The Diamondbacks should probably stop batting Tim Locastro leadoff

Tim Locastro is undoubtedly one of the most exciting players in all of baseball.

For one, he is arguably the fastest player in the majors. His 30.8 ft/sec sprint speed ranked as the highest in MLB (minimum 10 opportunities). Locastro attempted to steal 17 bases with the Diamondbacks in 2019. He was successful in all 17 of those attempts. The only other rookies this century to steal at least 15 bases without being caught once in a single-season are Quintin Berry (21 SB in 2012) and Craig Gentry (17 SB in 2011).

The other element that makes Locastro such an intriguing player is his knack for getting hit by the pitch.

Lowest single-season plate appearances totals for players w/ 20 or more HBP

• 250 Tim Locastro (2019)
• 306 Derek Dietrich (2019)
• 345 Brandon Guyer (2016)
• 369 Chief Roseman (1890)
• 370 Tyler Flowers (2017)
• 385 Brandon Guyer (2015)
• 408 Carlos Gomez (2018)
• 412 Derek Dietrich (2016)

22 out Tim Locastro’s 250 plate appearances ended in HBPs. That’s an 8.8 HBP%, which is higher than what the league average BB% was last season (8.5).

For more great insight into Tim Locastro’s game, I’d recommend reading this article (by Steven Pappas, one of our writers) and watching this video (by Foolish Baseball).

What I am going to explore in this article are Tim Locastro’s splits (batting order).

When Locastro batted leadoff in 2019, he reached base in 27.8% of his plate appearances (25/90).

When Locastro did not hit first in the order, he reached base in 40.0% of his plate appearances (64/160).

Here is a pie-chart showing his percentage of total plate appearances by spot in the lineup…

Locastro’s -12.2% (-.122) difference in on-base percentage for batting first in the order versus not batting first cannot be explained by HBP%, as Locastro was hit by a pitch in 7.8% of his PAs in the No. 1 position.

The difference can be largely attributed to Locastro’s extremely low walk rate when batting leadoff. In his 90 plate appearances batting first in the order, Locastro walked only once (1.1%). In all other plate appearances, he walked 8.1% of the time.

Contrary to what I would’ve expected, Locastro’s swing% wasn’t much higher when batting leadoff (46.4% vs 43.8% when batting elsewhere). With that being said, Locastro did see a higher percentage pitches in the zone (46.9% vs 40.9%). It’s possible that he made contact on a higher percentage of his swings and that a higher percentage of his contact was him putting the ball into play. His approach may have been entirely different.

To see if the difference in Locastro’s OBP when batting first in the order versus any other spot in the order is statistically significant, we will do a two-proportion Z-test (more AP Stats).

The null hypothesis is that there is no difference between Locastro’s OBP batting leadoff and his OBP in other spots in the lineup (OBP1 – OBPNOT1 = 0).

The alternative hypothesis is that OBP1 – OBPNOT1 < 0. In other words, his OBP in the No. 1 spot is lower. We have evidence to suspect that the alternative hypotheis might be true: the test statistic (.278 – .4 = -.122; -.122 < 0).

The frequency in which Locastro hit leadoff was fairly constant (all of his games there didn’t occur in one span).

For the purpose of this piece, we will assume the splits are independent of each other (which is not perfectly realistic).

Here is my work to find the Z-stat…

We can then calculate the probability of getting a Z-stat of 1.93 (standard deviations) or higher using a Z-curve where “x” represents the Z-stat for OBP1- OBPNOT1, with the mean equal to zero and the standard deviation equal to one.

The probability of getting a negative difference of .122 or smaller in terms of OBP between the 90 PA subset (OBP1) and the 160 PA subset (OBPNOT1) if the null hypothesis is true (OBP1 – OBPNOT1 = 0) is about 2.7%.

Ultimately, it appears as though Locastro was significantly less likely to reach base when batting leadoff. There was a sub-3% chance of him posting an OBP that much lower when batting No. 1 versus elsewhere in the order strictly by chance.

While we can’t definitively say that Locastro will be worse out of the leadoff spot in the future, we have reason to think that might be the case considering the statistically significantly difference between OBP1 – OBPNOT1 in 2019 (at a sig. level of alpha = 0.05 [5%]).

It is likely in the Arizona Diamondbacks best interest to refrain from batting Tim Locastro in the leadoff spot. Whether they will make that decision remains to be seen.

Note: Technically, the Z-stat should be negative and the curve should be shaded on the left side. This does not affect the p-value.