. Fortunately, this season recently offered a lesson on sign-stealing, as conducted by FSN Arizona's Bob Brenly during the ninth inning of a May 9th game between the Padres and Diamondbacks.

After a leadoff double by Padres outfielder Wil Myers, Brenly noticed Myers palming his helmet before a pitch was thrown, signifying, in baseball circles, that he knew the opposition's signs. Myers wasn't playing poker to unnerve pitcher Addison Reed and/or catcher Jordan Pacheco, either. Instead Myers correctly called out each subsequent pitch using a simple system: He placed his hands on his legs when a fastball was coming and kept his hands off his legs when it would be an off-speed pitch. Predominantly a fastball-curveball pitcher, Reed served as the perfect mark for the binary system.

That Myers broke the battery's code in such a short period raises many questions. Were Reed and Pacheco's signs that obvious to everyone but the common viewer?

Let's play a game to find out. Below are Pacheco's signs on three pitches, as shown by FSN Arizona. Each resulted in a fastball or curve targeted to Reed's glove side. To "win" the game all you have to do is figure out which set of signs called for which pitch. Simple? Try it:

Ready for the answers? Here they are, as signaled by Myers. Remember: hands on the legs equals fastball, hands off means curveball:

Get it? No? You weren't the only one. Let's run through some guesses made by various BP staffers.

Theory no. 1: The last sign is live
Why it works: It correctly identifies each of the pitches.
Why it doesn't work: It doesn't correctly identify location (e.g., the second pitch is a fastball away, yet Pacheco signaled with his index finger, or for a fastball inside).

Theory no. 2: The chest is the indicator
Why it works: If the knees correspond to a fastball in that location, and the helmet means a breaking ball, then the finger signs are decoys and the body touches are the real signs.
Why it doesn't work: You have to go back, earlier in the inning, to disprove this theory: Pacheco touches his chest then his left knee—seemingly signaling for a fastball inside—before setting up to receive a breaking ball away. Perhaps the signs changed during the interim period?

Theory no. 3: The last sign is live and the location is the opposite side of whichever knee is touched last
Why it works: Pacheco touches his left knee last on each sequence, which could mean to throw the pitch to the opposite side of the plate. Add in how the last finger correctly identifies the pitch type, and this has a chance.
Why it doesn't work: This is the best of the guesses. There's just one problem: Myers seems to signal before seeing the finger signs.

There were other theories, but nobody could definitively crack the code. That's to be expected; figuring out a team's signs based on a select few pitches should be impossible, or else pitchers would be in trouble. Besides, there are plenty of reasons why you don't see many hobbyists sign-stealers.

For one, batteries are constantly making their messages tougher to decipher. That can mean adding signs, such as the catcher touching his body like Pacheco did; installing dynamic sign packages that change based on variables, like the number of outs or the ball-strike count; embedding an indicator among the signs; having the pitcher call the pitches through subtle body language; having switchers and flippers that change the pitch after the fact; and even flying blind, with the catcher resigned to grabbing whatever the pitcher throws his way.

For another, while the odds were very much against any random individual figuring out the Diamondbacks' signs, there's power in numbers. Most teams assign bench players to watch for these things throughout the game, sometimes to great success: You've probably heard or read Mickey Mantle crediting Bob Turley for more than half his big-league home runs. Heck, even in this instance, you can guarantee the signs were cracked on the bench and not by Myers. Know why? Because Myers touched his helmet before Reed even threw a pitch with him on base. How could Myers have figured out the signs (and all the evidence suggests he did have the signs) without any validation?

Plus, in addition to having more eyes with greater experience than the typical onlooker, the Padres had something else working in their favor: a constant visual of the pitcher and catcher. Modern broadcasts are too herky-jerky, shifting from the pitch to various close-ups and back before the next offering. San Diego's bench didn't have to deal with overlays and advertisements for promotional giveaways. Those watching along at home do. As such, the answer to how the Padres stole signs could be as simple as a between-pitch gesture that the cameras didn't pick up.

So while there's no reason to doubt that everyone in baseball steals signs, the unintended implicit meaning—nobody outside of baseball steals signs—is nearly as accurate. How couldn't it be?

Special thanks to Colin Young for professional insight, Nick Wheatley-Schaller for visual assistance, and Jesse Sanchez, who first noted Brenly's comments on Twitter.

R.J. Anderson is an author of Baseball Prospectus. Follow @r_j_anderson
Click here to see R.J.'s other articles. You can contact R.J. by clicking here