February 24, 2010
Checking the Numbers
The Crystal Orb of SIERA
Few Phillies fans will forget the date Sept. 27, 2008, when the eventual World Series champions clinched their second consecutive National League crown in quite remarkable fashion. Leading the Nationals 4-3 in the top of the ninth inning, Brad Lidge had found himself in another one of his save opportunities requiring a bottle of Pepto Bismol. The bases were loaded with only one out as the dangerous Ryan Zimmerman stepped into the batter's box. Zimmerman then smoked a 1-1 pitch from Lidge up the middle. Jimmy Rollins dove to his left, flipped to Chase Utley, who fired to Ryan Howard to complete the 6-4-3 double play, ending the game, and sending the Phillies to the playoffs.
Rollins is one of the best defenders at his position and was able to snag a hard-hit ball that most other shortstops would fail to reach. For instance, if Phillies backup Eric Bruntlett were manning the position, the odds were great that Zimmerman's scorched grounder reaches the outfield, plating two runs and putting the Nationals in the position to steal a victory.
Aside from the obvious playoff implications in the above scenario, Lidge's ERA increases quite a bit with Bruntlett captaining the infield instead of Rollins. Why should this occur when Lidge or any other pitcher has no control over who plays in the field or how they are positioned? It makes little sense to penalize Lidge for the hypothetical runs that would have scored with Bruntlett up the middle when they would have been prevented had a better defender been playing the field. This is essentially the precursor for DIPS Theory-Defense Independent Pitching Statistics-a field of study that gauges pitchers based on the events they can control, for which they should be penalized or rewarded.
Cue SIERA, Skill Interactive Earned Run Average, a statistic created by Baseball Prospectus that generates a pitcher's ERA while bypassing the events that could fall under the defense or luck subheadings. The reason for avoiding these types of events is that they are not consistent from year to year but can result in a pitcher looking better or worse than what his actual skills would dictate. Another of its advantages is that, because it uses statistics found to be consistent from year to year, it is a very good predictor of a pitcher's ERA in the following season. Therefore, SIERA can be used to identify hurlers who could potentially surprise or disappoint in 2010, based on a 2009 disconnect between ERA and SIERA.
Using SIERA, let's take a look at which pitchers that had relatively poor 2009 campaigns who may surprise fans this season:
Aside from his rather ugly ERA, Nolasco really had a great 2009, striking out over nine batters per nine innings with an average of just 1.8 unintentional walks per nine innings. In fact, his 4.43 strikeout-to-walk ratio was identical to his rate in 2008, when he pitched to a 3.52 ERA in 212.1 frames. Barring a circumstance unforeseen, Nolasco should return to his dominant ways and battle Josh Johnson for the role of Feisty Fish ace.
At 26, the top pick in the 2006 draft still has time to prove he can stick in a big-league rotation, and if his SIERA is any indication, Hochevar is in line to have a stellar 2010. Plenty of bloops and bleeders plagued his outings last season, but a decent strikeout rate, slightly below-average walk rate, and high ground-ball rate make for a pitcher worth keeping an eye on. He might not make the All-Star team but a 4.20 ERA or one in that vicinity in the American League is very impressive.
Bush is a curious fellow in the sense that his underlying numbers are what we might expect from a mid-rotation starter and not the bottom-of-the-pile pitcher he has looked like recently. His K/BB ratio has remained strong over the last three years but his ground-ball rate has spiraled downward every year since 2005. Assuming he can regain some of those worm-beaters, Bush should rebound nicely.
Resisting the urge to snark about his disastrous tenure with the Yankees, Pavano actually pitched very well last season, with strikeout, walk and ground-ball rates resembling that which he posted in his final season with the Marlins in 2004. A .335 batting average on balls in play (BABIP) hurt his ERA, but if that rate can take a few steps downward he should be a very effective starter… as long as he is healthy.
It feels odd suggesting that someone with an above-average ERA could "surprise" this season, but, no matter how much one digs it is difficult to pinpoint numerically exactly where his 2008 and 2009 differed. He may not realistically be the 3.09 ERA pitcher as he was in 2008, but he is certainly better than what we would expect from a 4.32 ERA pitcher. Settling in the middle could give the Phillies the most potent 1-2 punch in the senior circuit with Hamels teaming with off-season acquisition Roy Halladay.
And, again with the help of SIERA, let's look at five pitchers who could disappoint this season:
Jurrjens generates plenty of grounders and whiffs batters at a slightly above-average rate, but needs to pare back his walks to remain as effective as his numbers look over the last two seasons. He won't be a bad pitcher this season but it is unlikely he resembles an ace or All-Star hurler. However, Jurrjens did just turn 24, so he could very well still be entering his own. This will make an interesting test case for SIERA.
What the defending NL champions will gain back with Hamels, they are likely to give back with Happ, who, like Jurrjens, won't pitch poorly this season but will struggle to keep up the reputation of a stopper or front-line hurler. Happ has relied on a sneaky fastball and tricky delivery so far, but it is only a matter of time before hitters catch up.
Wells and Happ found themselves intertwined for much of the 2009 seasons as NL Rookie of the Year candidates and also in that the controllable numbers indicated both would regress in 2010. Wells barely touches 90 mph with his fastball and he doesn't miss many bats, but he is stingy with walks and keeps the ball on the ground. All of that is well and good, but what one would expect from a third or fourth starter, which is what his SIERA indicates.Tim Lincecum to form a dominant duo atop the Giants rotation last season, but while Tiny Tim was every bit as good as his ERA suggested, Cain relied on above-average success with stranding base runners to beautify his mark. He substantially reduced his walk rate but saw his rate of strikeouts take a tumble as well. His true talent likely lies somewhere between last year's 2.89 and the 3.76 ERA he posted the year before.
Though the Arlington heat can be tough on pitchers, Millwood was able to strand plenty of base runners last season and keep his BABIP well below the .300 threshold en route to a shiny ERA that was not indicative of his actual skill set. His strikeout and walk rates took substantial hits and he served up more home runs. The move to Baltimore could lop a bit off of his expected ERA but not much.
A version of this story originally appeared on ESPN Insider .