Cliff Lee: 3.18 ERA, 3.03 SIERA
Lee looked like he might be on his way to another playoff gem through the first two innings of Game One of the World Series against the Giants, but things immediately soured. He gave up seven runs (six earned) en route to a shortened 4 2/3-innings start. He allowed eight hits, including five doubles. Of course, Lee’s peripherals from Game One look like a patented Cliftonian performance: seven strikeouts, one walk. This is where the philosophy of DIPS Theory and metrics like SIERA face a real challenge, because the Saber Orthodoxy would declare that Lee was unlucky in Game One and leave it at that. They would say that pitchers do not exert much control over balls in play, and the fact that he allowed 15 batted balls and eight went for hits would be an indication that he had bad luck. Well… he certainly was not receiving any good luck.

However, this is where the most important facts of SIERA and DIPS metrics need to be extracted from the party line and thought about more clearly. It was not hard to watch Game One and see Lee did not look like he was throwing as well. He was up in the zone. His curveball did not have the same bite. Does that mean DIPS theory is wrong? No. Lee was probably a bit unlucky and making mistakes.

DIPS Theory began when Voros McCracken discovered that BABIP was not repeatable. Pitchers with very bad BABIPs one year were unlikely to have very bad BABIPs the following year, and pitchers with very good BABIPs one year were unlikely to have very good BABIPs the following year. That does not mean it was not something that they exhibited any control over at the time. It means that the things they did to affect their BABIP are fleeting. If Lee gets a 1-2 count on a hitter and surprises him with a nasty cut fastball that breaks his bat and leads to a weak ground out, he made a good pitch. If a hitter sees that 1-2 count the next season and has well-documented data showing that Lee throws a lot of cutters on 1-2 counts, he may be better prepared to foul it off or pull it over the third baseman’s head for a single. Of course, if Lee has a tendency to throw a nasty curveball on 1-2 counts that looks like it will be a strike but is not and leads to whiffs, the batter may have less recourse to deal with this the next time—if hitters do not know where the ball is going to be when it reaches the plate, they are going to keep missing it. But if they can get their bat on it in the first place, minor adjustments can resolve the situation for the better the following time.

This is not to say that there is no luck in BABIP. We know that statistically there needs to be about a third of pitchers who fall further than .020 points from their true BABIP skill in any season. The sample size tells us that much fluctuation is due to happen—in other words, you do not get exactly five heads when you flip a coin 10 times and you do not get exactly 30 hits on 100 balls in play. But we also know that some of the variation in BABIP comes from adjustments. What that means is that Lee’s dominance that led to a .235 BABIP in the ALDS and ALCS is not going to prevent him from allowing a .533 BABIP in World Series Game One. It also means that his .533 BABIP in Game One is not going to imply he is any more hittable in Game Five. However, his 41/2 strikeout-to-walk ratio in his first four post-season games is far more repeatable. The Giants will need to pull another rabbit out of the BABIP hat Monday night if they are going to beat Lee again.

From Game One: After following up a Cy Young season with a stint as a playoff hero just a year ago, Lee actually improved his strikeout rate in 2010 from 18.7 to 22.0 percent while dropping his walk rate from 4.4 to a microscopic 2.1. Lee's batted-ball rates were reliably average yet again, but he was able to keep runs off the board thanks to his incredible walk rate. The key to beating Lee is getting hits because he is not going to let you on base on his own, as he walked only 18 hitters all year. They say that knowing is half the battle, but it is difficult to see how simply knowing that Lee is unlikely to walk you will get you halfway to a win. He has struck out 34 of 87 hitters in three post-season starts in 2010 and walked only one hitter, upping his game in the playoffs for the second year in a row. He has won three of the seven games that the Rangers needed to reach the World Series and will get a chance to give them two of the remaining four, beginning tonight. The Giants lineup is easier than that of the Rays and Yankees, but he will probably not be able to meet expectations this evening for no other reason than the fact that he has set the bar so ridiculously high for himself. At the same time, the Rangers have the better pitcher on the mound, and it is easy to envision another brilliant performance from the rent-a-mega-ace.

Tim Lincecum: 3.43 ERA, 3.16 SIERA 
There would not have been many Giants fans who would have smiled if you told them that Lincecum would allow four runs and be chased after 5 2/3 innings in Game One of the World Series. However, this ended up being more than sufficient for the Giants, who crushed Lee’s supposedly unhittable pitches and won 11-7. Lincecum threw 93 pitches en route to only three strikeouts and two walks, while generating 10 ground-ball outs of 22 balls in play. This was a sharp contrast to his NLDS performance against the Braves and even some of his weaker starts in the NLCS against the Phillies. Lincecum is still a great pitcher, and these small sample sizes say little about how fans can expect him to perform in Game Five. Look for one of the best pitchers in the game on the hill Monday night for the Giants.

 From Game One: This might have seemed like a more insurmountable challenge for the Rangers a year ago when Lincecum won his second straight National League Cy Young award, but the 25-year-old’s ERA increased by 0.95 runs in 2010. However, his SIERA only went up by 0.43 runs, indicating some of his return to mortality was rooted in bad luck. His strikeout rate did decline from his lofty 2008 and 2009 levels of 28.6 and 28.8 percent to 25.8 in 2010. While striking out as many hitters as any starting pitcher did during his Cy Young years, Lincecum was able to get away with mediocre walk and ground-ball rates. However, as his velocity declined, he became slightly more hittable and batters were able to get more runs off him. Lincecum did put up a career-best 50 percent ground-ball rate in 2010, suggesting that he is learning how to pitch smarter. However, he also had some bad luck as well—his BABIP was .315, primarily due to a 20.9 percent line-drive rate. This sounds bad, but line-drive rate is the least persistent pitcher statistic. In his career, Lincecum has allowed a .301 BABIP, so there is little reason to expect this to change. Lincecum will still strike out about a quarter of hitters he faces, and in the 2010 postseason he has struck out 34 percent of hitters he has faced. While this came in part due to striking out 14 of 30 Braves in Game One of the NLDS, he still struck out 16 of 59 Phillies he faced in the NLCS. The righty may not be the best pitcher in the league anymore, but he definitely is among the best. As the Giants go up against Cliff Lee in Game One of the World Series, Lincecum will probably need to be on top of his game, but there is little doubt that The Freak is a formidable opponent.  

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe