Scorecasting (4 page)

A shrewd batter armed with this information could—and should—use it to his advantage. Facing an 0–2 count and knowing that the chances of a pitch being called a strike are much lower, he would be smart to be conservative in his decision to swing. Conversely, on a 3–0 count, the umpire is much more likely to call a strike, so the batter may be better off swinging more freely.

From Little League all the way up to the Major Leagues, managers, coaches, and hitting experts all encourage players to “take the pitch” on 3–0. The thinking, presumably, is that the batter is so close to a walk, why blow it? But considering the home plate umpire’s omission bias, statistics suggest that batters might be better off swinging, because they’re probably conceding a strike otherwise. And typically, a pitcher facing a 3–0 count conservatively throws a fastball down the middle of the plate to avoid a walk. (Of course, if the pitcher also knows these numbers, he might throw a more aggressive pitch instead.)

There are other indications that umpires don’t want to insert themselves into the game. For as long as sports have existed, fans have accused officials of favoring star players, giving them the benefit of the calls they make. As it turns out, there is validity to the charges of a star system. Star players
are
treated differently by the officials, but not necessarily because officials want to coddle and protect the best (and most marketable) athletes. It happens because the officials don’t want to influence the game.

If
Albert Pujols, the St. Louis Cardinals’ slugger—for our money, the best hitter in baseball today—is up to bat, an umpire calling him out on a third strike is likely to get an earful from the crowd. Fans want to see stars in action; they certainly don’t want the officials to determine a star’s influence on the game. Almost by definition, stars have an outsized impact on the game, so umpires are more reluctant to make decisions against them than, say, against unknown rookies. Sure enough, we find that on two-strike counts, star hitters—identified by their all-star status, career hitting statistics, awards, and career and current salaries—are much less likely to get a called third strike than are nonstar hitters for the exact same pitch location. This is consistent with omission bias and also with simple star favoritism.

But here’s where our findings get really interesting. On three-ball counts, star hitters are
less
likely to get a called ball, controlling again for pitch location. In other words, umpires—already reluctant to walk players—are even more reluctant to walk star hitters. This is the opposite of what you would expect if umps were simply
favoring star athletes, but it is consistent with trying
not
to influence the game. The result of both effects is that umpires prolong the at-bats of star hitters—they are more reluctant to call a third strike but also more reluctant to call the fourth ball. In effect, the strike zone for star hitters shrinks when they have two strikes on them but expands when they have three balls in the count. Umpires want star hitters in particular to determine their own fate and as a result give them more chances to swing at the ball.

As fans, we want that, too. Even if you root for the St. Louis Cardinals, you’d probably rather see Pujols hit the ball than walk. As an opposing fan, you’d like him to strike out, but isn’t it sweeter when he swings and misses than when he takes a called third strike that might be ambiguous? We essentially want the umpire taken out of the play. Fans convey a clear message
—Let Pujols and the other team’s ace duel it out
—and umpires appear to be obliging.

The umpire’s omission bias affects star pitchers in a similar way. Aces are given slightly bigger strike zones, particularly on three-ball counts, consistent with a reluctance to influence the game by prolonging an outing. The more walks a pitcher throws, the more likely he is to be replaced, and that obviously has a sizable impact on the game and the fans.

In the NBA, home to many referee conspiracy theories, skeptical fans (and Dallas Mavericks owner Mark Cuban) have long asserted the existence of a “star system.” The contention is that there is one set of rules for LeBron James, Kobe Bryant, and their ilk and a separate set for players on the order of Chris Duhon, Martell Webster, and Malik Allen. But confirming that star players receive deferential treatment from the refs is difficult, at least empirically. Stars have the ball more often, especially in a tight game as time winds down, and so looking at the number of fouls or turnovers on star versus nonstar athletes isn’t a fair comparison. Unlike in baseball, where we have the Pitch f/x data, we can’t
actually tell whether a foul or violation
should
have been called. Did
Michael Jordan push off against
Bryon Russell before hitting the game-winning shot in the 1998 NBA finals? That’s a judgment call, not a call that current technology can answer precisely and decisively.

The closest thing to a fair comparison between stars and nonstars we’ve found is what happens when two players go after a loose ball. A loose ball is a ball that is in play but is not in the possession of either team (think of a ball rolling along the floor or one high in the air). Typically, there is a mad scramble between two (or more) opposing players that often results in the referee calling a foul. We examined all
loose ball situations involving a star and a nonstar player and analyzed how likely it is that a foul will be called on either one.
^*
A nonstar player will be assessed a loose ball foul about 57.4 percent of the time, a star player only 42.6 percent of the time. If the star player is in foul trouble—three or more fouls in the first half, four or more fouls in the second half—the likelihood that he will be assessed a loose ball foul drops further, to 26.9 percent versus 73.1 percent for the nonstar. But what if the nonstar player is in foul trouble but the star isn’t? It evens out, tilting slightly against the star player, who receives a foul 50.5 percent of the time, whereas his foul-ridden counterpart receives a foul 49.5 percent of the time. These results are consistent with the omission bias and the officials’ reluctance to affect the outcome. Fouling out a player has a big impact on the game, and fouling out a star has an even bigger impact. Much like the called balls and strikes in MLB for star players, it is omission bias, not star favoritism, that drives this trend. Star players aren’t necessarily being given better calls, just calls that keep them in the game longer.

Another long-standing fan accusation against referees is the use of the make-up call. When an obviously bad call is made, the thinking goes, the officials soon compensate by making an equally bad call that favors the other team. Or, in the next ambiguous situation, the refs will side with the team that was wronged previously. A few years ago there was a commercial for Subway that featured a football ref standing at midfield and saying: “I totally blew that call. In fact, it wasn’t even close. But don’t worry. I’ll penalize the other team—for no good reason—in the second half. To even things up.”

The stats do seem to confirm the reality of make-up calls, but again, this stems from officials not wanting to inject themselves into the game. If you know you’ve made a bad call that influenced the game, you may be inclined to make a bad call in the other direction to balance it out. The hope is that “two wrongs make it right,” but of course this means referees are consciously not always calling things by the rule book.

In baseball, we can look at make-up calls by the home plate umpire. If the umpire misses a strike call, how likely is it that the next pitch will be called a strike? It turns out that if the previous pitch was a strike but the umpire missed it and erroneously called a ball, the next pitch is much more likely to be called a strike even if it is out of the strike zone. If the previous pitch should have been called a ball but was mistakenly called a strike, the umpire is much more likely to call a ball on the next pitch even if the ball is in the strike zone. When umpires miss a called strike, they tend to expand their strike zone on the next pitch, and when they miss a called ball, they tend to shrink the strike zone on the next pitch.

The following graph shows the difference between the strike zones for pitches
immediately following
errant strike calls and errant ball calls. After an errant ball call, the strike zone magically
grows by 70 square inches. This pattern holds even for the first two pitches of the at-bat.

Also, the more obvious the mistake, the more umpires try to make up for it on the next pitch. If the pitch was dead center down the plate and the ump failed to call a strike, he or she
really
expands the strike zone on the next throw. If the ball is way outside and the ump doesn’t call a ball, he or she
really
tightens the strike zone the next time. Again, this is consistent with trying not to affect the game. Umpires are trying to balance out any mistakes
they make, and the more obvious those mistakes are, the more they try to balance things out.

It’s not just in MLB and the NBA that officials try to avoid determining the outcome. It also occurs in the NFL, the NHL, and soccer. The omission bias suggests that the rate of officials’ calls will decrease as the game nears its conclusion and the score gets closer.

In the NBA there is some evidence that fouls are called less frequently near the end of tight games, especially in overtime. (That includes the intentional foul fest that usually attends close games.) However, by looking deeper into the
types
of fouls called, or not called, late in the game, we get a more striking picture. Fouls more at the discretion of the referee—such as offensive fouls, which any NBA ref will tell you are the hardest to call—are the least likely to be called when the game is on the line. For some perspective, on a per-minute basis, an offensive foul is 40 percent less likely to be called in overtime than during any other part of the game. Certain “judgment call” turnovers, too, disappear when the game is tight. Double dribbling, palming, and every NBA fan’s favorite gripe, traveling, are all called half as often near the end of tight games and overtime as they are in earlier parts of the game. Remember the credo: When the game steps up, the refs step down.