In "An Invention of Sorts - Part I" I introduced PAG, my Points Average per Game scoring method, and explained how this creation is a measure to help find statistical equals amongst hitters in the minor leagues, not just magnify the industries top prospects.
This week my focus is the extension of PAG called APPA, which stands for "Average Points per Plate Appearance." Basically, what PAG does for games, APPA does for plate appearances.
Make no mistake, PAG trends tell me it's working. After years of research however, I loosely concluded that although PAG is a good measuring tool, I wanted something more. Something to put PAG to the test. Something stretched to even more diversity.
As I pointed out last week, each defensive position carries its own offense calling card, but PAG scores, collectively, regardless of position, over the last several seasons, appeared to be stretched as far as they could go. Extreme PAG scores appeared to be reserved only for small sample sizes of limited at-bats in a season. Therefore, PAG juxtaposition with fundamental stats, albeit effective, needed something to help prove its worth. So, to the drawing board I went again.I tested the theory of micromanaging PAG.
It occurred to me that by breaking PAG down, I might find the wider statistical variance I was originally anticipating. So, I started by looking at the amount of plate appearances, not at-bats, per game for players who posted a seasonal 3.50 PAG score or above. With PAG, a total score (TS) is created which is divided by games played. With APPA, TS is divided by plate appearances. I first tested APPA during the 2014 off-season.
I was running stats for the Texas Rangers and ironically, at the same time, a former fantasy league mate sent me a text message asking for my thoughts on Rangers star 3B prospect Joey Gallo, and how he matched up to former Cubs top prospect Kris Bryant. Bryant is lauded for his power, but the 6'5" - 235 pound Gallo's mammoth home runs are near immortal and entice chuckles of amazement amongst both baseball purists and stat-heads alike.
His light-tower power is reaching near legendary proportions. One absurd long-ball in particular is humorously dissected here by CBSSports.com's David Brown.
My casual breakdown turned into instant pressure, as a friend wanted a fantasy baseball question answered. I ran Gallo's and Bryant's APPA numbers on the fly. Much to his delight, I chose Bryant. Not because my friend is a die-hard Cubs fan, but because Bryant outscored Gallo in both PAG and the newly tested APPA.
Bryant posted a 138 game PAG score of 4.74 with an APPA of 1.10, while Gallo's 4.50 PAG in 126 games equated to a 1.06 APPA. This was the proof I needed.
I wouldn't call it being legitimized. But yet, I can no longer take away from the fact that writing partner Brown and I have created new metrics (that sit side-by-side with fundamental stats) that help us make conscience choices for fans and readers of UTRMinors.
Have Brown and I gotten lucky with some of our past prospect predictions? Of course. But we are improving each year as we work on our intellectual inventions, and that's the direction we've been working toward from the very beginning.
No comments:
Post a Comment