What has gotten into Taijuan Walker?

You might have noticed that Taijuan Walker was not very impressive over the first two months of the season. In his first nine outings, Walker allowed four runs or more five times. He walked four batters in nearly half of his starts, allowed eight homers in 43 frames, and only made it past the sixth inning once. His ERA was well over seven and hitters torched him for a .313/.393/.503 line. For sakes of comparison, Joey Votto is batting .282/.396/.495 in a hitters park. Walker looked overmatched and if the Mariners had had any confidence in their minor league starters he almost certainly would have been optioned to Tacoma by mid-May.

You may have also noticed that Walker has looked like an entirely different pitcher lately. In his last seven starts, his ERA is under two. He's struck out more than a batter per inning in that time and he's walked just three, none in his last four appearances. Opponents are hitting .199/.221/.318 in that stretch, and while that isn't quite as bad as Willie Bloomquist's line, it's a pretty good Rob Johnson impression. There are a number of factors behind Walker's development, and while it's difficult to isolate the specific importance of each individual component, the improvement in his pitch mix, ability to throw quality strikes, and raw arsenal suggests that his recent run of form is no fluke.

Pitch Mix
Walker has four pitches: a fastball that can reach 97, a slider that often moves more like a cutter, a tumbling changeup, and a curveball. The fastball has always been legit -- at least when he spots it well -- but he's struggled with his secondaries, and early in the season he had a tough time getting same-handed hitters to swing and miss at his slider or curve. He's always leaned heavily on his fastball, but he's increased his usage of the pitch as the season has gone along. In fact, when you compare his pitch mix from his first nine outings with his last seven, he's almost entirely abandoned his curve and slider. Take a look at the change in the tables below (and keep in mind that his slider is classified as a cutter because of it's shape; Walker calls the pitch a slider so for clarity, that's how I refer to it too. The same applies for his changeup):

(Thanks as always to Brooks Baseball for the help on pitch data.)

It would be easy to stop here. Walker stopped throwing his slider and curve and turned the corner on the strength of a fastball-changeup combination that has been too much for opposing hitters. Narrative crafted and done. '

The problem is that it's not quite that easy. While Walker has undoubtedly had success by predominantly throwing his fastball and change, his use of that approach pre-dates his recent dominance. In Baltimore on May 19th, he threw only five breaking balls and that was one of his worst outings all year. In his following start in Toronto, he threw three curves and eight sliders -- fewer breaking balls than average at that point -- but that wasn't a good game either. Per Fangraphs pitch values, his slider and curve haven't been successful this season and it certainly helps that he isn't throwing them as often but there's more behind his improvement than a change in strategy.

Throwing strikes
In those first nine starts, Walker threw strikes on 61.8% of his pitches, bottoming out with two sub-60% strike outings in Baltimore and Toronto. He flipped a switch in his following start, getting 76 strikes in 102 pitches against Cleveland, the highest strike% he's posted all year. From that day forward, Walker has hit the zone or induced a swing with just a bit less than 70% of his pitches. It's pretty easy to limit walks when you're only missing the zone with three out of ten pitches.

There are a couple of layers at work here. On the one hand, there's the obvious conclusion that more strikes leads to fewer walks, and fewer walks almost inevitably means fewer runs. Additionally, more strikes means more pitchers counts where Walker can nibble a little bit and entice a hitter to chase an elevated fastball or a dropping changeup. It's never a bad thing to throw strikes -- provided that a pitcher isn't just grooving fastballs on 2-0 and 3-0 counts for the sake of avoiding walks -- but it's even more important for Walker than other pitchers.

All year long, Walker has been poor when working with runners on base. Hitters are hitting about .80 points better per wOBA with men on than when the bases are clear and seven of the thirteen homers the right-hander has allowed has come with runners aboard. Interestingly, his walk and strike out ratios are significantly lower when he's pitching out of the stretch, which suggests that he's throwing a lot of bad and hittable strikes with runners on, almost as if he's afraid to compound a stressful situation by walking more hitters. It's normal for pitchers to have more success with nobody on, but the degree to which Walker struggles with men on base is unusual. The reduced walk rate he's posted lately thus has a second positive effect by allowing him to work out of the windup a little more often.

Throwing better strikes

Walker's first few games in 2015 were marred by poor breaking pitches, predictable sequencing, and an almost chronic inability to hit the lower part of the strike zone. Chris Young gave us a firsthand look at how to have success working the top of the zone, but Walker doesn't have Young's command, plane, or unpredictable plan of attack. Early on in the season, hitters could bank on getting a hittable fastball up in the zone within the first few pitches of their at bat, as you can see in the charts of three of his earliest starts:

Too many pitches in the middle of the plate, and too many elevated fastballs early in counts. He also liked to use his fastball early in counts and rely on his secondaries to chase strikeouts, a predictable pattern that didn't make life any easier for him.

Compare that to his most recent start in San Diego:

He still likes to work up in the zone, he still isn't the world's greatest sequencer, but he threw a lot more pitches near the corners and there's a nice barren spot in the middle of the plate. You can see something similar in the plot in his start last week against the Angels, where there's again quite a few pitches on the edges and an almost unblemished oval middle in to righties:

In short, Walker has done a much better job of not only throwing strikes, but throwing quality strikes. He's getting ahead in the count, and for the most part, is living out of the middle of the plate. He hasn't been perfect, of course, but there's tangible improvement and it's good to see him work the corners and not settle for throwing over the heart of the plate all of the time.

A better changeup
We saw above that Walker has slowly changed his pitch-mix over the course of the season to include more changeups. Encouragingly, he's not just using his changeup, he's employing it really effectively as a putaway offering. In his first nine starts, hitters swung and missed at the change about 11.5% of the time they saw it. In his seven starts since, his whiff rate has more than doubled on the pitch, which is particularly impressive given how often he's using it. The change also features about a half inch more drop on average now than it did earlier in the season. The changeup's predominantly vertical movement leaves it less prone to wide platoon splits than a typical change, meaning that even though he doesn't use his slider or his curve much, he still has an out pitch to attack righties with. Matt Kemp found out the hard way yesterday:

In the game against San Diego, Walker threw 14 changeups for 13 strikes, 12 swings, and a career high seven whiffs. Exciting stuff to say the least.

A better slider
It's always nice when a pitch f/x system starts classifying your slider as a slider, at least occasionally. In all serious though, Walker is starting to get more vertical drop out of the pitch, particularly when he throws it low in the zone or down in the dirt. It may or may not turn into a reliable offering for him going forward, but he's flashing an average slider right now, and if he can throw that consistently, he'll have another weapon to generate whiffs with.

The road ahead
It's amazing how quickly Walker has gone from being one of the worst starters in the league to a promising young contributor. For all of the improvements he's made though, the work isn't over. Development is never static and while the 22-year-old has had the upper hand lately, the league is bound to adjust to his adjustments before too long. Going forward, we'd like to see him sustain the gains he's made with his changeup and ability to work the corners while also incorporating an improving slider and more creative sequencing into his game plan. It's a tall order, and we can expect bumps in the road along the way.

But in the last few weeks we've gotten a glimpse of what a fully mature Taijuan Walker can do. A look at his last seven starts suggests that the improvement he's made over the course of the season is real, and that the Los Angeles native will be able to stick in the Mariners rotation going forward. If we're lucky, we're only seeing the start of his development at the major league level. Walker has the raw talent to be an all-star in this league, and if the past five weeks are any indication, he also the capacity to improve his game and get the most out of what his right arm will allow.