Same Song, Second Verse

Why has it been seven years since the New York Yankees won the World Series? Roster construction, situational hitting, bullpen depth, and managerial decision making have certainly played a roll in The Boss’ torment, but starting pitching stands out as the number one factor differentiating the three-peat Yankees from the $200 million albatross continually crapping out before the pennant. The continuity of Andy Pettitte, David Cone and Orlando Hernandez served them well through the late ‘90s, and more recent teams have scrambled to find the right mix throughout the regular season, and especially into the postseason.

My point here is not that the Yankees need to rededicate themselves to finding consistent starting pitching. While that may be a desirable end for New Yorkers, I believe it is more instructive to look at teams who have captured lightning in a bottle long enough to get good performances out of an entire starting rotation. The 2005 Pale Hoes blew people’s sox off by winning about 20 more games than anyone expected, mostly behind the simultaneous ascension of the entire starting rotation. In looking for the 2006 version of Cinderella, very few analysts fingered the Detroit Tigers, and it would be hard to blame them. The sheer improbability that Detroit would make the leap from mediocrity to dominance the year after reshaping their entire roster did not sound like a plausible proposition at the time, and it still seems at least a bit quizzical.

Looking at the teams’ components, though, it ought not shock anyone that the Tigers were up to the task. In broad strokes, the team featured a fly catching outfield and a rangy infield without any pronounced weaknesses. Offensively, they mimicked the White Sox top-to-bottom consistency, as well as their lack of an overpowering force that would distinguish them from other average offenses. The pitching staff, though, seemed to have far less upside than the one that broke through for Chicago the year before. Jeremy Bonderman and Justin Verlander both had high ceilings, but neither had achieved anything like greatness in the majors. And the rest of the rotation did not seem to provide the depth that kept the Sox in so many games the year before, as Kenny Rogers and Mike Maroth struck fear in the hearts of few. Nonetheless, the Tigers exploded early in the season and kept the pace long enough to get to the World Series. Here, I will take a deeper look at the numbers behind the similarities to see if there is some sort of trend that has benefited the Sox and the Tigers at the expense of teams like the Yankees.

Last	Year	IP	ERA	EqH9	EqBB9	EqSO9	EqHR9	VORP	BABIP
Buehrle	2005	198.3	4.47	10.2	2.0	5.1	1.0	35.1	NA
Garcia	2005	178.7	4.55	9.4	2.6	6.5	1.1	30.5	NA
Contreras	2005	129.3	4.91	8.9	3.6	7.3	1.1	17.1	NA
Garland	2005	169.3	5.05	10.1	2.9	4.7	1.2	19.7	NA
Hernandez	2005	113.0	4.35	8.8	3.2	7.0	1.1	24.2	NA
McCarthy	2005	82.3	4.77	9.3	2.4	6.5	1.4	12.7	NA
Buehrle	actual	236.7	3.12	9.1	1.5	5.7	0.8	54.8	0.295
Garcia	actual	228.0	3.87	8.9	2.4	5.8	1.0	45.7	0.285
Contreras	actual	204.7	3.61	7.8	3.3	6.8	1.0	42.1	0.263
Garland	actual	221.0	3.50	8.6	1.9	4.7	1.0	50.7	0.27
Hernandez	actual	128.3	5.12	9.6	3.5	6.4	1.3	6.9	0.306
McCarthy	actual	67.0	4.03	8.3	2.3	6.5	1.8	13.6	0.251
Last	Year	IP	ERA	EqH9	EqBB9	EqSO9	EqHR9	VORP	BABIP
Verlander	2006	153.0	4.20	8.7	2.9	6.5	1.0	19.4	.287
Robertson	2006	184.7	4.35	9.3	2.8	5.6	0.9	18.9	.295
Rogers	2006	167.0	4.87	10.2	2.6	4.3	1.1	7.1	.299
Bonderman	2006	194.7	4.05	8.8	2.8	6.8	0.9	28.1	.294
Maroth	2006	192.7	4.41	9.8	2.2	4.6	1.0	19.4	.296
Ledezma	2006	78.7	5.39	9.9	3.8	5.8	1.2	-0.5	.304
Verlander	actual	186.0	3.63	9.0	2.9	6.0	1.0	47.5	.297
Robertson	actual	208.7	4.28	8.9	2.9	5.9	1.3	42.4	.281
Rogers	actual	204.0	3.84	8.6	2.7	4.4	1.0	40.6	.265
Bonderman	actual	214.0	4.08	9.0	2.7	8.5	0.8	39.8	.323
Maroth	actual	53.7	4.36	10.7	2.7	4.0	1.8	10.0	.291
Ledezma	actual	60.3	3.58	9.0	3.4	5.8	0.8	12.6	.286
Miner	actual	93.0	4.84	9.7	3.1	5.7	1.1	9.6	.305

The data in the table compares PECOTA’s projections before each team’s breakout season with the starting pitchers’ actual contributions. Although the projections for the teams’ pitchers do not look wildly different, it is important to note that not everyone saw them on the same path going into their respective years. The White Sox were often seen in an unflattering light. Having just traded Carlos Lee for Scott Podsednik, it seemed that their powerful offense would have more trouble keeping up with a solid, yet unspectacular, pitching staff. Joe Sheehan predicted 71 wins, largely due to the fact that he did not believe in the starting rotation.

“The pitching staff isn't going to make up for the falloff in runs scored. Freddy Garcia and Mark Buehrle are mid-rotation innings guys who get paid like aces. They head a rotation that's long on hope--not just one, but two Cubans with shaky recent track records exiled from the Bronx--and short on anything you can rely upon.”

I do not mean to pick on Sheehan; I believe his comments are representative of many fans going into 2005. The White Sox came out of the dark to the point that I was still anticipating their collapse when they were ahead 2-0 in the World Series. The Tigers, on the other hand, at least looked as if they had some chance to improve. Sheehan picked them to win 78 games, commenting that they would be “an interesting team to watch.” While short of contention, he at least saw some positives coming out of the motor city.

Obviously, the results of the games told a different story, as each team came close to 100 wins and caught fire in the postseason. Is there some common theme that we should have identified to help us see these teams coming? Looking into the data, there does not seem to be any systematic shortcoming causing under-projections for either team. Chicago skinned the cat by dramatically limiting base runners: every started except El Duque outperformed both the walk and hit surrendered projections. On average, the White Sox were 9% better than the projections in hits allowed per nine, and a whopping 16% better in walks per nine innings pitched. Naturally, teams will have trouble scoring runs when most of their plate appearances end in outs.

Detroit’s projections were not so far off. Like Hernandez in Chicago, Mike Maroth was the lone disappointment standing out from a group that met or exceeded expectations, although this unit’s improvements were less staggering. As an aggregate, their actual performance was within 3% of the cumulative projections for all of the major rate stats- hits, walks, strikeouts and homeruns. They beat or met their ERA projections across the board, so some of the variance comes from factors outside of this relatively broad scope (such as stranding runners, turning double plays, and other factors over which the pitcher has less control). Despite these differences, I can identify five key similarities between the two staffs, some by design and some by happenstance.

1. An ace taking a big step forward, regardless of defensive help. Mark Buehrle’s career year was a major factor in Chicago’s 2005 success. He did not replicate that step in 2006, and has started shakily in 2005. Nonetheless, Buehrle was fantastic in 2005, far above the label of innings eater, primarily because he made very few mistakes (miniscule 1.5 BB/9 and 0.8 HR/9 rates). These numbers indicate that he trusted a very strong defensive unit, and did so without cutting down on his strikeout rate. Justin Verlander did not have the consistent track record that Buehrle had going into his breakout season, but he was a highly touted prospect who arrived all at once. Across the board, his numbers came very close to PECOTA’s projections, except for his ERA, which was more than half of a run better. That luck combined with his solid durability at a tender age made him the most valuable of the team’s starting pitchers, at least in terms of Value Over Replacement Player.

2. A mid-rotation starter who looked much better by getting defensive help and limiting walks. Jon Garland, meet Kenny Rogers. Sure, the gambler has a much longer rap sheet than Garland, but look at how much each player benefited from his defense. Even though there were no projections for batting average on balls in play in 2005, notice that every prediction for 2006 hovered between about .290 and .300 with small variations for extreme groundball or flyball pitchers. Garland’s defense took about 25 points off of that average, giving up only a .270 average on balls in play. Rogers got even more help with a .265 BABIP, though he did not slice a full walk per nine innings off of his rate projection like Garland did. In any case, each pitcher chipped in over 200 innings with an ERA below 4.00, and very strong defenses deserve some of the credit in both occurrences.

3. Beating innings projections by keeping their most effective pitchers healthy. One of the interesting aspects of the Yankees’ starting pitching is that they consistently had 4 starters making 27 or more starts in each of their four World Championship seasons in the ‘90s, and have returned to that level of consistency only twice in the last six years. With four 200 inning pitchers apiece, the White Sox and Tigers got consistent performances out of their best pitchers, which helps for the simple reason that each inning pitched by a good starter is one that does not have to be completed by a lesser replacement. I do not know if there is some demonstrable effect on a defense or a pitcher’s endurance after remaining consistent through the regular season, but these teams’ rates of success seem to indicate that there is even more benefit to keeping pitchers healthy than is readily apparent. The coaching staffs for the teams also deserve credit for getting the most out of their starters without overworking them and keeping them ready throughout the season.

4. Getting some luck by losing their ineffective pitchers to injury and having young starters replace them with greater success. It is a valuable skill to be able to keep your effective players healthy. It is considerably luckier to have the ineffective ones break down. Perhaps Orlando Hernandez and Mike Maroth were hindered by pitching with pain, which would explain some of their struggles. If not, the teams were quite fortunate that they went down, since Brandon McCarthy, Zach Miner (for whom no PECOTA projection even existed in 2006), and Wil Ledezma provided much stronger innings than the men they replaced. It is no surprise that pitching depth is a good thing, and these teams did well to stock up on major league ready talent that did not cost several million dollars to serve as insurance policies.

5. Little regression among older players. While there is a great deal of volatility with younger players, both teams have to be thankful that guys like Jose Contreras and Kenny Rogers did not start feeling the effects of age. El Duque showed some wear and tear, but even he was better than most pitchers in their late thirties (or maybe even mid 40s, who knows).

Ideally, these similarities would provide some sort of formula for projecting future breakouts. Unfortunately, I do not think it is quite so clean as that. A handful of teams have some of the characteristics, but lack some key components. The Phillies have a rotation that goes six deep with breakout candidates at the top in Cole Hamels and Brett Myers. Without any semblance of a defense, though, it is hard to see them impersonating the White Sox and Tigers. The Diamondbacks have a good defense and lots of name pitchers. On the other hand, I do not see Randy Johnson, Doug Davis, and Livan Hernandez all making it through the season healthy. Maybe the best candidate is yet another AL Central team, the Cleveland Indians. With lots of young pitching talent and an improved defense, it would only take some surprises from Paul Byrd and Jake Westbrook to change a pretty good unit into an exceptional one.