The Thing That Colby Lewis Does Better Than Anyone Else

Colby Lewis sits among the 20-worst pitchers by strikeout rate among qualified pitchers over the last two years. He has the second-worst ground-ball rate among that group. He has the fifth-worst fastball velocity. He basically only has two pitches, and only one of them rates as above-average on whiffs or grounders currently. He’s fifth on the Rangers’ starting-pitcher depth chart currently, and the team would probably admit that they are hoping that Chi-Chi Gonzalez and/or Nick Martinez take that job from him.

The point is, you wouldn’t think he was best in the league at anything.

The good news for our own personal senses of rankings and skills and value and Colby Lewis? That thing that he’s good at is something that people don’t think is really a skill.

Still. He’s been good at this thing, even as his ERAs over the last two years have been poor. And this thing? If you can repeat it, it’s good.

Defense-independent pitching tells us that pitchers have little control over balls in play, but we do know they can coax the ground ball. And if they can get ground balls by pitching low, couldn’t they also have some effect on whether those balls are pulled or pushed?

The results on pulled grounders against pushed grounders are starkly different. Opposite field ground balls are around 50% more productive than pulled grounders. It’s even starker for pulled fly balls, which are almost four times more productive than opposite field fly balls, when judged by slugging percentage.

So let’s say you could affect this much on a batted ball. By throwing low in the zone at certain times and high in the zone at other times, you could ostensibly get grounders on balls normally pulled, and fly balls on balls normally pushed. In this moment, your ground-ball to fly-ball ratio (GB/FB) to the push and pull fields become important.

It seems like it would be beneficial if you could have a high GB/FB to the pull field, and a low GB/FB to the push field. And so, the ratio between the ratios should be high if you put the pull field GB/FB on top. Looks what happens if you do that.

Starting Pitchers’ Pull and Oppo GB/FB
Name Pull GB/FB Oppo GB/FB Pull/Oppo GB/FB
Colby Lewis 2.84 0.25 14.7
Jake Odorizzi 2.96 0.32 12.2
Jered Weaver 2.14 0.25 12.1
Jesse Chavez 6.44 0.48 11.0
Marco Estrada 2.30 0.30 11.0
Cliff Lee 10.72 0.71 10.7
Mike Minor 3.46 0.35 10.6
Phil Hughes 2.96 0.34 10.6
Travis Wood 2.80 0.34 10.4
Dan Haren 2.97 0.36 9.8
Jason Vargas 3.83 0.40 9.6
Drew Hutchison 4.27 0.50 9.4
A.J. Burnett 13.00 0.70 9.0
Jeremy Guthrie 4.59 0.46 9.0
Aaron Harang 3.71 0.46 9.0
Hector Santiago 2.07 0.36 9.0
Jake Peavy 2.90 0.40 8.8
Jason Hammel 4.64 0.53 8.4
Trevor Bauer 2.88 0.38 8.4
Danny Duffy 2.62 0.35 8.3
Max Scherzer 2.81 0.41 8.3
Wei-Yin Chen 3.58 0.44 8.1
Nick Martinez 3.46 0.47 8.1
Bartolo Colon 5.00 0.58 8.0
Bud Norris 5.14 0.57 8.0
Pull/Oppo GB/FB is the pitcher’s pull gb/fb ratio divided by his opposite field gb/fb ratio.
Minimum 1000 balls in play since 2013. Overall n = 140

There’s Colby Lewis at the top. And a fair amount of interesting names, guys that have seemingly beat their peripherals the last three years, guys like Marco Estrada, Jered Weaver, Hector Santiago, Danny Duffy, and Nick Martinez. It’s tempting to call this a possible salve if you’re missing the strikeouts. This group, which represents the top 18% on this skill, has beaten its collective FIP by… six points since 2013.

A pitcher’s ground-ball to fly-ball ratio is highly sticky (.752 correlation year to year, which trails only strikeouts in strength). By cutting it up, of course, we are making it less sticky. And also, though the pitcher can elicit a ground ball reliably by throwing low in the zone, it’s a little bit less easy to show how they can get a pulled ball.

Here, for example, are two heat maps for the Pirates this year. In the one on the right, we see where pulled ground balls were thrown to righties. In the other, we see where opposite field ground balls were thrown. While pulled balls are generally thrown inside and opposite field balls are generally thrown outside, this isn’t a road map of any kind. Look how many pulled balls were on the outside part of the plate.


Looks like trying to get a pulled ground ball is a tricky thing. As Andrew Cashner told me once, “You can throw an outside corner fastball to a pull hitter and he’ll still pull it.” So living on the ability to get pulled grounders but not pulled fly balls, and adding to that the skill to get opposite field fly balls and not opposite field ground balls… that’s a tricky way to live.

Of course, if you’re Colby Lewis, living off an 88 mph fastball, a plus slider, and little else, maybe it is something to think about.

With a phone full of pictures of pitchers' fingers, strange beers, and his two toddler sons, Eno Sarris can be found at the ballpark or a brewery most days. Read him here, writing about the A's or Giants at The Athletic, or about beer at October. Follow him on Twitter @enosarris if you can handle the sandwiches and inanity.

Newest Most Voted
Inline Feedbacks
View all comments
8 years ago

What happens if you use a ratio of (Pulled GB + Oppo Fly)/(Pulled Fly + Oppo GB)? In other words, good outcomes vs. bad outcomes.

8 years ago
Reply to  tz

I was thinking something similar… What about (pulled GB + oppo FB) / (BIP + HR)… Basically something like “unproductive contact rate” (or non-ideal contact rate!)

I wonder if that would be a better predictor/estimator of a low BABIP than using Hard/Soft%.

8 years ago
Reply to  Eno Sarris

Sonny Gray! So THAT’S how he does it!!