David Phelps and the hybrid pitcher, part 1

Heuristics: An Introduction

Heuristics, in the field of Psychology, are often seen as "mental shortcuts". These shortcuts are the brain’s way of quickly determining solutions that ease a cognitive load. They can be made both consciously and subconsciously. We frequently utilize heuristics without even realizing it. The "Rule of Thumb", stereotyping, profiling and "common sense" are all forms of heuristics. While the benefits of heuristics are numerous, heuristics do have a tendency to cause cognitive errors.

In baseball, heuristics are frequently used in profiling players. We base past profiles of particular players to judge the value of current players. This can sometimes lead to categorizing certain players erroneously. Placing them in box that mislabels them and hides their true value. Manifestations of these categories from a baseball perspective include the terms "Quad-A" players, and "failed starters". On a basic level, our brains go through a process of:

"A" resembles "B"

"A" falls into "Category 1"

Therefore "B" must also fall into "Category 1"

The brain, being the incredible machine it is, organizes the information within milliseconds of registering it. It’s important to note that while that relationship might prove true the majority of the time (hence the usefulness of heuristics) baseball as a context-based sport doesn’t always allow for such tidy categorizations.

Categorizing Pitchers

Heuristics are frequently used in categorizing pitchers. In baseball, the basic delineations are: starting pitchers and relief pitchers. While there are various degrees between them (closers, LOOGY, spot starters, etc.) those are the basic categories by which we organize pitchers. But what about pitchers who fall smack dab in the middle? Pitchers that are used (effectively) as relievers AND starters. Do they even exist? Or are pitchers that oscillate between roles merely "failed starters"? Good enough to remain on the staff but sub-par when it comes to the rotation.

Below is a list of pitchers over the past five seasons that have accumulated multiple years of at least 10 games started and 10 relief appearances (RAPP):

As you can see, the list is short. Over the past five seasons, just 11 pitchers have accumulated multiple seasons of at least 10 starts and 10 relief appearances. Looking at the names, it’s a fairly diverse group. So, how then can we pull the list apart further to identify which pitchers profile as "failed starters" and which actually profile as "true hybrids"?

Immediately we should consider eliminating Tim Wakefield. Wakefield really doesn’t qualify for either term due to his knuckleball sorcery. Out of his 19-year career, Wakefield put up 13 years as a quality starter. Interestingly enough, there was a stint from 1999-2002 where Wakefield put up terrific numbers as a late inning reliever. But due to the fact that he primarily worked as a starter for the bulk of his career (and his aforementioned knuckleball-ness), I’m disqualifying him.

Moving forward, another way to identify a "failed starter" from a "true hybrid" is to identify the context by which they attained their roles. Pitchers that start the season in the rotation but inexplicably end the season in the bullpen could be seen as "failed starters". Especially if the move to the pen is caused by performance issues rather than injury. That, from my estimation, eliminates Duensing, Zambrano, Williams, and Gorzelanny. Which leaves Wood, Chen, Villanueva, Phelps, Medlen and Kendrick as "True Hybrids" (Though Medlen moved on to a full time starter position in 2013.)

The essence of a "true hybrid" must be more than context however. A hybrid must be effective both as a starter AND a reliever. They must possess a starters consistency with a reliever’s flexibility. They provide necessary depth to a pitching staff beyond the role of an occasional spot starter. "True hybrids" provide long term starting capabilities when the unexpected happens. They can simultaneously stabilize a rotation in a pinch while seamlessly creating a formidable bullpen at the drop of a hat. "True hybrids" are rare breeds and they are often mischaracterized as "failed starters" due to cognitive errors caused by heuristic processes. We view their ever-changing roles as sign for "failed starter" when in actuality context plays a large part in their "hybrid-ness". Whether there isn't enough room on a rotation or a managers preference of versatility, "true hybrids" are often lumped in with "failed starters."

Next week we will delve into the aforementioned list of "True Hybrids" and determine their value from a statistical standpoint. We will then characterize whether or not David Phelps should qualify as a "True Hybrid." From that point, we should be able to determine his value and role on the Marlins specifically. See you next week for Part II of this series!