July 25, 2008
The Unpredictability of Pitching
After 100 games in the 2008 season, the top three ERA leaders are Justin Duchscherer, Cliff Lee, and Edinson Volquez. There have certainly been surprises with hitters as well, but not to this extent. The day that baseball becomes predictable will also be the time it becomes boring, but 2008 has brought an especially erratic season for starting pitchers.
There are numerous aspects of this that are important to examine, if for the sole reason of discovering whether waiting to draft starting pitchers is prudent in fantasy baseball, something that's no doubt paid off handsomely so far this year. One obvious reason is the unnatural act of throwing a baseball, which leaves pitchers far more susceptible to injury than a hitter is, making it a position that's inherently riskier to rely on when building your team. That still leaves the main issue unanswered, though: How can so many unheralded and often undrafted hurlers be so instructive to winning fantasy baseball in 2008?
The Big Three
Justin Duchscherer: Entering 2008, Duchscherer hadn't started a game since 2003, and yet he's been baseball's most consistent starter this season. He's allowed two runs or fewer in all but one start, and he gave up just three in the other. He's used McAfee Coliseum's spacious foul territory to his advantage (1.36 ERA, 0.71 WHIP), but Duchscherer's ability to keep hitters off balance has produced solid results regardless of venue. Still, this is one example of a small sample size resulting in a mirage. Duchscherer's .215 BABIP is the lowest mark in all of baseball. His .82 strand rate is the highest mark in all of baseball. His 5.5 K/9 is below league average, while his 2.7 strikeouts for every walk ratio ranks 28th. Just 5.8 percent of his fly balls have turned into home runs, which is the second lowest in the league. In other words, no pitcher in baseball has been luckier in 2008. Add in the fact that he's already passed his career high in innings by 19 1/3, and has nearly doubled his total from the past two seasons combined, and Duchscherer is a candidate to both wear down and regress badly. It's unlikely he's still among the ERA leaders come season's end.
Cliff Lee: After posting a 6.29 ERA and 1.52 WHIP in 2007, Lee followed that up with a 5.68 ERA during six starts in spring training this year, gaining a spot the last rotation slot with the Tribe mainly for lack of alternatives. Fast forward to July 24, and he leads the American League with a 2.29 ERA. There's unpredictable, and then there's Lee's 2008 campaign. Of course, he was a former fourth-round pick who was solid in 2005, so some skills were clearly hidden beneath the surface. His first six starts this season were historically good (0.67 ERA, 44:2 K:BB ratio), but the question remained: Is he for real? With a major league-leading ratio of 5.5 strikeouts for every walk, the answer is yes. Only 4.2 percent of Lee's fly balls have left the park, which is the lowest rate in baseball. But that just raises the question: Can someone with a 2.29 ERA not have some luck involved? Unless he's striking out 15 batters per nine innings, probably not, and in Lee's case, his .301 BABIP is hardly lucky. Moreover, because of his newly-discovered pinpoint control (1.4 BB/9), even when more of those fly balls leave the yard, many will be solo shots, and Lee has done a solidly average job of keeping the ball on the ground anyway (1.1 G/F). Some luck has no doubt contributed to Lee's breakout 2008, but he's also developed a skill set that suggests he's now an elite hurler. He's here to stay.
Edinson Volquez: After going 13 starts to open the season without allowing more than two runs, Volquez has come back down to earth, as his ERA has nearly doubled since then. A former highly-regarded prospect who turned in a dominant spring after switching to the easier league, Volquez's 2008 season isn't as huge of a surprise as Duchscherer or Lee have been, but no one expected this kind of ERA and strikeout totals. Poor command (4.3 BB/9) has led to inefficiency, as he's yet to toss more than seven innings in a single game this year. Despite throwing almost nothing but fastballs and changeups, Volquez's 9.5 K/9 is truly elite, and his 1.4 G/F ratio is also fantastic. He doesn't have a lucky hit rate, but his .81 strand rate is the second highest in baseball, and likely unsustainable. Additionally, he's struck out three batters or fewer in four of his last five starts, so he's already beginning to wear down. Volquez has a ton of ability and the skills to be a top-notch starter for years to come, but more growing pains are in store for the immediate future.
The Next Tier
John Danks: Thought to be the lesser property in a trade involving Brandon McCarthy, much like Volquez, Danks has thrived since leaving Texas. Also like Volquez, Danks' 2008 breakout season has been a mixture of luck combined with an impressive true skill set. His ratio of 2.9 strikeouts per walk reveals he's no fluke, but a 6.5 percent HR/FB rate is the fifth-lowest mark in the league. That's especially lucky when you consider that US Cellular Field is one of the most homer-friendly parks in baseball. Danks is very unlikely to finish with a sub-3.00 ERA, but his component numbers suggest his breakout is legit.
Joe Saunders: Saunders is the poster boy for revealing how four months of good fortune can spurn Cy Young talk about an otherwise mediocre pitcher. He's done a fine job limiting baserunners (1.14 WHIP) and inducing ground balls (1.3 G/F), and good run support has led to 12 wins. However, his BABIP (.246) is the fifth lowest, his strand rate (.78) the eighth highest, and his 4.5 K/9 mark is "good" for 89th in baseball. It's not all luck, as Saunders has allowed the fewest line drives (13.5 percent) in the league this year, but his '08 season has been more a result of luck than of a changed skill set, meaning his ERA should soon head north toward his career levels.
Kyle Lohse: To get an idea of how Lohse was valued entering the season, consider that he went unsigned until the starter-desperate Cardinals offered him a one-year deal in mid-March. He now ranks second in wins in baseball (12), and boasts a tidy 3.35 ERA. Lohse had been pitching in the NL for a year and half before 2008, so that switch can't be credited. He's induced more ground balls and the Cardinals field a solid defense behind him, and Dave Duncan magically works wonders for veteran pitchers-but good fortune is nevertheless mostly to blame. Lohse's 4.7 K/9 mark is not commensurate with the rest of his numbers and sticks out like a sore thumb, and BP's LUCK metric, as measured by the number of extra wins and short losses the pitcher actually has versus his expected record, ranks him the fourth-luckiest starting pitcher in MLB.
Ryan Dempster: Most hurlers find it easier to pitch out of the bullpen as opposed to the starting rotation. Dempster, on the other hand, has transformed himself from a mediocre relief entity to one of the best starters in the league this year. Despite improvement in the area, he still walks too many batters (3.5 BB/9) to be an elite starter in the long term, and his current .259 BABIP is largely responsible for the sparkling 3.05 ERA. That's not to say that he can't be an effective starter, as his K rate is solid and he keeps the ball on the ground (1.3 G/F). Just don't expect any more All-Star appearances.
Gavin Floyd: The double-digit wins, 3.50 ERA, and 1.20 WHIP are all pretty, but the component stats are ugly: 3.8 walks and 6.2 Ks per nine are both below league average, while a .225 BABIP and .77 strand rate are both extremely lucky. With a DERA of 4.36, Floyd is clearly someone who is unlikely to keep up his pace, and a severe crash may be looming.
Johnny Cueto: An inconsistent rookie still learning how to pitch, Cueto has struck out more batters per nine innings (8.3) than Dan Haren, who has probably been the National League's best pitcher so far this season. Improved control is a must, but Cueto's stuff is future ace material. A remarkable 16 percent of his fly balls have gone for homers, which is the fifth-highest mark in baseball. Of course, Great American Ballpark is a contributing factor, but he doesn't give up an inordinate amount of fly balls, so regression towards the mean is likely. Cueto has endured bad luck and possesses huge upside moving forward.
A.J. Burnett: Pitching for a new contract, Burnett's 4.73 ERA and 1.44 WHIP have been a major disappointment. He's walking too many batters, but his .336 BABIP is 50 points higher than his career mark. Burnett's 9.0 K/9 is sixth best in the league, and his 1.4 G/F ratio is also impressive. He's already lowered his ERA a full half-run over his last three starts, so the correction has already begun. Health permitting, Burnett is likely to finish with numbers resembling his career marks.
Aaron Harang: Harang entered 2008 coming off back-to-back seasons in which he threw more than 3,500 pitches, so he was a health risk. A record of 3-11 with a 4.76 ERA ensued, followed by a trip to the disabled list. Still, Harang wasn't pitching nearly as poorly as the numbers indicate, partially a result from an abnormally high BABIP (.331). Before going down, his strikeout and walk rates remained intact, and his 3.2:1 K/BB ratio was elite (and better than Johan Santana's). It's ridiculous that 165 pitchers have more wins than Harang this season, so if he can return to health, expect his superior component stats to start improving his win and ERA numbers.
Since pitchers have less control over outcomes than hitters, and their stats will fluctuate more often. A pitcher can even go an entire season with incredibly good (or bad) luck, but for the most part, after 35 starts, the true pitcher will reveal himself. This makes the position the best to trade for in fantasy leagues. While timing the market is never advisable, looking at component numbers reveals far more than wins/losses and ERA. This season has produced more out-of-nowhere dominant starters than usual, but the key is deciding which ones are getting lucky, and which have truly improved their skill sets in ways which will directly lead to positive predictive measures for how they'll do moving forward. While some surprises are here to stay, the end of year ERA leader board is highly unlikely to resemble the current one, as the larger the sample size, the greater the likelihood the flukes are revealed. So remember, draft based on skill level (K rate, K:BB ratio, ability to limit home runs, etc.) and trust that the areas out of the pitchers' control, like wins and ERA, will follow.
Dalton Del Don is a contributor to Rotowire.com; he can be reached here.