by Eno Sarris //

We’ve talked in this space about different ways that a pitcher can improve his game. He can refine a pitch and increase his whiff rate on a specific pitch like Johnny Cueto, or alter his pitching mix like Mike Pelfrey. Usually that change happens right in front of our eyes in American baseball. In the case of Colby Lewis, we actually have a pitcher that seems to have done both by undergoing a sea change abroad.

Lewis left MLB for Japan after the 2007 season, following an ERA north of six with the Rangers and Athletics. He struggled to strike out batters at an above-average rate, couldn’t keep walks down (career 4.8 BB/9 before this year), and didn’t have a proclivity for worm-burning, as his career groundball percentage is below average (39.9%). There wasn’t much that suggested he was going to succeed in the major leagues other than his good minor league statistics (3.39 ERA, 1.185 WHIP, 8.7 K/9, 2.5 BB/9).

Then he went to Japan. In those two years, Lewis had a strikeout rate over one per inning, and walked a minuscule 1.16 batters per nine. His ERA was sparkling and under three both years, he led the Japanese leagues in WHIP one year, and he took the strikeout crown both years. You could say that he made good use of his time there.

Now that he’s back, he’s striking people out, not walking people as often as he did before, and has even become the subject of fawning love letters in the media. Okay, it’s not quite a love letter, but you get the picture. Despite his poor start on Sunday, Lewis is still striking out more than a batter per inning, and walking fewer than four batters per nine innings. The Bloomberg Sports Fantasy Tools spider chart shows us that his current walk and strikeout rates make him a more than capable fantasy as an SP2 in a deep league, or SP3 in a shallower league. The dot closest to him is Justin Verlander. 

First off, he’s been a little lucky, but not incredibly so. His BABIP will rise (.261 currently, around .300 MLB-wide), and he’ll strand fewer runners (78.1% this year, 70% across baseball). As more dinks and dunks fall into play, more runners will cross the plate instead of ending the inning on the basepaths. Lewis is also a flyball pitcher that is only giving up 0.88 HR/9 because of a slightly-generous 8.3% HR/FB rate. That number usually approaches 9-11% across baseball, so a couple more long flies could turn into homers soon, especially in the favorable hitting environment at Arlington.

All that said, Lewis’ xFIP (a number that focuses on strikeouts and walks and strips out batted-ball luck, and produces a number on the ERA scale) is still 3.99; that seems to be a good estimate of his true talent level. He’s obviously striking people out and has made real progress finding the strike zone.

Most of his progression came from altering his pitching mix to feature his slider. In 2003, Lewis used his fastball 74.6% of the time and his slider 2.4% of the time. FanGraphs tracks a stat called linear weights, which uses game states before and after a pitch to measure the effectiveness of each type of pitch. In 2003, Lewis’ fastball was ‘worth’ -25.8 runs. That number is legendarily bad, as Carl Pavano owned last year’s worst fastball with -23.6 runs and Zach Duke‘s -19.4 runs fastball was second-worst. Surprisingly for such a bad season, Lewis’ slider was still worth +1.2 runs that year.

After five years of featuring that below-average fastball and hiding his slider, which was the only pitch that was consistently positive by linear weights, it took two years in Japan for the light to go on permanently. Now that Lewis is back, he’s using the slider 30.3% of the time, and it’s his best pitch by linear weights (+7.8 runs). Altering his pitching mix has made all of his pitches more effective, as his fastball is finally a positive (+2.1 runs).

Lewis’ slider is his most effective pitch at getting whiffs this year (15%, 8.5% is average), and in particular it’s great low and away as Dave Allen showed on FanGraphs.com. Using Patrick Newman’s pitch f/x tracker for the Japanese leagues, we can see that he refined the pitch while in Japan. Take a look at the image below, which shows how often he threw the slider in a typical start (5/22/09 in this case, and the sliders are yellow).

The biggest remaining question is which way Lewis’ walk rate will go. Obviously, he was having trouble in that category before he left for Japan, as his career rate suggests (4.86 BB/9). And then he dominated in that category in Japan, where the strike zone is called a little bit larger and walks are not as prominent in the baseball culture. For example, Patrick Newman had this to say about the Japanese strike zone:

“My (unofficial) translation of the official rule is “the strike zone’s
upper limit is the point mid-way between the batter’s shoulders and the
top of his pants, the lower limit is the bottom of the batter’s knees,
and covers the area over homeplate”. So that’s not too far off the MLB
strike zone. In practice, I have noticed that the umpires can get a
little generous at times.”

Of course we can expect his walks to come in closer to the major league average (3.59 BB/9) than his elite Japanese rates, but if he does only walk batters at an average rate, his strikeout ability will play well enough to make that work as a package. On the other hand, if the walk rate starts creeping significantly over 4 (and after his bad start Sunday, it’s at 3.68 BB/9), Lewis may have some trouble.

In the meantime, Lewis is a hold in shallow leagues – you wouldn’t get much for such an unestablished non-prospect pitcher anyway – and a testament to the ability of pitchers to change. With that new focus on the slider, he’s practically a new pitcher.

For more on Colby Lewis and other surging pitchers, check out Bloomberg Sports’ Fantasy Tools.