The Baseball Economist: The Real Game Exposed (4 page)

With play-by-play data we were able to look at eight seasons’ worth of data (two four-season samples—1969 and 1972–1974, and 1989–1992) to estimate the determinants of hit batters.
⁹

Here are some interesting things we found.

Most importantly, we found that after controlling for these many factors—especially the quality of the batter, which should capture the unequal factor that generates the change-in-demand theory—having a DH increased the likelihood of a batter being hit by 11–17 percent. Even after we accounted for the fact that pitchers were much less likely to be hit than other batters, the existence of the designated hitter is associated with a greater incidence of hit batsmen. Therefore, it is very likely that the law-of-demand hypothesis is a partial reason for the difference in hit batter rates across the leagues, explaining between 60 and 80 percent of the league differences over these two data samples.

Unfortunately, at the time of our initial study, play-by-play data for games after the 1992 season were not freely available to the public. This would be a good sample of data to examine because the nature of the hit batter difference changed quite dramatically in the 1990s. Luckily, game-by-game data does exist from 1973 to 2003, the entire DH era. Although we could not control for play-specific factors influencing hit batters in this sample, we could look at the game-specific factors, such as score and the average offensive and defensive abilities in scoring and preventing runs.

The beginning of interleague play in 1997 provided a unique opportunity to see how the use of the DH by National League teams, and lack of it by American League teams, affected the hit batter rates. Because of the DH rule difference, Major League Baseball adopted the policy of having a DH based on the league rule of the home team. This allowed us to determine whether or not the identified impact of the DH on hit batters in our play-by-play study was due to a price response or if it was just a difference in the American League style of baseball that we had yet to identify.

The results from the game data were consistent with the play-by-play data. Having a designated hitter was associated with about an 8 percent higher rate of hit batters than without. And interestingly, when we looked at interleague games we found that the DH was associated with an 11 percent increase in the incidence of hit batters. This is strong support for the law-of-demand hypothesis. In regards to hitting batters, National League teams acted more like American League teams with the DH and American League teams behaved more like National League teams when they lacked the DH. We also found that for each batter the batting team plunked when pitching, the incidence of hit batters by the pitching team increased by 10–15 percent. This provides further evidence that teams do retaliate for pitchers hitting batters.

Economist Steven Landsburg wrote, “Most of economics can be summarized in four words: ‘People respond to incentives.’ The rest is commentary.”
¹⁰
Indeed, this chapter provides a clear example of how even small changes in incentives in a trivial sporting contest can affect human behavior. Pitchers respond to the price of hit batters in a predictable manner. When the “price” of hitting batters changes, pitchers change their actions accordingly.

Pitchers respond to other incentives, as well. If the batter isn’t much good, why risk hitting him with a harder-to-control pitch? Pitchers are more likely to hit good batters, who are more likely to punish pitchers for leaving balls out over the plate. Additionally, they are less likely to hit pitchers on the opposing team, who represent an extremely poor class of hitter. In these ways we begin to build the best intuition about how a pitcher or any player in any position will behave.

The law of demand, which is deemed a “law” because its powers of foresight are so great, is something we should never ignore when analyzing human actions in baseball or anywhere else.

2

The Legendary Power of the On-Deck Hitter

—STEVE HENSON
¹¹

AS MATT MORRIS toed the rubber during the first inning of the 2004 World Series for the St. Louis Cardinals, his focus should have been on the batter, Manny Ramirez. But how could it be? He’d just given up a double to Johnny Damon and hit Orlando Cabrera with a pitch—two runners on base. There were no outs and he was now facing a man who’d swatted forty-three home runs that season, more than any other player in the American League. As if the pressure weren’t enough, he couldn’t ignore the sight of David “Big Papi” Ortiz—who’d hit forty-one home runs—standing in the on-deck circle. Should he put some more gas on the fastball or go with something off-speed? He had to do something, because it wasn’t going to get any easier. While he certainly wished the situation were better, he must have been thankful that at least there wasn’t anybody on third. . . .

Baseball fans like statistics in part because the team game can be quantified into many mini-contests between individuals. Batting statistics especially seem to reflect individual achievements that don’t rely on teammates helping one another out, which is different from individual performance in other team sports. A basketball player might score a lot of points, but maybe it’s because other good players on his team freed up the defense. Or in football, a receiver might have a lot of receiving yards, but is it because of his speed and catching ability or the result of an excellent quarterback? In these sports it’s hard to evaluate players independently of other players.

In baseball, a hitter steps up to the plate to face a team of nine men all alone. The outcome of this event is surely the triumph or failure of the individual who stands against his rivals. And the statistics he puts up over a given time frame, whether it is a season or a career, indicate a certain level of individual skill demonstrated over that span. Therefore, baseball fans feel a degree of comfort in evaluating batters using well-worn statistics such as batting average or slugging percentage or on-base percentage. However, there does seem to be one exception to this view. In fact, it’s a widely held belief that a man who isn’t even participating in the game at a given moment can influence a batter’s success. Many people believe that the hitting prowess of the on-deck hitter—the batter who follows the current batter—influences the performance of the batter. How could that possibly be?

It’s the pitcher’s job, while facing a series of hitters, to minimize the runs produced by the opposing team. One way to do this, which we looked at in the last chapter, is to avoid giving good batters pitches right over the center of the plate, because they are easier to hit. On a team with very few good players, a pitcher may decide to “pitch around” a dangerous hitter—even putting him on base by walking him—in order to pitch to a less threatening on-deck hitter. The logic behind this move is that this limits the runs scored by the opponent by limiting the opportunities for good batters to produce runs through hitting the ball. Pitching around a good batter is something that gets more costly for a pitcher to do the better the on-deck batter is. Putting one runner on base so that he can throw to another good batter is something a pitcher tries to avoid. Therefore, it is often assumed that hitters can “protect” one another in the lineup. A hitter who is good enough to help the preceding batter get better pitches to hit is said to offer
protection
.

The concept of protection is so commonly thrown around between baseball commentators that it’s come to be accepted as fact. I often hear comments like “The Giants need to acquire a slugger to protect Bonds, or opposing teams are just going to pitch around him,” or “His numbers may look impressive, but that’s mostly a product of having a slugger hitting behind him in the batting order.” The impact of one person’s actions spilling over onto others is a phenomenon well known to economists. A discipline that assumes individuals always act according to their own self-interests ought to be concerned when the behavior of one individual has consequences for other individuals. Economists refer to any action that imposes costs or benefits on other parties as an
externality
.

The externality is simply the amount of the cost or benefit that spills over onto a third party. Externalities can be both positive or negative. Actions that have good consequences for third parties generate a positive externality, and those that have bad consequences produce negative externalities. It’s not a hard concept to understand, because we experience externalities in everyday life. When someone cuts you off in traffic, that’s a negative externality. Your next-door neighbor who keeps a well-manicured lawn generates a positive externality.

Whenever externalities are present, they generate problems for society. Negative externalities are overproduced, because individuals who engage in behavior that causes a negative spillover don’t always take into account the negative consequences of their action. For example, a woman who lights a cigarette in an elevator has the pleasure of the nicotine rush, but often doesn’t appreciate the irritation to the lungs of her fellow passengers. Positive externalities are underproduced. A man who paints his house will reap monetary benefits when he sells it, but he probably won’t factor in the benefit that his neighbors’ houses will sell for more because of the pleasant view he creates. Rather than paint it every year, he might do so every five years.

Sure, most individuals do care about other people, but it would be a stretch to say individuals care as much about others as they do about themselves. After all, the Golden Rule doesn’t say “love your neighbor as you love your neighbor.” And even if you did care about others as much as you did yourself, you would lack the information to know how your decisions impact others . . . that is, unless you can read minds. The problem with externalities is that people make decisions that they would otherwise not make if they reaped the rewards or bore the costs of those decisions in accordance with their impact on other people. For example, let’s say we have an elevator with eleven passengers, one of whom wants to smoke a cigarette. If ten passengers in the elevator value having their lungs clear of smoke for the ride at ten cents a piece, and the smoker would prefer to be a dollar richer than enjoy the cigarette on her ride, the world would be a better place if this transaction took place. The smoker gets a dollar and the passengers get clean air, and all parties are happier than they would have been had the trade not occurred. In a situation like this, a transaction doesn’t often take place, simply because the cost of organizing the transaction is too high. So the smoker will end up smoking and the passengers will bear the cost of her decision to do so. Therefore, we often adopt rules (like prohibiting smoking on elevators) to mirror the likely outcome if such a market did exist.

The house-painting positive externality is a little more interesting. Externalities in real estate transactions can have huge financial and quality-of-life rewards. Everyone likes a good neighbor, and we would all like to live in neighborhoods where people keep up the appearance of their property. Unfortunately, few people are willing to undertake projects because they benefit neighbors. A popular solution to this problem is the use of restrictive covenants in real estate developments, which require owners to adhere to certain appearance standards to guarantee positive spillovers to neighbors. While you might think that buying a piece of property with a catch is not as desirable as buying one without—most certainly you would prefer not to be burdened by rules—these covenants are popular because they ensure that all of the owners in the area do the things necessary to keep neighboring property values high.

But now back to baseball. The hitting externality in baseball doesn’t have the negative consequence of influencing market transactions. It’s just a spillover, and this spillover is interesting for a different reason. Quantifying the impact would be of great help to general managers, agents, and arbitrators in evaluating talent. Managers might learn how to better arrange their batting orders for the best run production. If player hitting abilities do influence one another, then we ought to learn how to evaluate the influence of talent properly.

In order to properly estimate any potential externality in hitting, we need to hypothesize about the possible external effects we are looking for. The traditional model of protection states that a good on-deck hitter prevents a pitcher from pitching around a good batter. This means that the batter is more likely to see good pitches to hit, which will result in more frequent and powerful hits. It should also lower the likelihood that he walks. For simplicity, we’ll assume that the improved hitting is more valuable to the hitting team that the forgone walks. Conversely, a poor on-deck hitter means a good hitter will see fewer pitches in the strike zone, which limits his ability to hit. He may walk more, but his hitting production will suffer. In economic terms, a good on-deck hitter generates a positive externality, while a poor on-deck hitter generates a negative externality.