The 2023 Twins surged to the AL Central title with a 42-29 second half. That stretch of strong play helped them go from a half-game out of first place in the standings at the All-Star break to winning the division with a nine-game margin at season’s end. Sparking that run was a resurgent offense that averaged 4.31 runs per game in the second half, after struggling to just 3.88 runs per game in the first half. Much of that improvement is attributable to four young hitters – Royce Lewis, Edouard Julien, Matt Wallner, and Alex Kirilloff – who stepped into regular roles and performed like seasoned veterans.
Heading into 2024, each of those four figures to again play prominent roles. They each also have opportunities for further development. Lewis and Kirilloff have been plagued by injuries. Julien and Wallner face persistent questions about their defense. Kirilloff, Julien, and Wallner — all left-handed hitters — have unanswered questions about their ability to hit left-handed pitching.
While all three have been heavily shielded from facing many lefties thus far in their major league careers, their results in the limited opportunities they have received hardly scream for them to be deserving of more:
I tossed in Trevor Larnach — another young lefty hitter — because this also applies to him. You can see that all four hitters have been much worse against lefty pitching.
With each of them in the club’s plans to varying degrees, I wondered if that was something that gets better with age and experience. Do young hitters learn to close their platoon splits?
The idea of the platoon split – that most batters hit better against opposite-handed than same-handed pitching – has been around in baseball for a long time. Baseball-Reference, in its summary of platooning, points out that switch-hitting, which allowed hitters to more or less always see opposite-handed pitching, dates back to the National Association in the 1870s.
In 1914, the Boston Braves, perennial bottom feeders managed by George Stallings, became what’s believed to be the first team to employ a platoon strategy with their starting lineups and surprisingly won the World Series over the heavily favored Philadelphia Athletics. The strategy caught on, as things like that tend to do, and it has been a part of the game to various degrees ever since.
Casey Stengel’s Yankees of the late 1940s and 1950s liberally deployed platoons, as one example. It was in that same general time frame that Dodgers’ boss Branch Rickey hired Allan Roth for statistical analysis and it’s thought that Roth’s work was the first to quantify the platoon effect.
Historical Platoon Splits
Statcast has data that goes back to 2008 and I’ve plotted the league average weighted on-base average (wOBA) by handedness splits for each season from 2008 through 2023 below.
You can think of the black and grey lines as when the batters are facing opposite-handed pitchers (i.e., right-handed batter facing left-handed pitcher, or vice versa) and the red shaded lines as when the pitcher and batter are of the same handedness:
The data is clear that there is an advantage for hitters when they are facing opposite-handed opponents. For left-handed batters, the average advantage when facing righty pitching over this period has been about 28 points of wOBA. For right-handed batters, it’s been about 16 points.
Those numbers align well with rigorous research findings of chapter three of The Book, which found platoon effects of 25 points and 14 points of wOBA on 1999-2002 data.
This effect is due, in large part, to same-handed, horizontal-moving pitch types, like sliders, sweepers, and sinkers as you can see from this table borrowed from Eno Sarris’ article about sweepers and platoon splits for The Athletic last year:
While pitch type drives a lot of this, there’s also something to be said for handedness alone. Look at the fastball row. Even the straightest, theoretically easiest-to-time pitch type shows a slight platoon advantage for the pitcher.
Taking this a step further, I calculated the platoon splits for every player that met certain plate appearance thresholds since 2008. For right-handed batters, the minimums I used were 1,000 PAs vs right-handed pitching and 500 PAs against left-handed pitching. For lefty-swingers, I used 500 PA against right-handed arms and 250 PAs against southpaws.
Those benchmarks gave me a dataset of 287 right-handed batters and 235 left-handed batters.
What I found was that it’s not just the average player who has a platoon split. The effect applies to almost every player.
Just more than 84% of the right-handed hitters showed a platoon split of any value and more than half of them (52.6%) had a split that was greater than 20 points of wOBA. For the lefties, 92.3% had a split of some sort, and 54.4% had a split that was more than 35 points of wOBA.
Here’s the summary:
The Southpaw Disadvantage
By now, you might have noticed the disparity in the size of splits between righties and lefties in the data I’ve shown. While the effect is significant for both, it’s quite a bit worse and more prevalent for left-handers.
Why is that?
Fortunately, that’s a question that has been investigated, not only in baseball but across many other sports and in the natural world, where left-handers seem to have an advantage. Consistently, studies have found that left-handers seem to benefit from an “unfamiliarity bonus,” especially in arenas where response times are very short — like hitting a pitch in baseball.
As Guy Molyneux and Phil Birnbaum, the authors of article in the the first link in the paragraph above, explained:
Scientists believe this pattern reflects a “negative perceptual frequency effect,” meaning that because athletes confront left-handed opponents much less often, their ability to perceive, interpret, and react to these opponents’ movements is less developed. A number of experimental studies in which athletes watch video and assess the direction of an oncoming ball support the theory:
The outcome of balls struck by left-handed volleyball players’ actions was significantly less accurately predicted than the outcome of right-handed attacks;
English Premier League goalkeepers are less adept at predicting the trajectory of balls kicked by left-footed penalty kickers.
Right-handed hitters get significantly more practice and exposure against right-handed pitching than left-handed hitters get against left-handed pitching, starting back in Little League and continuing through the professional ranks. Since 2008, per Statcast, 72.4% of all pitches thrown in the major leagues have come from right-handed pitchers.
In addition to those study results, both the article excerpted above and this one by Ben Lindbergh show that despite left-handed pitchers’ “stuff” being less imposing— as measured by velocity, spin characteristics, and movement — than their right-handed counterparts, their performance results in terms of allowing runs are more or less the same.
That gives credence to the unfamiliarity bonus idea, and Molyneux and Birnbaum estimated that the unfamiliarity bonus for left-handed pitchers is 0.75 runs per nine innings. (That estimate is against all batters, not just lefties.)
That bonus effect, when combined with the baseline disadvantage of hitting same-handed pitches and having to deal with horizontally-moving pitches is why left-handed batters’ platoon split is larger than right-handed batters. Lefty hitters are, in a way, extra-disadvantaged when facing left-handed pitching.
Can They Become Familiar?
The suggestion that left-handed hitters are disadvantaged as a matter of familiarity also suggests hitting left-handed pitching effectively is something that can be learned with sufficient repetition and opportunity. What does the data say about that?
One way to investigate that question is by reviewing the past two decades of Twins’ developed left-handed hitters. There have been several who have come up through the Minnesota system and then spent significant time on the major league roster.
The table below compares their early career splits against the rest of their careers. The rows defined as “early career” are my arbitrarily chosen cut-points — usually the player’s first two or three seasons. (Whatever was the most convenient with FanGraphs’ splits tool and where they had some reasonably sized sample of a few hundred plate appearances.)
I don’t see much of a clear pattern in that data. Mauer, Morneau, and Kepler narrowed their splits but still had larger-than-average differences even after improving. Kubel and Span’s splits got larger over time. Revere, Rosario, and Arraez more or less stayed the same. On the whole, it’s kind of a wash.
More broadly than the Twins, Shane Tourtellotte put together this 2019 Hardball Times article that looked at whether batters learn to narrow their splits as they age using all players who debuted between 1998 and 2003. Using their cumulative data through 2017, Tourtellotte concluded that they — both right-handed and left-handed — not only do not narrow their splits but instead see their splits grow larger over time.
The New Frontier
Before we take those findings to mean hope is lost for Wallner, Julien, Larnach, and Kirilloff it’s worth considering how hitter development is changing thanks to technological advancements. While much of the benefit of the recent advancements in data and technology has gone to the pitchers, things on the hitters’ side are starting to catch up.
Consider that for all of baseball history, the only way for a young hitter to get reps against game-like pitching has been to play in a game (or a simulated game) and face a live pitcher. Batting practice, as ubiquitous as it has been, has always been very little like facing live pitching.
That reality presented a challenge for clubs hoping to develop young hitters and win games simultaneously. Should they give their youngsters opportunities, even in unfavorable matchups, knowing that they would likely struggle and hurt the team, but that it might pay dividends years down the road? Or should they shield their young hitters from unfavorable matchups and try to get the short-term benefit of a better matchup with another player, at the expense of the young hitter’s development?
Now, another option is emerging because of data-powered products like iPitch and the Trajekt Arc Pitching Robot — smart pitching machines that can replicate the experience of facing just about any MLB pitcher, from velocity to movement, to release point and how the ball looks leaving the hand, even a specific pitcher’s delivery.
These machines are trying to make it so that in-game at-bats are no longer the only way for hitters to get live reps. That could prove critically valuable for young hitters, especially left-handed hitters who need exposure to left-handed pitches.
The Twins are known to be one of the MLB teams that use Trajekt and other similar products, including in the minor leagues, as Jeff Johnson reported about Cedar Rapids last summer. If lack of familiarity is one of the driving reasons the platoon split exists in the first place, here is potentially a new way to make up for that deficiency.
Historically, most hitters don’t close their platoon splits. Maybe in the future, they will.