By Tommy Rancel

Spring Training marks a period of experimentation. With a month of meaningless games to fill, it’s the perfect time to try things that you would never think of once Opening Day arrives. Most experiments go nowhere and end in mid-March. Occasionally one will work, such as last season’s move of Skip Schumaker from the outfield to second base for the St. Louis Cardinals.

The Texas Rangers are doing some experimentation of their own by giving set-up man and former closer C.J. Wilson a chance to compete for a rotation spot. In recent seasons we’ve seen Adam Wainwright, Ryan Dempster, and Braden Looper make successful moves from reliever to starter – albeit to differing degrees.

It was not too long ago when Wilson was the Rangers’ closer. After solid numbers in 2007 (66 games, 3.02 ERA, 12 saves), Rangers Manager Ron Washington tapped the lefty as the team’s bullpen stopper in 2008. However, things didn’t go according to plan. In 2008, Wilson saved 24 games, but finished with a dismal 6.02 ERA in 50 games.

The biggest difference between the two seasons was a disparity in home runs. After allowing just four homers in 68.1 innings in 2007, Wilson gave up eight in just 46.1 innings in 2008. 

The huge jump can be explained by a ridiculously high home run-to-fly ball rate (HR/FB). For his career, Wilson owns a HR/FB rate of 12.6%. But in 2008, his HR/FB skyrocketed to 16.8%, up from 2007 levels, when batters hit home runs on 8.2% of the flyballs hit off Wilson. 

Wilson benefited from regression last year, when his HR/FB dropped to a career-low 6.3%, in the process lowered his home runs allowed per nine innings rate (HR/9) to 0.37 – also a career best. Filling the set-up role and backup closer role in support of the Rangers’ top closer choice Frank Francisco, Wilson finished the season with 14 saves and set career marks in several categories, including: wins (5), games pitched (74), innings pitched (73.2), strikeouts per nine innings (K/9) (10.26), and ERA (2.81).

Despite the career year out of the bullpen, Wilson is still auditioning for a rotation spot in the Texas Rangers rotation. Why? Because in most cases 150 innings of average starting pitching is more valuable to a team than a 60 good middle relief innings. 

As strong as Wilson’s 2009 campaign was, his contributions added up to a decent but not overwhelming 2 wins above replacement level (WAR represents the number of wins a player adds to his team over your typical 25th man on a roster). According to fangraphs.com, Wilson was one of just one of 10 relievers to hit the 2 WAR threshold, showing the limited upside inherent in a role that severely limits innings pitched. On the other hand, 63 starting pitchers were worth 2 WAR or more.

According to B-Rank (Bloomberg’s proprietary ranking of all players), Wilson’s stellar relief numbers earn him an overall ranking of 336. In most leagues he is not even drafted. On the other hand, Joe Saunders ranks 271th despite posting a nearly league average 4.60 ERA and an ugly 5.17 FIP (Fielding Independent Pitching, a stat that runs along a similar scale to ERA, while stripping out factors a pitcher can’t control such as defense and luck). He is also drafted in almost every league regardless for format. This is just another example of the value of league average starting pitching over above average relief

The move from bullpen to rotation is a tricky one. It’s generally accepted that a pitcher can see increases in ERA, strikeouts and velocity moving from the rotation to the bullpen. New Yankee Chan Ho Park is a good example of his phenomenon. As a starter, Park owns a career 4.39 ERA. His ERA plunged to 2.58 last season, on the strength of a strikeout rate that reached its highest point for a full season since 2001. Park also saw an increase in velocity: He posted an average fastball velocity of 92.2 MPH last season, up from his career average of 90.3 MPH. Even if Wilson makes a successful transition, he will likely see the reverse effect that Park – and other failed starts like Eric Gagne and Jose Mesa – experienced in moving to the bullpen from the rotation. In 2009, only two qualified major league starters posted K/9’s above 10.00: two-time National League Cy Young winner Tim Lincecum (10.42) and Justin Verlander (10.09). 

In addition to the strikeouts, there are also the looming home run concerns. As mentioned, Wilson enjoyed career best marks of 6.3% HR/FB and 0.37 HR/9 in 2009. Beyond regression to the mean, neither are sustainable as a starting pitcher. In 2009, only five starters had HR/9 less than 0.50: Chris Carpenter (0.33), Clayton Kershaw (0.37), the aforementioned Lincecum (0.41), AL Cy Young Winner Zack Greinke (0.43) and Joel Pinero (0.46). Looking at HR/FB rates, five starters had rates below 6.0%, including four names from above: Carpenter (4.6%), Kershaw (4.1%), Lincecum (5.5%) and Greinke(4.5%), along with Carlos Zambrano (5.6%).

Outside of the statistical regressions, the biggest question is can Wilson handle the toll of pitching every fifth day over a 162-game season? After spending nearly five seasons in the bullpen, Wilson will be asked to throw at least 150 innings and upwards of 100 pitches per start. Last season, he threw 25 or
more pitches 14 times and never went over 38 pitches in one appearance. The good news is Wilson’s numbers reflect that he does better when pitching on more days rest. With zero days rest in between appearances, Wilson’s slash line against was .325/.385/.482 (opponents’ AVG/OBP/SLG) last year. When working on three days rest that went down to .190/.261/.238; on four days it was .190/.190/.238 on four days. (Please note small sample sizes in all cases.) 

In addition to the physical aspect of starting, Wilson will need to work on pitch selection. A relief pitcher can get by with one dominant pitch and a secondary pitch. A starting pitcher generally needs at least three pitches to deal with the opposing lineup three or four times a night. Wilson has an above average fastball and a decent slider, but is lacking a true third option. In 2009, he toyed with a cut fastball and a curveball, but neither was used more than 5.6% of the time. He will need to up the usage of both pitches, or concentrate on just one as his third pitch, to balance out his pitch selection. 

When taking everything into account – statistical regressions, the physical impact and pitch selection – Wilson is facing an uphill battle. But it’s a battle worth fighting. Despite being one of the games better relievers last season, Wilson was not a valuable fantasy player, because pitchers can only produce so much value (in fantasy or real life) as non-closing relief pitchers. However, if Wilson can parlay his solid historical strikeout rates into a league-average ERA as a starter, his value would jump considerably. I wouldn’t put Wilson on any draft boards as a starter right now; but you should add him to your watch list. 

To keep an eye on C.J. Wilson, and hundreds of other players, check out Bloomberg Sports’ fantasy kits