Documenting the strikeouts of Carlos Martinez

Roughly a year and a half ago, I constructed an exploratory piece titled "Adam Wainwright and the art of the set-up pitch" with the primary intention of explaining the importance of pitch sequencing and location in strikeouts. My intention can be summed up using the following quote directly from that July 2014 article:

"The final pitch of a strikeout is very important and is usually the pitch seen on highlight reels, but what about the set-up pitch? Isn't the set-up pitch just as important as the pitch that actually got the strikeout?"

Looking back, I feel like I did an okay job at fulfilling my intention, but I only documented 76 strikeouts for that article and had nothing yet available to serve as a comparator to my findings. Given that I have not written about Carlos Martinez in a while and we still do not have live baseball to discuss, I figured it was the perfect time to revisit the exercise. Thus, I documented each final two-pitch sequence of 157 Martinez strikeouts from the 2015 season, more than doubling the sample size I documented for Wainwright.

Now, if you look at Martinez's FanGraphs player card, you will notice that he recorded a total of 184 strikeouts last season. I did not just arbitrarily pick the number 157 or skip out on the remaining 27 strikeouts out of laziness. Instead, for whatever reason, Martinez's August 27th start against the Diamondbacks, in which he tallied six strikeouts, is not available for viewing on BrooksBaseball.net, and I excluded all 21 strikeouts of opposing pitchers since I truly believe pitchers approach fellow pitchers differently.

2015 PitchF/x Basics (via BrooksBaseball.net)

Remember: Regarding horizontal movement in right-handed pitchers, a negative value means arm-side movement, whereas a positive value means glove-side movement.

Pitch	Frequency	Velocity (MPH)	Dragless Horizontal Movement (inches)
Fourseamer (FB)	31.48%	96.53	-6.58
Sinker (SI)	25.63%	95.72	-12.99
Changeup (CH)	17.18%	87.69	-13.61
Slider (SL)	25.67%	85.92	9.87

In my four-part offseason series titled "Repertoires in review," Martinez garnered the most votes of the pitching staff in two of the four pitches analyzed: sinker (72% of 760 total votes) and changeup (30% of 466 total votes). Plus, no one will reasonably deny the fact that Martinez has phenomenal stuff, subsequently making it tough for him to necessarily go wrong from a sequencing standpoint considering each offering is at least above average.

Regardless, Martinez has four different pitches, meaning there is a total of 16 unique two-pitch sequences (SL-SL, FB-SL, FB-CH, etc.) at his disposal when trying to retire hitters via strikeout. Fifteen of the 16 possible combinations were represented in my data collection from the 2015 season, but, due to size constraints, I included only the top 11 two-pitch sequences in the chart below. For those curious, those sequences that did not make the chart were SL-SI (3), CH-FB (3), FB-SI (2), and SI-FB (2).

Final Two-Pitch Strikeout Sequences

Full Web Version

As you can see, Martinez was not at all afraid to throw the same pitch back-to-back in order to notch a strikeout. In fact, the slider-slider combination was his top final two-pitch strikeout sequence, appearing 28 times (or ~18% of his total strikeouts). Changeup-changeup, sinker-sinker, and fastball-fastball all made the chart as well, with 13, six, and six, respectively.

Remembering back to Wainwright, the 34-year-old ace used the same pitch back-to-back on only 11 occasions (or 14% of the documented strikeouts). Personally, I think the fact that Martinez was able to find success going back-to-back 53 times (34% of his strikeouts) should be seen as a representation of the quality and complexity of his repertoire. While Wainwright is still viewed as one of the best pitchers in baseball, his stuff is not as good as it once was (particularly his 90-91 MPH fastballs), and because of this, he cannot often throw two fourseamers or two sinkers in a row without being timed up by opposing hitters.

Percentages of Each Pitch Appearing Among Final Two-Pitch Sequences

Fourseamer	Sinker	Changeup	Slider
32% (51 out of 157)	29% (45 out of 157)	41% (65 out of 157)	64% (100 out of 157)

Once again, the slider leads the way, being used in 64% of Martinez's strikeouts in 2015. What is notable is that the slider is a pitch Martinez uses a little over 25% of the time, so it is clear that the pitch is an integral weapon for putting hitters away, as compared to the fourseamer appearing in only 32% of strikeouts despite being his most frequented pitch at a ~31% usage rate overall.

Velocity Implications of Final Two-Pitch Strikeout Sequences

Full Web Version

First, I must provide a few notes on the classifications used in the above chart. I considered pitches within two MPH of each other as possessing the "same" velocity. In other words, an 89 MPH changeup was classified as having the "same" velocity as an 87 MPH slider or a 91 MPH changeup. Martinez's pitches group into two very similar velocity chunks—95-100 MPH for fourseamers/sinkers and 85-91 MPH for changeups/sliders—so this led to a very different separation than what was seen when analyzing Wainwright (who has five different pitches with four noticeably different velocity ranges: 90-92 for fourseamers/sinkers, 82-84 for changeups, 75-77 MPH for curveballs, and 85-87 MPH for cutters).

The chart makes it clear that Martinez enjoys a lot of success by setting up his offspeed pitches with his fastballs, as 69 of his 157 (~44%) strikeouts had a final two-pitch sequence of going "faster to slower." Given Martinez's propensity to go back-to-back with pitches (discussed above), it is not surprising to also see that 66 of his strikeouts (~42%) came with two-pitch sequences clocking in at the "same" velocity. Honestly, the mere threat of blowing a hitter away with a 100 MPH fastball immediately following an offspeed pitch likely plays an integral role in these two velocity sequences being so effective.

Location, Location, Location

Here, I divided the pitching zone into four equal quadrants, and if interested, you can find an admittedly blurry scan of my manual pitch plotting here. If the set-up pitch fell into the upper left quadrant and the put-away pitch fell into the lower right quadrant, then, intuitively, I determined that the two pitches had different locations.

Remember, just as I stated in my Wainwright article, while pitches may be in the same quadrant, there is still the possibility that they look very different to the hitter. However, dividing the zone into more (and subsequently smaller) quadrants for more specificity is not feasibly done by hand. Plus, I think having bigger quadrants helps us to better appreciate the magnitude of difference in pitch location, should we find a trend. After all, there must be a noticeable difference in order to be successful in "changing the eye level" of the hitter.

Full Web Version

While ~67% (105 of 157) seems like a lot, we only really have one comparator for Martinez, and that is Wainwright, who, using this same exercise, placed his set-up pitch in a different quadrant than his put-away pitch an eye-popping 77% of the time. That being said, one must remember that, with Wainwright, I used a much smaller sample size (66 strikeouts) and excluded nine strikeouts that ended curveball-curveball, which almost certainly played to Wainwright's benefit in this measure.

Two GIF Examples (courtesy of the ever talented @daniel_doelling)

Sinker-Changeup Sequence to Paul Goldschmidt on May 25th (BrooksBaseball)

The main reason behind the inclusion of this GIF is to provide a visualization of just how difficult it must be for hitters to distinguish a difference between Martinez's changeup and fastball(s). Now, I was the one who picked the two pitches for the creation of this layered GIF, so I know which is which, but I feel confident in saying that many readers will be unable to confidently state which pitch is which. We have the privilege of watching the GIF on loop, too, just imagine how difficult it must be for hitters standing ~55 feet away from where the pitch is released.

Fourseamer-Slider Sequence to Howie Kendrick on May 31st (BrooksBaseball)

On this one, you will probably have to first train your eyes before being able to fully appreciate El Gallo's downright lethal approach. It is tough to ignore the rising fourseamer for the first few loops, and that's fine because it is a very well-executed set-up pitch by Martinez. However, do your best to then focus on the nasty slider that put Kendrick away for the strikeout. This two-pitch sequence involving two very different locations, combined with two very different velocities (97 MPH fastball followed by an 85 MPH slider), is virtually unhittable.