May 10, 2013
Extending the Process
Much of statistical analysis in baseball involves the study of outcomes. Hits, walks, strikeouts—these are the results of what an athlete accomplishes on the field. The focus of scouting and coaching, by contrast, is on process.
In my evaluations of pitching mechanics, the outcomes that get most of the attention are those that immediately result from the pitcher's delivery, such as velocity, movement, and command. By extension, a pitcher's rates of walks, strikeouts, and home runs allowed have been labeled the Three True Outcomes due to the lack of influence by fielders on those events. However, only 30 percent of plate appearances end with one of these true outcomes, and to ignore the other 70 percent of plays that do involve the defense is to paint an incomplete picture of a pitcher's performance.
If the lessons of pitching mechanics carry any weight, then over time we should be able to see their impact on box scores beyond the Three True Outcomes. With this in mind, I decided to challenge a major tenet of the philosophy that I have been advocating for the past nine years: namely, the advantage of a deep release point. The functional benefits of a deep release point have been established by the general principles of physics, including a higher perceived velocity from the hitter's point of view as well as later break on pitches with movement. The second point is key; the absolute timing of break may not change with respect to pitch release, but an extended distance will allow the baseball to creep closer to the hitter as the spin dynamics take effect during the ball's flight path, effectively shrinking the window of time in which a batter must identify the incoming pitch.
Pitchers with great extension at release point can be expected to induce weaker contact, on average, but to assess such a skill without the benefit of HITf/x requires that we dive into the murky waters of batted-ball data. The multitude of confounding variables creates a volatile cluster of data points, but perhaps we can learn something from the extreme outliers.
Batting Average on Balls in Play
(Stats thru 5/8/13, min 40 IP)
Pitchers who can consistently induce weak contact will make life easier on the fielders behind them, allowing for greater defensive efficiency and thus a lower batting average on balls in play. However, the use of BABIP obscures the related ability for a pitcher to generate more strikeouts with a deep release point.
Consider the classic example of Randy Johnson, whose long limbs combined with great momentum and near-perfect posture to produce possibly the greatest release distance of the modern era. Opposing lefties had virtually no chance to distinguish the Big Unit's slider from his fastball before they had to initiate a swing. The impact of his extension was not evident in Johnson's BABIP, with a career mark of .295 that perfectly matched the league average during his 22-year career, but his ability to sap power was reflected in a .353 opponent slugging percentage, 61 points lower than the league standard.
Opponent Slugging Percentage