Phil Hughes as postseason starter

When Phil Hughes faced off against the Minnesota Twins in the final game of their Division Series on October 9, many analysts questioned the approach of having Hughes pitch at home. Anyone who has followed Phil Hughes and the New York Yankees this season probably is aware of Hughes’ dreadful 2010 home/road splits. In 18 home starts, opposing hitters have batted .248/.312/.443, and Hughes has allowed 20 HRs in 106 innings at Yankee Stadium.

It’s also worth noting that Hughes did struggle against left-handed hitting (.235/.311/.417 with a K/BB ratio of 1.97, and 17 of his 25 HRs allowed). With the Twins having six of the nine hitters in the lineup batting from the left side, things didn’t look too promising.

Looking at PITCHf/x data for October 9, among his 99 pitches, Hughes threw 72.7 percent fastballs, 18.7 percent curveballs, 7.1 percent cut fastballs, and 2.0 percent changeups. According to Fangraphs, Hughes threw fewer cutters than usual on this night, since his overall season average came in at 63.6 percent fastballs, 16.4 percent cutters, 16.5 percent curveballs, and 3.5 percent changeups. Other than that, nothing too alarming in terms of pitch selection and velocity.

Yet, what was a bit alarming regarding Hughes was his pitch location. Against opposing left-handers, Hughes – for the most part – wisely stayed on the outer part of the strikezone.

This is nothing new considering his pitch location trends to left-handed hitters over the 2010 season:

On a few occasions, Hughes did get lucky concerning batted ball in play data against left-handers. On a few fastball counts where Hughes was behind, many pitches thrown to left-handers were fastballs over the middle part of the plate. Since most PITCHf/x databases don’t allow you to choose multiple counts for location purposes, descriptions of major at-bats will be listed below:

—At the top of the second inning with two outs and no one on, Jim Thome gets a 1-1 high fastball over the middle of the plate. Thome fouls it off.

—At the top of the fourth, Denard Span leads off and sees a 2-2 curveball dropped in on the lower middle part of the plate. Span turns on it and singles a line drive to right field.

—Later in the top of the fifth, Jason Kubel leads off, and on a 3-2 pitch Hughes places a fastball over the top middle of the plate. Kubel weakly flies out to left field.

—In the same inning with one out and Delmon Young on first, Jim Thome walks on seven pitches, two of which crossed over the middle of the plate (both fastballs), and both were fouled off.

—Finally, in the top of the sixth, Orlando Hudson is up with two outs and no one on. Hudson sees a 1-1 fastball over the middle of the plate and hits it to right for a single. Next batter is Joe Mauer; he gets a 1-1 fastball over the same spot, but fouls it off. Later in the same at-bat, Mauer is fed a 2-2 fastball middle-in and sends a line drive to right for a single.

Obviously, by the end of the fourth inning, Hughes had the luxury of pitching with a 5-0 lead. Most pitchers feel confident pounding the strike zone, especially with two outs and no one on base. But the pitch selection to these key left-handers could have been dangerous, and luck did play in some batted-ball situations.

Finally, what I found incredibly curious given Hughes’ 2010 success vs. right-handed hitters was that a majority of pitches thrown that night to righties were a series of fastballs and cut fastballs pretty much around the middle of the plate:

I was shocked by this and decided to look over Hughes’ pitch location data to all right-handed hitters over the entire 2010 season.

Moving forward to last Saturday’s ALCS Game Two match-up between the Rangers and Yankees in Arlington, lots of mention was made regarding Hughes’ strong start and his relative success in Arlington, as he hadn’t allowed a run in 15.1 total innings pitched.

Facing a larger number of righties (six out of nine) on the road and coming off a huge confidence-boosting win, things looked positive for Hughes on the mound. Triple-slash stat line of importance: in 2010 opposing right-handed hitters hit .253/.292/.381… not too shabby.

Despite Elvis Andrus’ single (2-2 fastball over the upper part of the plate) and subsequent steal of home off a delayed double steal, Hughes did go on to strike out three of the first five batters faced (called outside 0-2 fastball to Michael Young, swinging outside 2-2 fastball to Vladimir Guerrero, swinging outside 3-2 fastball to Nelson Cruz.

After that, things became steadily worse for Hughes, as he was knocked out of the game by the fifth inning with a line of 4 IP, 10 H, 7 ER, 3 BB, 3 K and 1 HR in 87 pitches.

Looking over the data for the day: 52 percent were fastballs, 28.6 percent cutters, 17.9 percent curves, 1.2 percent chang-ups. Facing a few more righties, Hughes expectedly threw a few more cutters, where 19 of the 24 thrown came with righties up to bat (however, it’s worth noting that the sole home run allowed was a 2-0 cutter to lefty David Murphy that was middle-in).

Looking over the pitch location to Ranger right-handers:

Except for a few more cutters and total pitches (obvious since he faced more righties than he did in Minnesota), this chart looks rather similar to the one created for the ALDS start on October 9.

Besides pitch location (or concurrent with it) Hughes also struggled with two strikes against opposing hitters. Against right-handers with Hughes ahead 0-2, three fastballs and one curve were thrown. One hundred percent of these pitches were swung at, and 75 percent were put in play.

Against Minnesota, Hughes was ahead of Twin hitters 0-2 a total of seven times. In this situation, Hughes threw five fastballs and two curves, generating a swing rate of 100 percent. Opposing hitters whiffed at a 40 percent rate and put the ball in play 20 percent of the time.

Below is a chart of Hughes’ pitch location against Minnesota hitters down 0-2:

Some bloggers have called Hughes’ approach to opposing hitters with two strikes a prevailing problem, but overall data have shown that Hughes is capable of getting opposing batters to swing and miss with two strikes. In fact, with two strikes opposing batters are only hitting .189/.266/.301 with 146 strikeouts in 417 plate appearances this season.

Putting this into context, 2010 strikeout leader Jered Weaver held opposing batters with two strikes to a line of .150/.204/.234 while striking out 233 in 535 plate appearances. Obviously, Hughes isn’t among the leaders in this category, but to say he is lacking in an ability to put away batters with two strikes isn’t exactly correct.

Comparing two starts against two separate teams and trying to foster some sort of conclusion probably isn’t fair. Besides looking at pitch locations and selections, other factors probably need to be addressed. However, it should be safe to draw the conclusion that Texas hitters were ready for Hughes last Saturday. He and the Yankees get another shot against the Rangers Friday night in ALCS Game Six, and it would be wise for Hughes to think twice about his pitch location, especially against this crew of right-handed hitters.

* All charts were provided by texasleaguers.com.