As the 2013 minor league season kicked off, I posted 2 FanPosts here that evaluated how the Astros organization’s minor league starters performed in 2012 versus their league and level peers on 4 fielding- and field-independent statistics (Righties link, Lefties and Righties link). Since then I have performed similar analyses for all organizations' minor league starting pitcher prospects who received a straight B grade or better from SBNation minorleagueball’s John Sickels pre-2013.
Five Astros' starters received a B or higher grade from Sickels: Jarred Cosart, Nick Tropeano, Lance McCullers, Mike Foltynewicz, and Asher Wojciechowski. They and the other organizations' B or better prospects who faced at least 200 batters in 2012 at or above the High Rookie level form the study population. Since McCullers fell short of qualifying based on his 2012 batters faced total, his 2012 Appalachian League stats were combined with his 2013 Midwest League stats to date and for the sake of what follows I will treat that pooled sample of data as if it were compiled during 2012 in the same leagues.
The four stats examined are as before:
- BB&HBP %: the sum of walks and hit by pitches, divided by plate appearances
- K %: strikeouts divided by plate appearances
- LD&OFFB %: the sum of line drives and outfield fly balls, divided by the sum of nonbunted balls in play and home runs
- OFLD&OFFB Pull %: the number of line drives and fly balls that went to the batter's pull-field third of the outfield, divided by the total number of outfield line drives and outfield fly balls
The stats of each prospect are then compared to those of his peers, with "peers" defined as all starting pitchers of the same throwing dexterity who faced a sufficient sample of batters at the same level of the minors. Five scores are generated based on those statistical comparisons:
- BB&HBP Score
- K Score
- LD&OFFB Score
- OFLD&OFFB Pull Score
- Overall Score (20% BB&HBP Score, 36% K Score, 22% LD&OFFB Score, 22% OFLD&OFFB Pull Score)
Each score is assigned using a 20-to-80-style scouting scale, with a 50 score indicating level-average performance, any value above 50 indicating better than level-average performance, and each 10 points amounting to one standard deviation (Overall Score requires scaling to get 1 standard deviation back to 10 points).
Since each of the Astros’ fivesome throws righthanded, only the results of righties will be shown in the graphs.
McCullers’ BB&HBP % was 16% in the Appalachian League part of his 2012 (67 batters) but has dropped to 11% (43 score) in his 2013 campaign (157 batters). Cosart would check in with a relatively unchanged 40 score for 2013 as of May 20. More generally, there aren't many "control artists" among the 47 starters in the graph.
Cosart, Foltynewicz, and Wojciechowski have each had upticks in their K % in 2013, and if they can sustain those increases we should see them climbing north on this graph over the course of the year. One can infer just how much weight strikeouts carry in subjective rankings of pitching prospects - only 5 of 47 pitchers on this graph rated below level-average at K %.
This graph reinforces how elite Cosart is at avoiding outfield fly balls and line drives; the Rays’ Guerrieri (pitched in short-season A ball during 2012) is the only highly-rated prospect that was within sniffing distance of where Cosart sat in AA and AAA. McCullers' score is quite good also- I would suspect that he is placing well above where many readers would have expected.
OFLD&OFFB Pull Score
It is reassuring to see McCullers faring well on this stat at the start of his pro career, as it provides some indirect evidence that he is beating hitters with his fastball and that his offspeed pitches are seldom pulled aerially. As was true for the general sample of Astros' minor league starters that I examined in the cited posts, in this group of top-shelf prospects I am finding that this relatively ignored metric explains slightly more of pitchers' performances on batted balls than do their linedrive, flyball, groundball, and popup rates per batted ball (or combinations of the same).
McCullers, Cosart, and Tropeano each graded out more than one standard deviation better than an average minor league starter in terms of performance on these metrics versus peers. Better than one standard deviation is quite good as it conveys that the trio outperformed 87% of 2012's minor league starters. Foltynewicz’ youth affords him plenty of development time for him to climb upward.
Bonus: Pitcher Similarity Scores
For prospectphiles, I’ve developed an algorithm that rates how similar pitchers are in terms of the 4 individual component scores described above.
Below are the 3 most similar Sickels B grade or higher righthanded prospects to the would-be Astros and their percent match (100% would indicate identical from a statistical perspective).
Tropeano: 96% M. Wisler (SD), 91% T. Thornburg (MIL), 91% Wojciechowski
Foltynewicz: 94% B. Maurer (SEA), 91% D. Corcino (CIN), 90% J. Taillon (PIT)
Wojciechowski: 91% B. Maurer (SEA), 91% Tropeano, 91% M. Wisler (SD)
McCullers: 90% A. Sanchez (TOR), 89% W. Peralta (MIL), 89% A. Webster (BOS)
Cosart: 85% A. Webster (BOS), 80% T. Guerrieri (TBR), 80% C. Martinez (STL)
Tropeano and Wisler, at 96%, are the 3rd highest match of the 1081 possible pairs of the 47 pitchers featured on the graphs. Tropeano and Wojciechowski rated similarly versus peers in 2012 with K Score the main differentiator. Cosart’s uniqueness at LD&OFFB % confounds identification of a close match - his best has only 85% strength.