Key Losses and the Next Men Up: Ohio State Buckeyes Preview
August 18, 2015Are you ready for the Ezekiel Elliott E:60? (You will be.)
August 18, 2015We tend to view fortune in sports in binary terms. This player has been lucky. That player has been unlucky. However, is it possible for a player to both be fortunate and unfortunate at the same time?
Such has seemingly been the case with Danny Salazar’s post All-Star break starts, as his sterling 1.76 ERA, seventh best in MLB, stands in contrast to his peripheral numbers and FIP1 , while the Indians’ 3-3 record in those Salazar starts, with his incredibly shrinking ERA, lead him far from needing to re-wear the same socks in consecutive outings (or whichever his favorite superstition might be).
On Tuesday, Salazar continued his post-break ERA excellence by holding the Boston Red Sox to just four hits, one walk, and one run in seven innings during the Cleveland Indians’ 8-2 victory over the Red Sox.
Regardless, the most interesting part of Salazar’s pitching line is to determine if there is something concrete with which he has built his newly dominant ERA during this stretch or if a return to the norm should be expected over the course of the remainder of the season.
Note: Please note that the numbers from this point forward only take into account Danny Salazar’s first five post-break starts and do not include his start against the Red Sox on Tuesday. However, the same pattern that Salazar has had in the second half continued in Boston as Salazar gave up fewer runs than his peripheral numbers suggest he should have.
The Numbers2
The standard set of statistics show that Salazar has been allowing less than a hit every two innings, and he is allowing less than a baserunner per inning, even when adding in walks and batters hit by a pitch3 . However, that alone does not explain why his ERA has been cut in half4 .
Salazar’s peripheral numbers show a completely different story. He is striking out fewer batters, he is walking more, and he is allowing about the same amount of home runs. In fact, his FIP is up a half of a run at the same time that his ERA dropped almost two full runs.
The explanation for the difference lies in the BABIP5 and LOB%6 . Both of these statistics are considered to have a large component of luck involved with them as BABIP tends to normalize around .300 for most pitchers, while LOB% is a component of “when” a pitcher gives up hits rather than if he gives them up7 .
However, that is a surface view of those numbers. Certain pitchers have been able to induce weak contact in a few different ways including a deceptive release, developing a new pitch, altering pitch-type ratios, et cetera.
A-ha! Now, we get to see the first positive indication of what Danny Salazar is doing to batters. While his ground ball rate has remained relatively static, he has converted nearly four percent of his previously hit balls from line drives to fly balls. Adding in that we know his home run rate has also remained relatively static, the simple deduction is that Salazar has turned those line drives into lazy fly balls. Definitely a positive step, but a four percent change does not explain his BABIP nearly dropping 50 percent either.
The contact velocity on batted balls against Salazar also reveals some intriguing information, though it’s difficult to draw conclusions from them.
As learned from a previous Bode Plots on the Indians hitters, converting batted balls from soft to non-soft is supposed to correlate to BABIP more than converting batted balls from non-hard to hard. Salazar has had six percent less soft contact in his second half starts, while he also has had four and a half percent less hard contact.
The expectation would be that the reduction in soft contact (largely easy outs) would overpower the reduction in hard contact (largely hits) and BABIP would rise here. Yet, we see the opposite occur. Part of that oddity can be explained in having less line drives, but certainly not all of it.
Pitch Location8
The next obvious place to look for explanations into Salazar’s increased effectiveness is the placement of his pitches, as we saw when we noticed that Trevor Bauer was hitting the outside corners of the plate effectively with some of his early 2015 success.
There are certainly some differences in the pitch location to right-handed batters in the second half9 . The heat map is from the catcher’s POV, so seeing the numbers all slightly rise to the right side means that Salazar’s pitches seem to be gravitating away from the hitters. The greatest decrease is the second column, which is the inside portion of the plate. Avoiding the inside of the plate might help explain why less batters are able to get line drives off of Salazar.
Again, there are some differences in the pitch location to left-handed batters in the second half. Salazar has somewhat reduced the number of inside pitches as well, but to a much smaller degree. The most striking contrast to left-handed batters seems to be that he is throwing a far higher percentage of his outside pitches for strikes.
Overall, the heat maps demonstrate reasons for Salazar’s batted ball differences, but certainly are not drastic enough changes to indicate why he his overall effectiveness has increased while his peripheral numbers have decreased.
Pitch Types10
Salazar has been tinkering with his pitch variation all season. He began the season very much relying on his four-seam fastball before mixing in his off-speed variations along with a two-seam fastball that shows up as a sinker at much higher rates in May and June (including a brief affair with his slider in June). However, it appears that part of his recent success has been due to a return to his four-seam fastball.
One huge difference between his April reliance on the fourseam fastball and his recent stretch is that he is mixing in his sinker at a much higher rate, while reducing the number of curve balls thrown.
Well, the first noticeable pitch on the chart shows why Salazar has gone away from his slider. It apparently was not fooling many hitters. His split-fingered fastball appears to be his most effective pitch though the lower usage might have something to do with it11 .
The other obvious component of the batting average on his pitches is that hitters are having a much, much more difficult time with his bread-and-butter fourseam fastball.
Conclusions
It is simple to determine what is happening with the multitude of data that is now available for public consumption. Danny Salazar has seen his ERA drastically drop in the second half of the 2015 season despite peripheral numbers that indicate an expected rise in earned runs given up. At the same time, Salazar is peppering the corners of the plate more often, while also going back to a reliance on his four-seam fastball. In addition, to left-handed batters at least, he is throwing more strikes. Hitters are hitting less line drives, but they are also hitting the ball softly less often.
The more difficult portion is figuring out why Salazar is having the paradoxical success12 . It is possible that Salazar has increased his deception with his delivery, the sequence of pitch types has been better, or that he is a huge beneficiary of the vastly improved Indians defense in July and August.
There are enough positive changes to discern that Salazar’s success is not merely luck. Hitting the outside portion of the plate and inducing less line drives are both inherently good things. However, there are enough worrisome traits that tend to indicate that there will be a correction coming for Salazar. It will be intriguing to see sharp that correction winds up being.
Of course, if the Indians are not winning when he starts, then all of his supposedly good fortune or increased skill is going to waste anyway.
- Fielding Independent Pitching, which was 3.91 in those same starts [↩]
- Courtesy baseball-reference.com and fangraphs.com [↩]
- In other words, his WHIP is below 1.00. [↩]
- And, his WHIP pre-break, while not quite as good, was not grossly out of line with what he has done in these last five starts. [↩]
- Batting average on balls in play [↩]
- Percentage of baserunners left on base [↩]
- If hits are given up in clusters, then it will be low. If hits are spread out, then it will be high. The effect is called cluster luck. [↩]
- Courtesy brooksbaseball.net [↩]
- First half table is on the left and second half table is on the right if you cannot read the font on your device. [↩]
- Courtesy brooksbaseball.net [↩]
- If Salazar can make his split-finger fastball at first appear to like his four-seam, then batters will start their swing and need to adjust when it is likely too late. [↩]
- ERA dropping, but FIP rising and still losing games [↩]