Rich Harden by ground

The scope of most Rich Harden articles usually ends up including his gaudy whiff rates. As this Cubs fan bids farewell to the impressive, yet fragile, Mr. Harden, it seems like a good time to talk about the complete pitcher. Or, at least, another part of the pitcher—his ability to get outs on the ground.

Land Ranger

Harden’s ground ball rates, taken from Fangraphs, match up with the Gameday data from 2007-2009. Other than this list back to 2003, I’ll be referring to Gameday data.

In 2008, the year Harden was traded from Oakland to Chicago, he was no longer a slightly above average ground ball pitcher (43 percent is average in MLB from 2007-2009). Turns out, he was a fly ball pitcher, of the extreme variety. Things regressed in 2009, but not all the way.

Rich Harden ground ball rates

2003  .46
2004  .45
2005  .42
2006  .43
2007  .39
2008  .30 
2009  .38  

Lacking pre-2007 PITCHf/x, there are two explanations I’m left to consider. First, Harden stopped throwing his splitter in his later days in Oakland. Second, Harden was not throwing a two-seam fastball much, if at all, in the past few seasons.

In a late-season interview with WGN Radio, Harden told Judd Sirott he was planning on bringing back his two-seamer as a means of reducing contact quality. Arming himself with something to pitch to contact with, Harden also expressed hope the two-seamer would help reduce his pitch counts.

With Gameday’s rich data, I can break down the outcomes of those grounders in more detail. Given the lack of playing time, and PITCHf/x, for Harden in 2007, I’ll stick his two most recent seasons. First up, slugging rate on ground balls.

Rich Harden total bases per ground ball

2008  .253
2009  .318

Harden’s increased ground ball rate in 2009 came with an increase in slugging on grounders. Let’s put that in terms of run values instead. Without regard to base state or number of outs, but with regard to the count, Harden fared far better on grounders in 2008 than in 2009.

Rich Harden ground ball rv100 (E)

2008  -8.7 (-7.3)
2009  -4.4 (-8.0)

The run values are based on hits and outs, as recorded by Gameday. The rv100E reflects the expected or MLB average values for ground balls on the same counts from 2007-2009. In 2009, the expected runs were slightly less (more negative, more runs saved), which would reflect a few more grounders yielded on hitter’s counts.

In any case, the actual outcomes hint at some bad luck in 2009 and a little bit of good luck (but not much) in 2008.

The rate state is telling, but so would be the total run values. This blends/confounds innings pitched and ground ball rates, but we’re okay on the former (148 and 141 IP) and actually interested in the impact of the latter.

Rich Harden ground ball rvaa (E)

2008  -8.3 (-6.9)
2009  -5.8 (-10.5)

This is more interesting, to me at least. Ground balls cost Harden 2.5 more runs in 2009 than in 2008. Using the “expected” values, he should have saved himself 3.6 runs. Even using the actual 2008 and the expected 2009, he should have seen an improvement in total run value saved.

How much of that is dumb luck, and how much is courtesy of the Cubs’ defense? On that note, will things be any better for Harden with the Rangers’ infield behind him in 2010?

Once again, we turn to Fangraphs and Ultimate Zone Rating per 150 games (UZR/150)—a measure of runs saved against average—for a comparison of the 2009 Cubs and Rangers, along with the 2008 Cubs.

UZR/150

     CHC   TEX   08CHC
1B   4.2   -0.2    4.6
2B   2.5    8.0    2.5
SS   5.6   14.8   -3.3
3B  -6.2   -9.5    1.6
     6.1   13.1    5.4

Not what I would’ve guessed. The Cubs were actually a little better in 2009 than 2008. But they have nothing on the strong-up-the-middle Rangers. Still, even with a slightly better infield and a higher ground ball rate, the worms burned Harden more often.

Omar Vizquel has moved on, but most of those shortstop innings went to Elvis Andrus. Harden will notice a difference with Andrus/Ian Kinsler over Ryan Theriot/Various Guys.

How much of a difference? Using Baseball Projection’s TotalZone projections and my own guesstimates at playing time, a seven-run gap emerges (nearly the same as the UZR numbers above). Apportion one of those runs to Harden for 2010. Where the chips may actually fall, or where the balls may actually roll, remains to be seen.

Diving back into Gameday, we can run a few comparisons on 2009 ground ball outcomes.

You can read Harden’s line at least three ways, and they’re not mutually exclusive, either. His ground balls allowed were hit harder than average. He was unlucky on ground balls. The Cubs defense took his days off, counting on strikeouts to win the day. No matter what, you’d expect things to get at least a little better for him in 2010, and that’s without considering the potential impact of the two-seam fastball.

A brief thought on the two-seam’s impact. My pitch classifications are not exhaustive, and I may be more likely to tag two-seamers as such for known sinkerballers, so there may be some selection bias….but my calculations show four-seam fastballs tend to yield a ground ball rate around 35 percent and two-seam’s yield a 53 percent rate. That’s topped only by splitters, at 56 percent. The best ground ball pitches are both pitches Harden used to throw regularly.

Wait and see

I’d take a little bit of optimism away from looking at these numbers. If Harden continues to regain his ground ball tendencies of old, he will have a better defense behind him to turn them into outs. And his luck could turn, too. Each little piece should save a tenth of a run here-and-there from his ERA. Harden of the 4.10 in the NL could easily be Harden of the 3.80 in the AL—more two-seamers, better defense, better luck is not too much to ask for, is it? And I haven’t started talking about fly balls and home runs yet.

References & Resources
PITCHf/x data from MLBAM and Sportvision.
Batted ball classifications from MLBAM.
Pitch classifications by the author.