In 2010, Brandon Morrow produced what most baseball fans would view as mediocre results, with an ERA of 4.49. Yet heading into 2011, the sabermetrics community touted Morrow as a potential ace, pointing to impressive strikeout-to-walk ratios and bad luck on batting average on balls in play. Morrow’s predicted breakout year did not come close to fruition, as his ERA continued to rise, resting at 4.72 in over 179 innings. His strikeout-to-walk rates were just as impressive and his batted ball luck appeared neutral, but this time a failure to strand runners and work out of jams had torpedoed his results (He stranded 65.5% of runners- 71% is about average).
Heading into this year, many still view Morrow as a breakout candidate, but the narrative has changed a little. No one has soured on his ability to rack up strikeouts, but there seems to be a lurking suspicion that Morrow is ineffective with runners on base (as a result of his low 2011 strand rate). Poor mechanics out of the stretch and inability to work under pressure are commonly cited reasons for this issue. The difference between Morrow’s results with the bases empty and men on base seem to support this idea:
Morrow allowed a much worse batting average, on base percentage, and slugging percentage with men on base, suggesting that he is indeed worse out of the stretch.
Don’t be fooled by these results. Many different factors can affect outcomes for a pitcher- some are within the pitcher’s control and some are outside of the pitcher’s control. A pitcher has relatively more control over his strikeout, walk, and home run rates- these underlying skills are pretty good reflections of how well a pitcher is actually pitching. On the other hand, pitchers exert relatively little control over their batting average on balls in play (BABIP), the amount of balls in play that they allow to fall in for hits. Quality of defense, ballpark dimensions, weather, plain old luck, and other factors can skew a pitcher’s batting average on balls in play. So if Brandon Morrow really is fundamentally worse out of the stretch, we would expect his strikeout, walk, and home run rates to be worse with men on base than with the bases empty.
We can see above, though, that his strikeout, walk, and home run rates are all about the same whether or not there are runners on base. This tells us that he exhibits a constant set of underlying skills regardless of situation.
The reason for his low strand rate is not differences in these core skills, but the difference in BABIP seen in the table above. Morrow allows hits on balls in play a whopping 34.2% of the time with men on and a relatively low 27.3% of the time with the bases empty. It is, of course, worse for a pitcher to allow hits on balls in play with men on, because more runs can be scored on a hit if there are already men on base.
So, to put it in simpler terms, last year, Morrow had equal amounts of “bad luck” and “good luck”, but most of his “bad luck” happened to occur when men were on base. If Morrow’s luck were to even out, and he were to allow hits on balls in play both 30% of the time with men on and 30% of the time with no men on, he would strand a lot more baserunners and his ERA would plummet.
Since his BABIP is largely outside Morrow’s control and driven by randomness, it is reasonable to expect this to happen. We can see from the splits of his strikeout, walk, and home run rates that he basically profiles as the same pitcher whether or not he is pitching from the stretch. In both scenarios he is racking up K’s, and limiting walks and homeruns. If his luck evens out on batted balls with runners on and he maintains this strong K/BB/HR profile, Morrow should finally make good on the breakout many have been expecting for the last two years.