This serves as the wrap-up to my offseason series of FanPosts that analyze how the starting-biased pitchers of each organizational minor league affiliate performed versus their league's (or leagues') starting pitchers in 2013, as determined using a fielding- and ballpark-independent statistical evaluation system. The prior team-specific FanPosts provide more detailed player evaluations (spray-chart-derived event frequencies and batter handedness splits data, plus commentary) and can be accessed via these links: Oklahoma City, Corpus Christi, Lancaster, Quad Cities, Tri-City, and Greeneville. Here I will mostly just present scores for each pitcher's complete 2013 minor league season.
What the Scores Mean
"Performance Score" amounts to the pitcher's overall grade relative to other league starters. The first step is to break the plate appearances against each league starting pitcher down into 12 possible outcome categories [walk or hit-by-pitch, strikeout, infield flyball, groundball (GB) to batter's pull-field third of the diamond, GB to center-field third, GB to opposite-field third, line drive (LD) to pull-field third, LD to center-field third, LD to opposite-field third, outfield flyball (OFFB) to pull-field, OFFB to center-field, OFFB to opposite-field]. Then, for each such event they allowed, the pitcher is charged with the average number of expected runs that the corresponding event was worth per that league's season (the strikeout or IFFB is typically the most beneficial event for the pitcher, whereas the pull-third OFFB or pull-third LD is the most calamitous). At that point, the pitchers in a league can be ranked from best to worst in terms of expected run avoidance and their individual Performance Score is the number that expresses how well they fared overall relative to their peers.
To pinpoint each pitcher's relative strengths and weaknesses (and better understand how they achieved their Performance Score), 3 subscores are computed to rate the pitcher's performance versus that of league peers looking only at control (Control Subscore), strikeouts (Strikeout Subscore), or batted balls (Batted Ball Subscore). An Age Score is also tabulated to quantify how young the hurler was relative to the league's starting pitcher sample. A 50 score on any measure amounts to matching the league's average value (the pitcher topped or equaled 50% of peers), a 60 amounts to 1 standard deviation (SD) better than league-average (topped/equaled 83% of peers), a 70 amounts to 2 SD better than league-average (topped/equaled 97% of peers), a 40 amounts to 1 SD worse than league-average (topped/equaled 17% of peers), and so on with any 10 point swing in a score amounting to 1 SD. For players who pitched for multiple affiliates, the scores calculated for each stop are averaged out based on the number of batters faced in each league to arrive at a single value for the year.
The 59 pitchers whose names appear in the tables that follow faced at least 100 batters in the 2013 minors (above the transitional Gulf Coast or Arizona Leagues) and averaged at least 10 batters faced per game, and thus qualify as "starting pitchers" for the sake of the analysis. Table values that bettered the league average mark by at least 1 SD are highlighted in green text ("very good") whereas those that trailed it by at least 1 SD are in red ("very bad"). Asterisks indicate lefthanded throwers. A red line has been placed through the names of those who were no longer in the organization as of the time of this post.
The first table below presents the top 30 of the 59 starters evaluated, sorted by Performance Score. King Feliz lays claim to the organizational Performance Score crown for 2013. Try as I might, and having seen Feliz pitch live twice in 2013, I cannot identify a flaw or nit to pick at present. That, along with how he stacks up versus reference standards that would be appropriate for assessing his present development and performance, makes it rather easy for me to envision him pitching alongside McCullers and Velasquez early in 2014 and performing comparably to them, if not better.
The table numbers leave me craving some Peacock in the major league rotation, and soon (as do faint memories of him pitching respectably after his August recall). Oberholtzer sure grades out promisingly and with some movement back towards these sorts of numbers in the 2014 majors he may not look quite so much the overachiever that many have pegged him as. This minor league data also shows the potential for Cosart to be better at the strikeout in 2014 than he was in his 2013 big league debut.
Below are the scores of the remaining 29 pitchers. Data that Hader and Smith accumulated in 2013 prior to becoming Astros farmhands was processed recently and the results have been assimilated into the table. Though the duo rated similarly overall with a 52 Performance Score, one can appreciate that batted balls were Hader's relative strength and Smith's relative weakness. Above all else, the pair excelled at being young relative to their league and level peers, as noted by the greenness of their Age Scores.
More than a handful of starters who would be expected to see at least some full-season action in 2014 failed to accumulate enough 2013 data to merit tabulation of scores; that group includes Rudy Owens, Alex White, Kyle Weiland, Kent Emanuel, Luis Ordosgoitti, and Joe Musgrove. For the rest, the values in these tables should serve as useful frames of reference against which to track their progress during and after the 2014 campaign, as measured via the same method.
The hearty souls who have been following along with this series may appreciate that I have yet to present any data on the organization's minor league relievers. The next post will closely resemble this one, but with those relievers as the study cohort.