clock menu more-arrow no yes mobile

Filed under:

First and worst: Sliders of 2013

Continuing our series of best and worst pitches of the season, we now look at sliders with the help of linear weights and PITCHf/x data.

Jared Wickerham

Continuing our miniseries on the best and worst pitches of 2013, let's now shift our focus to sliders (SL). Like we did previously with four seam fastballs, we will use FanGraphs data on slider pitch type linear weights (wSL/C) as well as velocity (vSL), vertical movement (SL-X), and horizontal movement (SL-Z) data to create a z-score for each in order to compare the variables to one another. A summary score -- the Sum Z Score -- will then be calculated to give a final, single value for measuring the value of a given pitcher's slider. The ten percent criteria will also be used, with only pitchers who used the slider at least ten percent of the time will be considered.

Just looking at linear weights, these are the best and worst sliders of 2013, broken down by starters (SP) and relievers (RP):

Name wSL/C
SP, First James Shields 4.83
SP, Worst Cliff Lee -7.23
RP, First Jamey Wright 7.54
RP, Worst Luke Hochevar -8.73

Right off the bat, we see old friends from the four seamer article Cliff Lee and Jamey Wright; however, in a twist, we find the venerable Lee in the 'worst' column, something that comes as a mild surprise.

However, we aren't done just yet. Applying our ten percent criteria and then calculating the necessary z-scores for the wSL/C data and including pitch movement velocity data, we get the following (arguably) more accurate best and worst sliders of 2013; I have also included pitch frequency data (Pct Use):

Name Sum Z Score Pct Use
SP, First Yu Darvish 4.37 37.4
SP, Worst Joe Saunders -4.05 12.6
RP, First Craig Kimbrel 6.28 28.2
RP, Worst T.J. McFarland -5.11 20.2

Unlike the four seamer data, our initial suspects don't survive the cut for best slider. We also see the best sliders go to two pitchers -- Yu Darvish and Craig Kimbrel -- who are known for their power offerings and in essence, pass the eyeball test. Our worst slider honorees -- Joe Saunders and T.J. McFarland -- are both lefty 'thumbers' who reply upon locating pitches and changing speeds to get hitters out more so than an above average pitch in terms of velocity or movement.

For those curious, here is the z-score breakdown of our four hurlers; keep in mind that since we standardized the data, the mean/average for all variables is zero:

Name vSL_z SL-X_z SL-Z_z wSL/C_z
Darvish -0.88 2.70 1.31 1.24
Saunders -0.59 -1.08 -1.34 -1.05
Kimbrel 0.73 1.21 3.01 1.33
McFarland -1.26 -0.78 -0.78 -2.29

Right away, we see some subtle differences in each pitcher's slider, even between Darvish and Kimbrel. Looking at velocity (vSL_z), we find that only Kimbrel's score is above average, with Darvish's slider velocity being slightly sub-par, compared to other starter's sliders. What Darvish lacks in above average velocity, he makes up in movement in the vertical plane, bettering Kimbrel's slider in the SL-X_z category twofold. Kimbrel has an advantage not only in velocity, but also in horizontal movement and tilt, coming in with a z-score of 3.01. Overall, Kimbrel's slider can be considered double trouble, not only having dominant velocity on the pitch, but also elite horizontal tilt to it, making it all the more difficult to hit.

Talking about hits, let's briefly take a look at how hard each pitch is to put bat to:

Name K% BAA
Darvish 46.2% 0.141
Saunders 42.1% 0.262
Kimbrel 63.9% 0.099
McFarland 20.3% 0.424

Again, we see evidence Kimbrel's slider being a dominant, plus-plus, 80-grade offering. We also see the disparity not only between the best and worst sliders, but also how 'deserving' McFarland is of worst slider in 2013 (sorry, T.J.!). Poor swing and miss/strikeout numbers added to a gaudy batting average against (BAA) all add up to not only the wSL/C value we saw, but also his overall z-score.

Considering pitch selection and sequencing data (which can be found over at Brooks Baseball) for a moment, we come to another distinction. For McFarland, he uses his slider most often against lefties and in the first pitch of an at bat, while Kimbrel uses his ably against both lefties and righties, but mostly in pitcher friendly counts. For the starters, Darvish is similar to Kimbrel in that he uses it against both lefties and righties and in pitcher friendly counts, particularly two-strike counts. Saunders, not surprisingly, uses his slider much like McFarland, using it almost exclusively against lefties and sparingly against righties, but deviating from McFarland by using it in any count and not just pitcher friendly situations.

Let's return to the discussion of velocity one last time, while also thinking about how each pitcher uses the slider, whether it's a true off speed/change of speed offering, or if it's another power pitch, just with a wrinkle. To quote The Neyer/James Guide to Pitchers:

A slider is 6 to 8 mph slower than a fastball, and it breaks horizontally (or laterally) and vertically...designed to look like at fastball as it approaches the plate.

So do our quartet follow this rule, or are their sliders something else altogether? By comparing the difference between the average velocities of their fastballs and sliders, we can get a better grasp of what style of slider each of the four uses:

Name Diff FA-SL Velo, MPH SL Min-Max, MPH
Darvish 11.8 72.9-88.2
Saunders 7.8 77.4-87.3
Kimbrel 11.1 82.1-89.5
McFarland 8.6 76.0-81.2

I also included the minimum and maximum slider velocities of the the quartet, to see if there is variation in how hard they throw the pitch, perhaps lending some evidence as to whether the slider is a power pitch or a change of speed for each of the four.

Overall, the best sliders of our four pitchers have a pronounced average difference in velocity as compared to the fastball -- each right around 11 miles an hour slower than the fastball -- while the worst sliders are a more traditional slider in terms of velocity difference off of their heaters. That being said, we do see some interesting distinctions in terms of min-max speeds; both relievers throw it with a tighter velocity envelope, while both starters appear to be more inclined to take off a good amount of velocity on the pitch on occasion, with Darvish having an impressive 15 MPH difference between his hardest and slowest sliders in 2013. Anecdotal evidence shows that when applying a correlation test to all pitcher data on the variables 'Diff FA-SLVelo' and 'Sum Z Score' to see if pitchers with a larger difference between their average slider and fastball velocities had 'better' sliders, no statistically significant correlation was found.

While it does have some detractors due to the stresses it can induce on the elbow in order to throw it effectively, we find that the slider is an effective weapon for both starters and relievers and displays a surprising amount of flexibility and variability in how its proponents use the pitch.


All data courtesy of FanGraphs and Brooks Baseball.

Stuart Wallace is a writer at Beyond The Box Score. You can follow him on Twitter at @TClippardsSpecs.

More from Beyond the Box Score: