Baseball Theory: Rotation Identity
In a traditional MLB pitching staff, specific roles are designated to pitchers regardless of whether they are a starter or reliever. For years teams have focused on filling the identities of those roles on their roster in a checklist fashion. Such examples of identities include “workhorse”, “innings-eater”, and “long-reliever.” But as baseball becomes more progressive, pitcher identities are beginning to lose meaning for how they are used. Which calls into question, what is the performance and identity of a #3 starter? Why does a team need a “#3 starter” identity in a pitcher and what does it exemplify? Why can’t rotations operate without identity akin to a positionless NFL offense? Teams like the Tampa Bay Rays are challenging this through their organizational philosophies that focus on innovation and improving the modern game.
An example is the Rays’ groundbreaking use of the “opener”. Used for the purpose to reduce the run expectancy of an opposing lineup in the 1st inning and innings after. The opener fuels the idea that giving the batter 1 PA to get acclimated to a pitcher reduces the likelihood of the opposing lineup producing and picking up patterns. Think of it as a parallel to NCAA Men’s Basketball and the NBA Draft. The wave of one-and-done players has increased from year to year and talent evaluators have to decide from ONE season of playing if a player is worth their draft capital and millions of dollars. More often than not they don’t feel 100% comfortable that their investment will pay to their expectations. The parallel shows that the reduction of sample size makes an achievement of the end-goal less probable (getting on base or drafting an above-average NBA player). The goal I’m getting at with the breakdown of the opener is that identities are changing in baseball and are ever-evolving. Which consequently, will kill-off prominent baseball tropes in a rotation and bullpen as the goal is to create your own identities that will lead to winning and success, and not fill existing ones.
I believe that the opener will eventually be mass acceptance in MLB and will be used by a majority of the league. With this, I believe that the idea and identity of #3 starters need to keep pace with the progressiveness of baseball. Winning teams don’t explicitly care for having someone fill the established identity of a #3 starter, but merely someone who has little dropoff from their top starter’s performance.
Let’s assume that there is a world of baseball where the roles of #1, #2, #3, #4, and #5 starters need to be filled by players who fit the bill of being in those roles. Specifically, the #1 starter must be someone who is in the top 20% of pitching performance, the #2 starter being in the top 20–40%, the #3 starter being in the top 40–60%, etc. In theory, all SP staffs would be normalized and all would be on a level playing field; which isn’t an ideal way to achieve elite status as a ballclub. If the construction of an MLB SP rotation assumes that your #3 starter is merely a player who hovers around average, what is the appeal?
What I want to challenge is the idea that a rotation spot needs to be filled by someone who represents the status quo or identity of that spot. The idea and concept of a player with a “#3 starter ceiling and identity” is merely a facade for an identity without substance and meaning. Pitchers as a whole are improving every year and the concept of a fluid and “positionless rotation” is becoming more and more likely for major league baseball. A “positionless rotation” represents the future of an MLB rotation as it carries a higher ceiling than any traditional starter identity.
Extrapolation
While this is a theoretical and conceptual piece, I did want to show some graphs that display the unpredictability and identity of #3 starters. The graphs show that there is a wide variance of performance among pitchers in the #3 spot in a rotation and that there is room to innovate with how the spot is filled and what the true expectations should be. Nevertheless, a true #3 starter seems to be a pitcher who fails to generate a lot of whiffs and struggles to consistently limit hard contact. Essentially a true #3 starter is playing with fire in an age where hitters are striking out more and hitting the ball harder. While some can succeed on this traditional skill set based on their xFIP and WAR correlation, it isn’t desired for someone who is supposed to be your third-best starter.
Parameter and Sample Details:
For each team in the 2019 season, I selected the pitcher with the 3rd most starts on the season regardless of trade or injuries. I went off the idea that you will give your best starter the most starts as that will give you more opportunities to win a game, and so on for the rest of the rotation. In the event of a tiebreaker, I chose innings pitched.
I also chose a core group of Tampa Bay’s openers to compare the opener to normal #3 starters in the league. The openers were chosen by observing the starting pitchers for Tampa Bay in 2019 who averaged 2 innings or less per start. Indicating that they were used as an opener and not a typical starter. However, not all data from every single “opener” start was used as I just wanted to estimate the performance from the core group that TB used.
*Max Scherzer was exempt from this study even though he had the 3rd most starts for WAS as his numbers skewed the data too much.
*TB openers used were: Ryne Stanek, Andrew Kittredge, Diego Castillo, Hunter Wood, and Jose Alvarado.
Graph #1
Graph #1 shows the wide distribution of WAR and xFIP amongst #3 Starters and the openers. This shows that the performance of #3 starters vary across the league and how they compare to an alternative of a #3 starter, the opener.
Graph #2
Graph 2 shows how #3 starters compare to the openers to pitching to a three-outcome hitter; walk, strikeout, or homerun. So I chose to display the relationship of whiff rate and hard hit rate as a higher hard hit rate suggests that a ball is more likely to result in a HR. Keep in mind due to the nature of the opener being a relief pitcher, their hard hit rate may be inflated due to the psychological change in their preparation as a SP vs a RP.
All data courtesy of Fangraphs and Baseball Savant
