Looking Back At Jordan Lyles' 2012 Season

One season. 167 strikeouts in 144 innings as a precocious 18-year old in Low A ball. That was enough to turn Jordan Lyles into a super prospect, the savior of the Astros farm system and a future ace in the making.

Things haven't really turned out that way, have they? Lyles has now pitched in two major league seasons with ERAs over 5 and strikeout rates below 7. He's 7-20 in 40 career starts and has amassed a huge workload before his 22nd birthday.

While the results haven't been favorable for Lyles, is there any evidence that he made progress last season? At such a young age, Lyles is still five years away from getting close to his peak years. All we really need is to see him making strides in some areas to feel better about his ability to bounce back in 2013? Will he become more like the pitcher we expected after that stellar season in Lexington, or is he destined to be the fourth or fifth starter that we've seen the past two years?

To answer that, we're going to have to go deeper than the numbers. Those numbers show us an ineffective pitcher, with an ERA over 5.00 in the past two years. His SIERA, however, is quite a bit better than that. In fact, Lyles could be seen as massively unlucky in 2012, posting a SIERA 1.33 lower than his ERA.

So, what are the knocks on Lyles? What would he need to improve upon to become a front line starter? Let's start with his velocity.

Velocity

In his first season in the big leagues, Jordan Lyles posted an average fastball velocity below 90 MPH. That's a bad sign for any pitcher hoping to have big strikeout numbers. For example, David Price and Justin Verlander (two names you NL-centric fans may not be familiar with yet...) both had some of the highest average fastballs out there, with Price topping all qualified pitchers at 95.5 MPH, according to FanGraphs and Verlander clocking in at 94.3 MPH.

Last year, Lyles and the Astros talked about having him work on different things with his delivery that will add to his velocity. He went down to the minors after working on it in spring training, but we had no idea whether it would pay off soon or not.

Turns out, it did. Whether by physical maturity or new mechanics, Jordan Lyles saw his average fastball rise from 89.8 MPH in 2011 to 91.8 in 2012. That two MPH jump is pretty significant, and puts him firmly in the middle of the pack in the majors for fastball velocity, which also is a much better sign for him being able to turn the corner and become a league average or better pitcher.

If you're watching for signs that Lyles could take another step forward in 2013, start here. Another increase in fastball velocity, even up to 93 MPH on average could mean better results.

Pitch Selection

Obviously, though, Lyles doesn't just throw fastballs. Did anything change in the kinds of pitches he was throwing?

While Lyles still featured the same five pitch repertoire, he did change how he implemented them. Basically, Lyles junked his changeup in favor of the two-seam fastball. In 2011, he threw the change 13 percent of the time. That wasn't a lot, but it did make it his third-most used pitch.

His two-seam fastball was his least-used fastball, behind his four-seamer and his cutter and he barely featured the two-seamer at all in 2011. That changed in 2012, when his change usage fell to about 5 percent and his two-seamer usage jumped to almost 20 percent.

As you'd expect, Lyles accomplished something very interesting for his increase in sink: he raised his ground ball percentage from 41 percent to 53 percent. If his increased usage of the two-seamer is legitimate and sustainable, then Lyles may be able to also sustain a higher ground ball rate. Couple that with a strikeout rate around 7.5 (which is attainable from Lyles' 6.30) and Lyles could be a very effective pitcher very soon.

Changes By Month

What about Lyles through the season? Did he wear down? Why were there times he hit bumpy spots and did he make progress from the beginning of the season to the end?

Turns out, he did. His two best months were June and September, as he had ERAs of 3.72 and 3.58 in them, respectively. His worst month was in July, when he had a 7.22 ERA. But, the smart TCBer by now realizes that ERA is a bad judge of whether a pitcher was effective or not. After all, Lyles didn't have Jed Lowrie behind him in August and Matt Dominguez had yet to spread his defensive light to the team at third base.

June, though, seems to be a complete aberration. It's the only month where Lyles had a K rate less than 6 and he had his highest walk rate in June too. The fact was, Lyles was just plain luckier in June than any other month. He had a much lower line drive rate in June and a much lower BABiP than in other months.

In September, though? We see Lyles post very good strikeout rates, somewhat lower BABiP than average, but higher than in June, and a sharp decline in his home run rate. With that, he also posted a season-high in ground ball rate at nearly 57 percent.

Oh, and did I mention that his velocity numbers didn't dip by the end of the season, either? His September and October velocity numbers, thanks to Texas Leaguers, are right at 92 MPH for both his two-seam and four-seam fastballs.

What happened to the cutter?

Why did Lyles shutter the cutter? Did he really stop using the pitch, or was it a trick of Pitch F/X's wild imagination? Well, judging from the Pitch F/X data from Brooks Baseball on his last start in September (a pretty good complete game shutout against Milwaukee), the pitch is well and truly gone.

Lyles featured two fastballs this year with similar profiles. His four-seam fastball has a vertical drop of 7 and horizontal movement of -4.5. That means it stays in on the hands of right-handers with quite a bit of sink for a four-seamer (most live around 9-10). The two-seamer has more horizontal movement towards right-handers and more vertical sink, down to about 5.

The cutter he used in 2011 profiled differently. It had similar vertical drop to the four-seamer (slightly less, actually, which puts it in a dangerous spot of "flattening out"), but much more run to the left-hand side of the plate than either of the other two.

So, if we're looking for the pitch in data around here, it should have the movement horizontally of Lyles' slider with the drop of his fastball. Oh, and the spin is important. His cutter spun at an angle of 186 last season, while his four-seamer was at 210 and his two-seamer was at 235. This year, those two held similar, with his four-seamer spinning at 213 and his two-seamer spinning at 236.

In that September start, only two pitches spun around that 200 angle and they were both his four-seamer. They both had tons of movement in on the hands of right-handers but with a drop that resembled his two-seamer more than anything.

So, it appears that the cutter is well and truly gone, at least for that game.

Conclusions

What does all this mean? His fastball velocity has improved, and his reliance on his two-seamer over his cutter is a good sign, affecting his ground ball rate in a positive way. He didn't see a corresponding drop in his home run rate, but we can expect that to go down if he continues to throw fastballs with good sink.

He also held up well to the end of the season and seems to have hit on a formula for success in the future. I'd like to see some more growth from him in his overall velocity and more use of his slider (which was pretty good last year), but overall, I'm a lot more encouraged by Lyles' progress than I was.

Even if he doesn't improve in the offseason, Lyles' 2013 campaign could be improved by just having a few statistical glitches go his way again (a 10 percent infield hit rate, for one or a rise in his left on base percentage for another). That's enough for me to view him as a "breakout" candidate for next season.