Safety In Numbers: Platoon Splits

facebooktwitterreddit

I’m not sure when it started, and not sure who came up with it at the time, but somewhere in the development of baseball and strategy, somebody decided that when you have a right-handed pitcher on the mound, a left-handed batter seemed to do better than a right-handed batter.  Likewise, they must have also noticed that a right-handed batter seemed to hit better with left-handed pitcher throwing than a righty.

Since that bygone era, the platoon split has become gospel. Managers will adjust their lineup to move batters around to avoid a run of left-handed batters, lest a lefty reliever come in and wipe them out in a row.  In part, that’s a strategic game designed to either force a lefty to throw to a righty or to force the opposing manager’s hand to make a pitching change.

In any case, I’m going to take a little look at the extent of the splits and also look at why they develop that way.

Let’s consider the numbers:

SplitPAABH2B3BHRSOBAOBPSLGOPStOPS+
vs RHP as RHB7073563797162163111266182013334.254.313.397.71095
vs RHP as LHB6286355471145092932404155311309.262.337.413.750106
vs LHP as RHB3735233106868718651289236371.262.333.410.744104
vs LHP as LHB14575129793142578683173292.242.313.370.68388

Provided by Baseball-Reference.com: View Original Table Generated 1/12/2010

Those are statistics from the entire 2010 Major League season.

Right away this shows a few interesting things. First, you should notice that the idea behind platoon splits looks to be valid.  In cases where a batter faced an opposite-handed pitcher, their OPS was above the league average.  This occurred in about 54% of all plate appearances in 2010.

Next, you see that other 46% of the time when a batter faced a pitcher with the same handedness as themselves.  In those cases, right-handed batters performed better against righties than left-handers did against lefties.  It’s a decent, but not overwhelming difference.  On base percentage was exactly the same, so the difference in OPS comes more from batting average and slugging percentage.

This kind of thing is where the marriage of baseball and statistics really gets interesting.  In this sample, we’re looking at 185525 plate appearances – a huge batch of results.  It’s large enough that even the best matchup – left-handed batter vs. right-handed pitcher – is just 6% better than average.  That’s a fairly stable bunch of data and a lot of the numbers are very similar, with specific exceptions.

For instance, the opposite-handed splits are nearly identical in every way.  A mere .006 in OPS separates the two matchups.  Considering plate appearances per home run, a right-handed batter against a left-handed pitcher hit a homer once every 40.48 plate appearances.  The opposite matchup produced a homer every 40.47 appearances. A difference of 00.01 PA/HR.

That’s just nuts.

But that’s just the numbers from 2010.  Sure, it’s a lot of appearances but it’s just one season.  Big deal, right?

2009:

SplitPAABH2B3BHRSOBAOBPSLGOPStOPS+
vs RHB as RHP6857961508157903090280186312668.257.319.407.72593
vs LHB as RHP6725059021158853225452178711731.269.347.430.777107
vs RHB as LHP35982317558503178415310185925.268.341.430.771105
vs LHB as LHP15249135653346638643743267.247.317.386.70388

Provided by Baseball-Reference.com: View Original Table Generated 1/13/2011.

Almost the same percentages – the R/R matchup favored pitchers a bit more and the opposite hand splits were a bit more juicy, but the general idea is the same, even down to the similarity in production from lefties vs. righties. The PA/HR differential was a little larger in favor of the RHB/LHP matchup than its counterpart (37.63 vs. 35.35) but not by a very significant margin.  In both cases of favorable platoon matchups, the slugging percentage in 2009 was .430.

Without getting into too many past seasons, I think it’s safe to assume that these kind of splits are fairly constant year to year.

In all cases, the matchup of lefty vs. lefty lagged behind the others.  It’s often the matchup that gets the most attention when determining lineups, player substitutions and other strategic in-game maneuvers.  It’s correct to do so, based on the numbers, but it’s not as drastic as it might seem.  Still, it’s clear that the platoon advantage does exist.  The lefty/lefty matchup is 12% less productive for a batter than average, but relative to other matchups, a LHB’s OPS against a LHP is just 3% smaller than a RHB/RHP matchup.  That makes it a valid measurement over the course of a large amount of plate appearances, but it seems like a small difference in one instance in one inning in one game.

More interesting to me, the more I thought about that, was why that would be the case.  If there was a sharp divergence between a left-handed batter’s production against a right-handed pitcher relative to the opposite matchup, I may not have thought about it, but over both 2010 and 2009, those matchups were nearly identical in overall production in every way.  So what makes the difference in same-handed matchups?

Let’s say you’re standing in a batter’s stance – we’ll assume you’re right-handed in this exercise – and I’m standing two feet in front of you.  Now, if I bring my right hand forward to grab your front shoulder, my right hand has a shorter distance to travel to get there than my left hand.  Also, if I give you the opportunity to bring your back hand up to block me from grabbing your shoulder, you have an easier path if I’m trying to do so with my left hand than with my right.  You’d have to move your back hand faster and at a more pronounced angle to get to my right hand than you would my left.  Meanwhile, my left hand has to cover additional distance and you have more time to see it and react.

It’s similar with a batter. When a batter stands at the plate, the bat is on their back shoulder.  As they swing, the bring it forward towards the ball.  We’ve all seen it a billion times at least.  When the ball is leaving a pitcher’s right hand, to hit it well, a right-handed batter has to open up a bit quicker to get the bat around and present more of the “sweet spot”.  It’s similar with a left-handed batter against a lefty. Take this diagram for example:

In this case, if a right-handed pitcher is throwing, the ball will come in at an angle that’s less likely to get a lot of the sweet spot from a right-handed batter’s bat (and likewise from the lefty/lefty matchup). It’s similar to how a player shoots a basketball.  You can still make a basket with a shot with a low arc, but you have to be much more precise about it.  A basketball rim has enough clearance to fit about three basketballs in it, so to take advantage of that width a higher arc is best so that the trajectory puts it at a better angle to go straight down through the hoop.

That relates to the bat striking the ball in a similar way.  If you can turn around fast enough, you get a wide enough plane of the bat on the ball, but you have to be quick and have to get it just right.  Too quick and you get the first bat in the diagram – you hit a small patch of the sweet spot and don’t have all your swing going into the ball.  Too slow and the ball’s on top of you and it glances off the bat.  When you get it just right, you get a wider patch of the sweet spot on the ball. Now when the pitcher throws from the other side, it’s coming in at a better trajectory for a swing.

A right-handed batter waiting on a left-handed pitcher can get a better angle (unless the pitcher throws a sick Rivera-like cutter) when the ball comes in.  The pitcher has the decision to go inside or outside, but if they go inside, it’s coming in at an angle that’ll meet a well-timed swing.  If they go outside, they might cover too much of the plate and too much of the batter’s swing.  It’s a small difference, but I think it’s there.

Also to consider is the sight of the batter.  If a righty is against another righty, as the ball is delivered, it’s more likely to be obscured until just before the ball gets delivered.  It might just be a split second more, but against a lefty, a right-handed batter can see a bit more of the ball as it’s delivered.  That extra bit of pitch recognition can make the difference. I think that accounts for the decreased hitting performance in same-handed matchups.  But then why do lefties hit lefties worse than righties hit righties?

In this case, it’s demographics.

About 7-15% of the world’s population is left-handed and while baseball has a higher number of left-handed performers than a general population (that’s market demand there), they still make up the minority of players.  In a larger scale, with smaller left-handed populations, you’ll have a smaller number of baseball players developing out of that group.  For the most part, you can’t be just any sap off the street who happens to be left-handed and get a major league tryout, so you have to also have talent, a work ethic, and the ability to perform at a high level.  That narrows things down quite a bit.

A left-handed batter coming up is then less likely to see a left-handed pitcher than a right-handed one.  When they do face a lefty at the prep level, perhaps, that pitcher might not even be good.  They spend their early years crushing righties and destroying bad lefties, but when they get to a professional level, even up to the big leagues, the platoon split generally still takes effect.  In 2010, left-handed batters made about 42% of the plate appearances for the league.  Left-handed pitchers only faced about 28.5% of batters in the league.  Even though lefties are more prevalent in baseball, they’re still much more scarce, and that has to have an effect on a batter.

If you face a good lefty less frequently coming up, you don’t have the same comfort level a right-handed batter might who’s faced a right-handed pitcher about 3/4 of the time (or more).  Anyone who isn’t sure there’s a necessary comfort level can just ask Larry Walker or John Kruk how they felt facing Randy Johnson.

Baseball is a game of small advantages that build up over the course of a long season, a long career, a long era.  If a batter’s 2% less comfortable in the box against a pitcher of the same handedness, that’s going to impact the final results over the course of nearly 200,000 plate appearances for a league.

To that same end, with the idea in place that a lefty struggles against other lefties, baseball has evolved a specialized role for left-handed pitchers.  Whether that pitcher’s individual number suggest he’s more effective against left-handed batters, if there’s an important moment in a game and a left-handed batter at the plate, the manager is instinctively going to call for the southpaw.  It’s a percentage play.  What evolves from that is the LOOGY (left-handed one out guy) who exists solely to get David Ortiz or Joe Mauer out.

Generally, you don’t see that level of specialization from a right-handed pitcher and a manager is more likely (both by the constraints of his bench and by habit) to leave a right-handed batter  to face a righty (the R/R matchup happened in 38% of plate appearances in 2010, highest of the four matchups), even if it’s an important situation and often even if there’s a left-handed batter available.

Conventional wisdom says that these matchups are important, and the numbers support the idea.  Pitchers throwing to batters that share their handedness have an advantage, even if it’s not very dramatic.  Over time, it adds up.  It’d be interesting to see if left-handed batters would improve their OPS if they faced lefties more often over time (and got more comfortable) or if there is just something about a left-handed pitcher that would cause that performance to decline even further.

Still, if you’re a new parent out there, get your kid throwing left-handed, and they might have a shot to pitch out of the bullpen for 25 years.  Southpaws have a home in every bullpen.

You can stay current on all the Call to the Pen content and news by following us onTwitter,Facebook, or by way of our RSS feed. Michael Engel is the lead writer for KingsOfKauffman.com, a Kansas City Royals blog on the Fansided network