clock menu more-arrow no yes mobile

Filed under:

Dallas Keuchel in perspective

This year, the Houston southpaw has accrued a massive amount of ground balls, while also putting up strikeouts and limiting walks. But it's the extent to which he's done these things that has set him apart.

Keuchel has blazed out of the gate in many statistical categories.
Keuchel has blazed out of the gate in many statistical categories.
Bob Levey

What is important for a pitcher? Traditionalists would answer wins. Other, marginally more advanced bipeds would cite the prevention of runs as a crucial task. But the truly erudite among us recognize that there are three things for which a pitcher ought to strive: ground balls, strikeouts, and walks (or rather, the lack thereof). The best pitchers will maximize their outputs of the former two while minimizing their output of the latter; poorer hurlers will lag behind in one or more of these areas.

Let's shift our focus for a minute. Although the Houston Astros are the worst team in the world for the fourth straight year, they possess a few players who have performed admirably. One of said players is Dallas Keuchel, whose 2.67 xFIP ranks fifth in the majors. He's now the most improved pitcher in baseball, and he's done well enough that we must become cognizant of him. But what's caused his evolution, and from what has he evolved?

In his 2012 debut, Keuchel was objectively terrible. While hitters put the ball on the ground 52.1% of the time against him, they also walked 10.3% of the time while striking out 10.1% of the time. Last year, the walks decreased to 7.6%, while the strikeouts and grounders increased to 18.0% and 55.8%, respectively. This year, though, he's been on another planet: Giving free passes to 5.0% of batters while fanning 22.7% and amassing a 67.7% (!) ground-ball rate, Keuchel and his all-encompassing hegemony have taken the league by storm.

There are some underlying causes here, to be sure (in 2013, he phased out the curveball in favor of a whiff-inducing slider) but right now, I'd like to discuss results — specifically, the three aforementioned results. As you might suspect, that 67.7% worm-burner rate leads the majors this year, as well as for every year, ever*. While not historically dominant, the strikeouts and walks are also impressive: The former ranks 29th, and the latter ranks 18th. When viewed collectively, however, these numbers take on a whole new significance.

*Well, since 2002, but, y'know.

As the above footnote mentions, easily available ground ball data goes back to 2002. Henceforth until now, there have been 1,148 qualifying seasons. Using z-scores, we can make their statistics era-neutral, to compensate for the meteoric rise of strikeouts in recent years. Through this lens, Keuchel's GB, K, and BB rates are reduced to simple numbers: 3.08, 0.58, and -0.92, respectively. But while they may be bland, each of these is also notably better than league average. How common is that?

There are a few ways we can look at this; let's begin with the balanced approach. Of those 1,148 starters, 335 have strikeout z-scores of 0.50 or greater, 375 have walk z-scores of -0.50 or lower, and 321 have ground ball z-scores of 0.50 or greater. However, far fewer player have done all three things simultaneously — 22, to be exact:

Year Name K% z_K BB% z_BB GB% z_GB
2014 Felix Hernandez 23.6% 0.78 4.7% -1.05 50.0% 0.52
2014 Dallas Keuchel 22.7% 0.58 5.0% -0.92 67.7% 3.08
2013 Matt Harvey 27.7% 1.86 4.5% -1.37 47.7% 0.53
2013 Felix Hernandez 26.3% 1.51 5.6% -0.75 51.4% 1.24
2013 Adam Wainwright 22.9% 0.66 3.7% -1.82 49.1% 0.80
2012 Felix Hernandez 23.8% 1.14 6.0% -0.58 48.9% 0.54
2012 James Shields 23.6% 1.09 6.1% -0.53 52.3% 1.10
2011 Cole Hamels 22.8% 1.02 5.2% -1.07 52.3% 1.12
2011 Roy Halladay 23.6% 1.21 3.8% -1.91 50.9% 0.89
2010 Roy Halladay 22.1% 0.89 3.0% -2.72 51.2% 0.89
2010 Adam Wainwright 23.4% 1.21 6.2% -0.80 51.6% 0.95
2009 Roy Halladay 21.6% 0.66 3.6% -1.93 50.2% 1.01
2008 Roy Halladay 20.9% 0.84 4.0% -1.60 53.7% 1.53
2006 Chris Carpenter 20.5% 0.95 4.8% -1.21 53.3% 1.18
2006 Brandon Webb 18.7% 0.52 5.3% -0.96 66.3% 3.02
2005 Andy Pettitte 19.5% 0.77 4.7% -1.09 50.2% 0.85
2005 Roy Oswalt 18.4% 0.51 4.8% -1.03 49.1% 0.68
2005 Chris Carpenter 22.4% 1.47 5.4% -0.72 54.5% 1.50
2004 Chris Carpenter 20.4% 0.79 5.1% -1.14 52.2% 1.04
2003 Roy Halladay 19.1% 0.62 3.0% -2.12 58.4% 2.07
2003 Andy Pettitte 20.1% 0.84 5.6% -0.85 52.2% 1.18
2002 Roy Oswalt 21.8% 1.10 6.5% -0.56 47.1% 0.55

So this isn't all that uncommon; Keuchel's not even the only pitcher this year to do it, as the King has continued carrying out his royal duties. Nevertheless, this methodology might sell Keuchel short; while his strikeout and walk rates are solid, his ground ball rate has been otherworldly. Although that 67.7% figure isn't actually the best ever (it's sixth), it's still one of only seven that's three standard deviations above average:

Season Name K% z_K BB% z_BB GB% z_GB
2002 Derek Lowe 14.9% -0.49 5.6% -1.03 66.8% 3.68
2008 Brandon Webb 19.4% 0.47 6.9% -0.27 64.2% 3.20
2003 Derek Lowe 12.4% -0.88 8.1% 0.37 65.9% 3.15
2006 Derek Lowe 13.5% -0.73 6.0% -0.60 67.0% 3.12
2005 Brandon Webb 18.2% 0.46 6.3% -0.25 65.0% 3.09
2014 Dallas Keuchel 22.7% 0.58 5.0% -0.92 67.7% 3.08
2006 Brandon Webb 18.7% 0.52 5.3% -0.96 66.3% 3.02

For the most part, the other pitchers on that list either didn't punch out batters or handed out bases on balls too frequently. Aside from Keuchel's 2014, only Brandon Webb's 2006 season featured strikeout and walk rates a half of a standard deviation better than average.

Even if we broaden the scope to, say, two standard deviations, there still aren't many pitchers for whom the across-the-board-0.5 platitude applies:

Season Name K% z_K BB% z_BB GB% z_GB
2002 Derek Lowe 14.9% -0.49 5.6% -1.03 66.8% 3.68
2008 Brandon Webb 19.4% 0.47 6.9% -0.27 64.2% 3.20
2003 Derek Lowe 12.4% -0.88 8.1% 0.37 65.9% 3.15
2006 Derek Lowe 13.5% -0.73 6.0% -0.60 67.0% 3.12
2005 Brandon Webb 18.2% 0.46 6.3% -0.25 65.0% 3.09
2014 Dallas Keuchel 22.7% 0.58 5.0% -0.92 67.7% 3.08
2006 Brandon Webb 18.7% 0.52 5.3% -0.96 66.3% 3.02
2004 Brandon Webb 17.6% 0.21 12.8% 2.35 64.3% 2.87
2003 Brandon Webb 22.9% 1.47 9.1% 0.85 63.9% 2.87
2010 Tim Hudson 15.1% -0.86 8.0% 0.28 64.1% 2.85
2005 Derek Lowe 15.6% -0.17 5.9% -0.46 63.1% 2.80
2009 Joel Pineiro 12.1% -1.41 3.1% -2.18 60.5% 2.77
2007 Derek Lowe 17.7% 0.07 7.1% -0.16 65.0% 2.77
2007 Roberto Hernandez 15.6% -0.38 6.9% -0.27 64.3% 2.67
2003 Kevin Brown 21.6% 1.18 6.5% -0.41 62.5% 2.66
2004 Jake Westbrook 13.0% -0.74 6.8% -0.37 62.6% 2.62
2005 Jake Westbrook 13.3% -0.72 6.3% -0.25 61.7% 2.59
2008 Derek Lowe 17.3% -0.04 5.3% -1.01 60.3% 2.58
2004 Derek Lowe 12.5% -0.85 8.5% 0.40 62.3% 2.57
2012 Trevor Cahill 18.6% -0.20 8.8% 0.85 61.2% 2.56
2006 Chien-Ming Wang 8.4% -1.95 5.8% -0.70 62.8% 2.52
2002 Roy Halladay 16.9% -0.03 6.2% -0.71 59.5% 2.52
2013 Justin Masterson 24.3% 1.01 9.5% 1.45 58.0% 2.51
2005 Mark Mulder 12.8% -0.84 8.1% 0.69 60.5% 2.41
2007 Tim Hudson 14.3% -0.67 5.7% -0.91 62.0% 2.36
2007 Brandon Webb 19.9% 0.55 7.4% -0.01 61.8% 2.33
2011 Jake Westbrook 12.9% -1.36 9.0% 1.20 59.3% 2.26
2006 Jake Westbrook 12.1% -1.06 6.1% -0.55 60.8% 2.24
2013 A.J. Burnett 26.1% 1.46 8.4% 0.83 56.5% 2.23
2011 Derek Lowe 16.5% -0.50 8.4% 0.84 59.0% 2.21
2010 Justin Masterson 17.5% -0.26 9.1% 0.94 59.9% 2.21
2014 Tim Hudson 16.7% -0.75 1.8% -2.33 61.7% 2.21
2014 Justin Masterson 20.1% 0.00 10.6% 1.55 61.6% 2.19
2007 Felix Hernandez 20.4% 0.66 6.6% -0.43 60.8% 2.19
2011 Charlie Morton 14.3% -1.03 10.0% 1.80 58.5% 2.13
2005 Tim Hudson 14.1% -0.53 8.0% 0.64 58.6% 2.12
2005 A.J. Burnett 22.7% 1.54 9.1% 1.21 58.4% 2.09
2004 Tim Hudson 13.0% -0.74 5.6% -0.91 59.1% 2.09
2003 Roy Halladay 19.1% 0.62 3.0% -2.12 58.4% 2.07
2009 Derek Lowe 13.0% -1.21 7.4% -0.01 56.3% 2.05
2012 Jake Westbrook 14.1% -1.36 6.9% -0.12 58.1% 2.05
2010 Derek Lowe 16.5% -0.51 7.4% -0.08 58.8% 2.05
2002 Greg Maddux 14.4% -0.60 5.5% -1.08 56.4% 2.03
2003 Tim Hudson 16.8% 0.10 6.3% -0.51 57.9% 2.00

Of these 44, three — Keuchel, Webb, and Halladay's 2003 — had strikeout and walk z-scores of 0.50 or greater. It just seems to be that when a player dominates in one area, he invariably falls back in other areas. This applies to high-strikeout and low-walk pitchers as well:

Season Name K% z_K BB% z_BB GB% z_GB
2002 Randy Johnson 32.3% 3.51 6.9% -0.35 46.9% 0.52
2002 Curt Schilling 31.1% 3.24 3.2% -2.29 40.8% -0.45
2013 Yu Darvish 32.9% 3.15 9.5% 1.45 41.0% -0.76
2014 Jose Fernandez 34.2% 3.12 6.3% -0.34 48.8% 0.35
2002 Pedro Martinez 30.4% 3.08 5.1% -1.29 42.2% -0.23
2003 Kerry Wood 30.0% 3.05 11.3% 1.93 41.5% -0.36
2007 Erik Bedard 30.2% 2.79 7.8% 0.21 47.9% 0.41
2004 Johan Santana 30.1% 2.79 6.1% -0.68 40.6% -0.71
2004 Randy Johnson 30.1% 2.79 4.6% -1.37 43.5% -0.27
2003 Curt Schilling 28.8% 2.78 4.8% -1.24 41.5% -0.36
2008 Tim Lincecum 28.6% 2.72 9.1% 0.75 43.9% -0.02
2004 Oliver Perez 29.7% 2.71 10.1% 1.13 35.3% -1.51
2003 Mark Prior 28.4% 2.69 5.8% -0.76 41.0% -0.43
2012 Max Scherzer 29.4% 2.59 7.6% 0.24 36.5% -1.50
2005 Mark Prior 26.8% 2.53 8.4% 0.85 36.9% -1.17
2014 Max Scherzer 31.5% 2.52 8.6% 0.67 42.0% -0.64
2003 Pedro Martinez 27.5% 2.49 6.3% -0.51 41.2% -0.40
2005 Jake Peavy 26.6% 2.49 6.2% -0.30 44.0% -0.09
2004 Ben Sheets 28.2% 2.40 3.4% -1.91 43.3% -0.30
2005 Johan Santana 26.2% 2.39 5.0% -0.93 39.4% -0.79
2006 Johan Santana 26.5% 2.38 5.1% -1.06 40.6% -0.62
2011 Zack Greinke 28.1% 2.30 6.3% -0.41 47.3% 0.31
2004 Jason Schmidt 27.7% 2.30 8.5% 0.40 44.6% -0.11
2009 Tim Lincecum 28.8% 2.22 7.5% 0.05 47.5% 0.55
2014 Jon Lester 30.0% 2.19 6.6% -0.21 37.2% -1.33
2006 Jake Peavy 25.4% 2.12 7.3% 0.07 38.0% -0.99
2013 Max Scherzer 28.7% 2.11 6.7% -0.13 36.3% -1.67
2003 Javier Vazquez 25.7% 2.09 6.1% -0.61 35.7% -1.20
2011 Clayton Kershaw 27.2% 2.08 5.9% -0.65 43.2% -0.36
2007 Scott Kazmir 26.9% 2.07 10.0% 1.37 43.1% -0.26
2014 Yu Darvish 29.4% 2.06 7.2% 0.05 35.3% -1.60
2007 Johan Santana 26.8% 2.05 5.9% -0.80 38.0% -0.96
2007 Jake Peavy 26.7% 2.03 7.6% 0.10 44.0% -0.13
2005 Pedro Martinez 24.7% 2.03 5.6% -0.62 37.9% -1.02
2003 Jason Schmidt 25.4% 2.02 5.6% -0.85 37.0% -1.01

Season Name K% z_K BB% z_BB GB% z_GB
2010 Cliff Lee 22.0% 0.86 2.1% -3.26 41.9% -0.52
2005 Carlos Silva 9.5% -1.64 1.2% -2.92 49.2% 0.70
2010 Roy Halladay 22.1% 0.89 3.0% -2.72 51.2% 0.89
2003 David Wells 11.4% -1.10 2.3% -2.46 43.0% -0.14
2004 Jon Lieber 13.6% -0.62 2.4% -2.36 48.2% 0.44
2007 Greg Maddux 12.5% -1.06 3.0% -2.34 51.5% 0.90
2014 Tim Hudson 16.7% -0.75 1.8% -2.33 61.7% 2.21
2004 David Wells 12.6% -0.82 2.5% -2.32 48.0% 0.41
2002 Curt Schilling 31.1% 3.24 3.2% -2.29 40.8% -0.45
2011 Josh Tomlin 13.4% -1.24 3.2% -2.26 38.2% -1.18
2002 Rick Reed 15.6% -0.33 3.3% -2.24 39.4% -0.67
2014 David Price 26.6% 1.44 2.1% -2.20 43.2% -0.46
2009 Joel Pineiro 12.1% -1.41 3.1% -2.18 60.5% 2.77
2004 Brad Radke 15.9% -0.14 2.9% -2.14 42.8% -0.38
2005 David Wells 13.7% -0.63 2.7% -2.13 48.2% 0.54
2007 Paul Byrd 10.5% -1.49 3.4% -2.12 38.3% -0.92
2003 Roy Halladay 19.1% 0.62 3.0% -2.12 58.4% 2.07
2011 Dan Haren 20.2% 0.40 3.5% -2.08 42.5% -0.48
2005 Brad Radke 14.1% -0.53 2.8% -2.08 42.2% -0.37
2014 Bartolo Colon 17.3% -0.62 2.4% -2.07 35.9% -1.52
2010 Carl Pavano 12.9% -1.41 4.1% -2.06 51.2% 0.89
2011 Brandon McCarthy 17.8% -0.18 3.6% -2.02 46.7% 0.21
2003 Brad Radke 13.5% -0.63 3.2% -2.02 39.4% -0.66

None of the 2-z K guys had 0.5 Zs in both other areas, and only two 2-z BB seasons had it: Halladay in 2003 and 2010.

Again, universal superiority is hard to maintain, which shouldn't really come as a surprise — after all, if being exceptional wasn't difficult, then it wouldn't be exceptional. This fact only makes Keuchel's achievements thus far all the more superb; if he can sustain this start, and pitch to the level of 2006 Webb or 2003/2010 Halladay, perhaps the Cy Young (which went to those pitchers) will also be awarded to him. It's unlikely, but hey, if you're a 'Stros fan, you've gotta hope, right?

. . .

All data courtesy of FanGraphs, as of Thursday, May 22nd, 2014.

Ryan Romano is a featured contributor for Beyond the Box Score. He also writes about the Orioles on Birds Watcher and on Camden Chat that one time. Follow him on Twitter at @triple_r_ if you enjoy angry tweets about Maryland sports and live tweeting about Veep, Sundays at 10:30/9:30c on HBO. Boldly running for president. Proudly standing for everything.