clock menu more-arrow no yes

Filed under:

Is Lucas Duda objectively streaky?

New, 65 comments

Time to examine the years-long debate: is Lucas Duda a streaky hitter?

Robert Deutsch-USA TODAY Sports

There has been a longstanding debate among Mets fans over whether Lucas Duda is a streaky hitter, or at least a streakier hitter than most. But what does "streaky" even mean? Is it a player whose hot and cold streaks occur with greater frequency? Or is it a player whose hot and cold streaks are more extreme (i.e., they last longer)? There isn't really an accepted answer to this question, so it's difficult to declare whether or not Duda is objectively streaky if there is no objective way to measure it.

I have a method in mind on how to answer our question. However, I'm always open to criticism or debate, so please share your thoughts in the comments.

Lucas Duda's wRC+ last season was 133. That will be our baseline for his true 2015 talent level at the plate. His monthly wRC+ marks were:

March/April: 161
May: 164
June: 64
July: 107
August: 169
September/October: 158

I'm using monthly splits because they're conveniently available and because you have to divide the season in one way or another. Monthly splits may mask short streaks, and they're also a bit fuzzy for streaks that cross month boundaries. If we used much smaller samples the data won't tell us anything because all players will have wild fluctuations among small samples. If we used samples larger than a month we wouldn't have much data to work with.

Duda's 2015 wRC+ had a standard deviation of 42.67. But that doesn't mean anything to you or me; we need to compare it to other hitters. Here are the standard deviations of each of the Mets' hitters last season who had at least four months' worth of data to analyze. (Note: I used 50 plate appearances as the minimum cutoff for a month to count as a sample. Also, Cespedes's numbers include those with the Tigers.)

Name Months in Sample Size 2015 wRC+ Standard Deviation
Lucas Duda 6 133 42.67
Curtis Granderson 6 132 24.95
Yoenis Cespedes 6 135 24.08
Daniel Murphy 5 110 36.87
Wilmer Flores 6 95 23.00
Juan Lagares 5 80 29.94
Michael Cuddyer 4 97 57.31
Ruben Tejada 4 95 14.04

What does this tell us? Well, Duda is indeed streaky, at least compared to the average Mets hitter last season. Unfortunately, I don't have data to tell you the standard deviation of an average MLB hitter's wRC+ by month, otherwise I would provide that as well.

In fact, Duda had the second highest variance on the team, behind only Michael Cuddyer. Cuddyer is a story of his own; his September/October missed my cutoff by eight plate appearances, but he posted a ridiculously awful 23 wRC+ over that time, which would have skewed his number even more. Duda's standard deviation was considerably larger than that of every other Met save Cuddyer, though. For reference, Duda's standard deviation in 2014 was 35.61, so he was streakier in 2015 than in 2014, but he was a pretty streaky hitter in 2014 as well.

On the flip side, Ruben Tejada was remarkably consistent all season, almost eerily so. Wilmer Flores, Cespedes, and Curtis Granderson were also very consistent in 2015. Whether a player being extremely streaky is a good or bad thing is actually a separate debate, and one that I'm not trying to answer here.

Still, though, there was nothing cherrypicked, selectively endpointed, or pre-molded in the data regarding wRC+ standard deviation by month. At least in one measurement, we have an objective answer to the longstanding debate—yes, Lucas Duda is in fact a streaky hitter.