The Baseball Economist: The Real Game Exposed (6 page)

I listen closely when I hear commentators discuss their explanations for the lack of left-handed catchers. Here are some that I find unconvincing.

The first two explanations involve skills that are too minor to explain the complete exclusion of lefty catchers. All baseball players can throw hard and straight without much fading. Framing pitches is more about acting than the positioning of the catcher and offers no obvious advantage to catchers of either handedness. The last explanation doesn’t fit with what we know about players learning new skills at the major-league level. Many players learn new positions and skills that they never played or used before. If a lefty could play catcher, the fielding instructors would be able to teach him how to do it.

The answer most commonly given, and the one I’ve found most convincing, is that left-handed catchers have a harder time throwing out runners at third base on steal attempts. The way the catcher positions himself to receive the pitch creates a difficult throwing angle for left-handers. A lefty must pivot and possibly throw behind a right-handed batter. This is similar to the difficulty that a left-handed shortstop faces when throwing to first. It’s possible that the difficulty in throwing to third base allows runners at second a greater opportunity to reach third on a steal. Problems solved, right? I’m not so sure.

Is this really that much of a problem? Right-handed catchers don’t seem to have too much of a problem firing pickoff throws to first base (the equivalent of left-handed catchers throwing to third base). Just ask Manny Ramirez, whom right-handed catcher Javy Lopez picked off first base at a crucial time during the 1995 World Series. To examine the potential advantage that right-handers have at preventing opposing runners from producing runs by stealing third, we need to know how valuable it is to keep runners on second base from getting to third base.

We can gauge the value of third base by quantifying the change in the runs a team is expected to score in a given base/out situation. The farther a runner is around the bases, the more likely it is that a runner will score. Additionally, each subsequent out in the inning lowers the number of runs we expect a team to score in an inning, because each out limits the opportunities for a team to knock in runners on base.

Thankfully, the tools we need to measure the trade-off are available and were first presented by George Lindsey in 1963. At the time, Lindsey was an Operations Research Scholar for the Canadian Department of Defence, who had an interest in baseball. With the help of his father, Lindsey tracked the individual plays of baseball games to gather the frequency of different events in baseball.
¹⁷
There are eight possible runner configurations (empty, first only, first and second, etc.), three out situations (zero, one, or two), for a total of twenty-four possible states in an inning. From his data, Lindsey calculated the frequency of and the expected runs to be scored from every base/out situation. Lindsey’s chart is a helpful tool, because baseball games produce a multitude of outcomes from the same initial starting state. The chart is simply a compilation of the average of all of the run outcomes at the end of the inning in which the state occurred. For example, with a runner on first with no outs, the expected number of runs to be scored from this situation
is .813; however, a runner on second with one out can be expected to produce .671 runs in that inning (that’s right,
on average
—not in all instances—sacrificing an out for a base in this situation
lowers
the expected number of runs scored in the inning).

We also need to know how frequently runners are in a position to steal third base. Even if keeping a runner off third base is valuable, opportunities to steal the base happen infrequently. A player can steal third only when there is a runner on second with third base open. By adding up all of the event frequencies for a runner on second and runners on first and second, italicized in Table 3, stealing third is an option only 13.7 percent of the time.

What we need to do with the information in these tables is to estimate the change in expected run-scoring from stealing third, which catchers try to prevent. A successful steal of third changes the game situation in three ways:

Since the change in the third state is rare—almost always the runner on first goes to second when the runner on second goes to third—let’s exclude it from the analysis. Table 4 shows the change in expected
runs per inning from changes in the base/out situation for all out configurations, weighted for the frequency with which they occur. We can use this table to compare the expected runs gained by stealing third to the expected runs from staying on second. The final columns of the table show the expected raw gain in runs from stealing third and the gain weighted for the frequency of times the initial state occurs in a game. The bottom right corner of the table sums the runs gained on a per inning and per game basis. The numbers tell us how much stealing third is worth if a runner on second successfully steals third every time the opportunity arises. The result: on average, stealing third nets the team about 0.4 runs per game, which conversely means keeping a runner from stealing third base lowers the opposing team’s expected runs by that same amount. The difference is not huge, but it’s large enough to be relevant—about sixty-four runs a year. If you are having trouble thinking of this number as small or large, you can view it as we view a pitcher’s ERA difference of 0.4 earned runs per game. As a manager, would you have preference between pitchers with 3.01 and 3.41 ERAs? I would.

However, this difference is actually very small, probably too small to be the only explanation for the bias against left-handed catchers. Four-tenths of a run is the amount of runs a team gains if it
successfully
steals third base
every time
a runner reaches second base with third base open. Teams don’t steal third base every time the opportunity exists. Failing to steal third is costly because the team gives up a runner on second and an out, which are both valuable inputs to run production. An extra out means fewer opportunities to push runners across the plate, and the forgone runner removes a player that was previously in “scoring position.” That is why steals of third base are rare.
¹⁸

So the gains from keeping runners on second from stealing third are small, and this is where the advantage of the right-handed catcher is supposed to be. Even if we made the extreme assumption that runners successfully stole third every time they got the opportunity against lefty catchers, and righties always put them out, it would net the opposing team only 0.4 runs per game. Since we expect teams to steal less than this, the expected gains are less.
¹⁹

However, why don’t teams put lefties behind the plate if the benefits from excluding them are so small? As George Stigler warned, simply claiming managers have just been making a mistake for so long isn’t the best path to understanding. So let’s dig a little deeper. While the benefits of using right-handed catchers are small, maybe the costs will yield some answers. What is the cost of excluding left-handers from catching?

A large benefit of playing a left-hander in the field is gaining his left-handed bat in the batting order—nearly all left-handed throwers possess the valuable trait of batting left-handed. Lefties can play four positions in the field: all of the outfield positions and first base. All of these positions, except for center field, are considered to be easier to play than the infield positions reserved for right-handers. This gives a team three positions to stash even a defensively poor left-hander in the NL, with the DH as a fourth option in the AL. Unless a manager has an overabundance of left-handers on his team, it’s very unlikely that he will need the catcher’s spot to get a left-handed bat in the lineup, especially considering the fact that many right-handed fielders possess the ability to bat left-handed. For all non-catching fielding positions played from 1998 to 2004, left-handed throwers have manned those seven positions 24 percent of the time. This means that it’s the average baseball manager’s job to find a place for about two players (24 percent times 7 equals approximately 1.7 positions) to play among the seven positions in the field. Even after allotting three of those positions to right-handers only, the manager still has four slots to play lefties in the field without having to squeeze one into the catcher position.

An outfielder has to be able to run and a catcher must have a good throwing arm, while first base requires the least amount of fielding ability on the diamond. The only real reason a manager would be tempted to put a lefty behind the plate is if he had a not-so-agile lefty with an excellent throwing arm and first base reserved for another player. This is a situation that very few teams face. There may be situations in the career of a left-handed player that would make his playing catcher a worthwhile move for the team to do at the moment. For example, a team has an up-and-coming lefty with a good bat, but he can’t run well enough to play outfield and is blocked by an injury-prone All-Star at first base. If he’s got a gun for an arm, you might consider putting him behind the plate just to get his bat in the lineup. However, the investment in training a lefty to catch probably isn’t worth the effort. Catcher is not an easy position to play, and it takes a toll on the bodies of the men who play the position. Why train a left-hander to catch when the odds are that he won’t have the role for long? The long-run returns to the player’s effort would be higher in working on other areas of his game, and the team would rather have its catching instructors working with long-run catching prospects. It’s just cheaper for the team to trade the player to a team without a logjam of players at first base, the traditional home for not-so-agile lefties.

We could conclude that the main reason that there are no left-handed catchers is that there simply isn’t a need to use that position to get a left-hander playing time. Even though the advantage of being right-handed at catcher may only be slight, there seems to be no real cost to keeping left-handers out of the role. Using only right-handed catchers is like locking your car door in a small town with no crime. The chances that someone will break into your car are tiny, but the cost of protecting yourself against the small likelihood of the negative consequences is so low (pressing the lock button) that it’s still worth locking the door. Similarly, in those rare instances where the advantage of having a right-handed catcher is the difference between winning and losing a game—runner on second base in the ninth inning of a one-run ball game—you’ll be glad you have one.