Major League Baseball

Should pitchers bat eighth?

Updated Mar. 4, 2020 8:26 p.m. ET

share

facebookxredditlink

foxsports

Every season has its sabermetric bellwether issue. Trout vs. Cabrera. The infield shift. Catcher framing. Joey Votto in the two-hole. But before all that, there was Tony La Russa hitting the pitcher in the eighth spot in the lineup. La Russa, when he managed the Cardinals, was known to be willing to experiment a bit to gain an edge. Then again, during his A’s days, La Russa was credited with “inventing” the modern bullpen and Dennis Eckersley. In 1993, he even tried a pitching strategy that had three groups of three pitchers each that worked a three-day rotation. The experiment lasted a week, but he gave it a shot. But now, the La Russa gambit of hitting the pitcher eighth is back.

Joe Maddon, a National League manager for the first time in his career, seems to be sticking by the strategy in the early going. Maddon comes with a pre-installed reputation as a free thinker, so maybe it’s not surprising that, on Opening Day, Jon Lester hit eighth for the Cubbies. But that doesn’t explain how Jacob deGrom ended up in the eighth spot of the Mets lineup last week. (Actually, Terry Collins put the pitcher eighth three times last year.) Or why Walt Weiss put Tyler Matzek in the eight hole — against Maddon’s Cubs!

Hitting the pitcher eighth seems to defy conventional wisdom. The pitcher is almost always the weakest link in the lineup, and since there are nine spots, he should be in the one that is least likely to come up, right? It seems to make sense, but like a lot of things in life, it makes sense until you think a little deeper about it. A manager usually puts his best hitters at the top of the lineup, hitters who are usually good at not making outs. The problem comes from the fact that in baseball, a lot of scoring depends on stringing a couple of hits together consecutively before the out clock runs out. Home runs are nice because the batter can do all the work of getting around to score on his own, but home runs don’t really happen all that often. On average, they happen in one out of about 40 plate appearances, and even the really gifted home run hitters only hit them 5 or 6 percent of the time. To win baseball games, a team needs to have runners on to knock in.

The problem with the pitcher hitting ninth is that those really good hitters find themselves without runners to knock in and with more outs on the scoreboard when they come up, because the guy in the No. 9 spot, the pitcher, is generally a bum with the bat. In fact, in 2014, pitchers in the NL put up a combined .124/.156/.155 line. Even lowly No. 8 hitters managed a combined slash line of .233/.299/.326, and while that’s nothing to crow about, it’s a lot better than the pitcher line. The eight-hole guy gets on base almost twice as often as a pitcher. Maybe it makes sense to put the actual major-league hitter in the nine spot. Yes, the pitcher would come up slightly more often, but it’s more likely that the top of the lineup would have a runner on to work with.

It turns out that batting the pitcher somewhere other than ninth is historically pretty rare. It didn’t appear in the modern (post-1900) era until 1916, although in four years 44 games featured this lineup quirk. Almost all of them featured either Walter Johnson or some Red Sox pitcher who eventually got traded for an off-Broadway musical and wasn’t much of a pitcher after being traded. And then the idea went into hibernation. The 1920s saw a total of four games with a non-ninth pitcher, the 1930s saw nine such games and the 1940s featured only two games with the pitcher out of the lineup cellar. In the 1950s, though, pitchers hitting ninth became a little more optional. For example, in 1952, there were nine times where the manager wrote the pitcher’s name somewhere other than the bottom, and that manager was Lou Boudreau of the Indians each time. By 1957, the pitcher hitting somewhere other than ninth happened an astounding 66 times with Boudreau, Casey Stengel and Bobby Bragan experimenting with the idea. And then just like that, it vanished again with not a single instance in 1958 or 1959. It happened a total of five times in the 1960s and 1970s combined and not at all in the 1980s. It didn’t happen again until 1998, when La Russa brought the idea back. While La Russa did it consistently in ’98, he (or anyone else) didn’t again until 2007. But La Russa is by far the patron saint of the movement. Since 1900 (through last year), there have been 803 games that featured a pitcher penciled into the lineup somewhere other than ninth. Tony La Russa wrote more than half (423) of those lineups.

But is it a good idea? This isn’t a new question. The fact that La Russa was hitting the pitcher eighth nearly two decades ago means that someone noticed and ran a study, in this case John Beamer at The Hardball Times. He found that in terms of scoring runs, the benefit that a team gets is small, but since there’s no cost to doing it, why not?

Warning! Gory mathematical details ahead!

We know that lineup construction itself is vastly overplayed. In a study that is not quoted nearly enough, Tom Ruane found that assuming the same nine players, the difference between the perfectly optimized lineup and the “He might actually be trying to sabotage his own team” lineup is actually pretty minimal. People place far too much emphasis on lineup construction relative to the added benefits that it can bring to a team. Still, if it’s just as easy to hit the pitcher eighth as ninth, you might as well pick the option that brings back the most runs, even if the effect is slight. I pulled out my lineup simulator and created a lineup composed of the composite average of a 2014 MLB leadoff hitter, a No. 2 hitter, a No. 3 hitter, etc. and then put a composite major-league pitcher in the ninth spot. The model that I use is a Monte Carlo Markov simulator, which essentially uses a bunch of dice rolls to simulate a baseball game. It includes parameters for speed and for baserunning advancement and double plays, but doesn’t include any context (what the score is). It’s not perfect, but it gives us a rough idea of what we can expect a specific lineup to produce. I ran that model through 100,000 simulated nine-inning games. Afterward, I flip-flopped the pitcher and the eight-hole hitter.

The results (runs per nine inning game):

Pitcher Hitting 8th – 3.7118

Pitcher Hitting 9th – 3.7079

We see that the pitcher hitting eighth wins … once you get to the third and fourth decimal place. Still, it’s a victory. Over the course of 162 games, a team using the “Pitcher Hits Eighth” strategy would be 0.6 run better off. Not 0.6 win. Run. And at that, some of that is blunted by the fact that rarely do teams let their pitchers bat past the sixth or seventh inning (my model plays all nine), so the difference between the two models should probably be chopped by about a third. Additionally, in this simulation, the pitcher was always swinging away. Managers often bunt with the pitcher up, so that’s going to blunt the differences a bit more. So, really we’re talking about maybe half a run over the course of a season by hitting the pitcher eighth. Still, that’s better than nothing.

If we assume that there’s no cost to hitting the pitcher eighth, then we can stop there. But it’s worth noting that there is a cost. And it’s one that anyone who’s ever played a baseball simulation as a National League team is well aware of. Or someone who doesn’t like the DH. It’s that decision. In fact, when anti-DH advocates make their case, they often point to the delightful agony of this particular situation. It’s the sixth inning and you are down 1-0, but there are runners at second and third with two outs and the pitcher is due up. He’s only given up the one run and could keep going if you needed him to. He’s not much of a hitter, so chances are that sending him up there kills this rally. But if you pinch hit here, you lose the pitcher and have to hope that your bullpen is solid. And that the pinch hitter doesn’t just pop out to second and make you look like an idiot. Now batting… who?

Let’s for a moment take a look at that decision. In games played under National League rules from 2010 to 2014, let’s take a look at how often it happened that the pitcher’s spot in the lineup came up and how often the pitcher was allowed to hit for himself.

Some of these may be relievers. Some of them may not be the nine-spot, because of double switches, but right now I just want to establish something. Look at the “pitcher hit for self” column. We notice that early in the game, it’s pretty much a given that the pitcher will hold a bat when his spot comes up. In the fifth inning, it becomes less automatic, and by the sixth inning, we’re starting to approach an even-up shot. In the seventh inning, the rate is down to roughly one-quarter! So, it’s the fifth and sixth innings where that decision gets a little more agonizing. We’ll focus on innings 5, 6 and 7, because that’s when it’s actually a decision.

Now that decision is a pretty big one. A manager has to decide either to let a guy who is basically an automatic out hit instead of a perfectly good pinch hitter on the bench, or he has to add another inning to the bullpen’s workload, both today and in general for the season. Maybe someone has to get four outs today instead of three or maybe the fourth-best reliever has to pitch today. We know that the bullpen overall is worse when a pitcher doesn’t go as deep into a game, and there are some days where that fourth-best pitcher just isn’t as good as that starter, even in the seventh inning. But I guess that’s NL baseball for you. Pretty much, no matter what he does, the manager gives up value somewhere.

Let’s look for situations that could make for an interesting managerial decision. The pitcher’s spot is due up. His team is losing, although the game is close (within two runs). Our pitcher has only given up three or fewer runs. This is a situation that a manager faces in 13.5 percent of all fifth innings, 4.9 percent of all sixth innings and 5.9 percent of all seventh innings. In other words, a manager is going to have multiple games during the year in which he has to make a choice between some much-needed offense or keeping the pitcher in the game and where the game could be very much in the balance. He has to make that decision.

That’s how often a manager has to make that decision with the pitcher’s spot, generally the nine-hole. But how often does this set of circumstances visit the eight-spot in the lineup? A quick comparison:

The eight spot in the lineup gets one of these close-game-should-he-or-shouldn’t-he situations in the decision zone (innings 5-7) in roughly 2 percent more games than does the pitcher’s spot now. If the pitcher is batting eighth, that means that the manager will have to make that decision two or three more times per year than he otherwise would have to. That’s roughly two or three plate appearances that will likely be moderately high-leverage situations (it’s getting later, the score is close, there might be runners on) in which the manager might have to send up an automatic out to the plate or he might have to sacrifice his pitcher before he otherwise would. In some cases, the pitcher’s pitch count might make the decision for the manager. He might be pitching well, but why risk an injury? But even if we allow that one of those situations might have the decision made for him, we still have one or two extra times where the manager is in a tough spot tactically, all because he hit the pitcher eighth and that decision sneaked up on him. If he had hit the pitcher ninth, he wouldn’t be in this spot.

We assume that a team clears roughly half a run over the course of a season by having a better hitter in the nine spot, but it’s about to give some of that back because of these extra tough decisions. If the manager hits the pitcher ninth, then in these extra situations, it would be the eighth-spot hitter at the plate, rather than the pitcher. (Note: I looked to see how often real eighth-spot hitters are intentionally walked in these situations. The answer is — surprise! — only 1.2 percent. Even if you look at situations with runners in scoring position and two outs and/or first base open — again, in a close game with the batting team trailing — the IBB rate is only 4.9 percent. Intentional walks aren’t as big a factor as I figured they’d be.) The average eight-hole hitter in 2014 had a .143 point edge in OBP over the average pitcher. Changing an out into some sort of on-base event is worth roughly a run in context-neutral value, although that value will be magnified because we’re likely in a higher-leverage situation. So, we’re looking at a loss in value of roughly .15 run each time one of these “Oops, had to make that decision a little too early” situations happens, and that’s before we adjust for the context of the game itself. Then, there’s some value in being able to send the starter back out for another inning. All of a sudden, most, if not all, of what little value there was to be gained from hitting the pitcher eighth is gone.

Sometimes “different” is different from “good”
I get the fascination with pitchers hitting eighth. It’s new. It looks daring. It's just that once you take everything into account, it doesn’t really buy you anything more than a little cool quotient. It doesn’t really help all that much in a best-case scenario, and it doesn’t hurt all that much if everything goes wrong. Compared to a traditional pitcher-hits-ninth lineup, it’s pretty much break even. The biggest effect might be that it probably annoys the guy who has to hit ninth behind the pitcher. If there’s something to be said for it, it’s an aesthetic choice, really. Pitchers hitting eighth is the bumper sticker of baseball. It doesn’t really affect how the car drives overall, but it’s the thing that people will remember about you if they pass you on the highway.

Russell A. Carleton is an author of Baseball Prospectus. Follow @pizzacutter4 
Click here to see Russell's other articles. You can contact Russell by clicking here.

share

Get more from Major League Baseball Follow your favorites to get information about games, news and more

in this topic

Major League Baseball