Tuesday, February 20, 2007

Best of the East?

Jimmy Rollins' comments that the Philles are the team to beat in the NL East seem to have stunned ESPN during a time when they're desperate for baseball news. Could a team that's been out of the playoffs for 13 years really rise up and take down the powerful Mets next year?

In short, yes. My analysis shows the Phillies are not only projected for the most wins in their division, but also the most in the NL. Admittedly, this is by the slimmest of margins--the Mets and Diamondbacks are each within one win of the Phils' projection, and four other teams are within three games.

This does not, of course, mean that Philadelphia will definitely take the flag. With the Mets nearly neck-and-neck and the Braves and Marlins within striking distance, they'll probably win the East less than 40% of the time, which is still a very good number in this era of competitive balance.

Without a flurry of offseason activity, or a bunch of hot youngsters coming up, why are the Phillies now (slightly) ahead of the Mets in the division race?

Lineup: Talent and Balance

The Phillies are one of the NL's only teams that has both a good middle of the batting order and no big holes in the lineup. In Rollins, NL MVP Ryan Howard, and Chase Utley, the Phillies have three of the 30 most valuable hitters in baseball relative to their positions. They are supplemented by the solid Pat Burrell, who has become oddly underrated due to all the flak he gets from the media.

Aaron Rowand, Shane Victorino, and Wes Helms all are projected by PECOTA to provide roughly league-average production. When your seventh-best hitter is an average player at his position, you're happy with it, unless your name is Steinbrenner. At catcher, the Phils should get acceptable production from Rod Barajas and Carlos Ruiz, though they would probably be better off letting Ruiz play full-time.

Top-to-bottom, I would argue that only the Mets have a better everyday lineup in the NL. The Phils do have a weak bench except for Jayson Werth, but this isn't a huge concern.

Rotation Depth

Much has been made of Philadelphia's rotation going six deep this year. In Cole Hamels, Brett Myers, and Freddy Garcia, they have one of the best 1-2-3 combinations in the NL. Adam Eaton, Jon Lieber, and Jamie Moyer are less promising, but they represent a significant competitive advantage over the back ends of the other NL rotations. By comparison, look at the Nationals' staff, which recently prompted this John Patterson quote, courtesy of the Washington Post:

"I mean, I am excited. This is an opportunity that everybody looks for, whether it's a team that only has one starter or not. It's still the opportunity to pitch on Opening Day and be the head of a staff."

The Phillies are rumored to be shopping Jon Lieber, but it seems clear to me that Jamie Moyer should be the odd man out; I doubt they would have re-signed him had they known Freddy Garcia was on his way over. Moyer is going to allow roughly one HR every six innings this year, and he doesn't have the strikeout potential to offset that handicap. You'll often hear Moyer be called "crafty," but that word belies his lack of the skills that traditionally correlate well with pitching success. When was the last time anyone called Johan Santana or Ben Sheets crafty?

Regardless of whether Moyer or Lieber takes the last rotation spot, it's likely the other will see regular time as an injury replacement. Either way, the Phillies will feature one of the top rotations in the NL.

The Weakness

Now we come to the Phillies' Achilles' heel, their bullpen. Yes, it is very thin behind Tom Gordon. No, it doesn't matter that much. While a stud closer can be worth several wins to a contending team, middle relievers are simply not that big a deal.

In baseball, preventing ten additional runs is generally worth about a one-win difference in the standings. For a relief pitcher who throws 60 innings per year, that's a gap of 1.50 runs of ERA. Any halfway decent farm system has a surplus of guys who can contribute a 4.75 ERA in middle relief. To gain one marginal win, you have to sign someone who's projected for a 3.25 ERA. There are very few pitchers like this, and they will usually demand closer money, a multi-year deal at $7 million or more per annum.

Go shopping on the free agent market, and you're more likely to pick up someone like Jamie Walker, who will be worth about .5 wins this year for Baltimore--and who knows what in 2008 and 2009--at a total cost of $12 million. When the Orioles miss the playoffs again this year, they may scratch their heads and wonder why their shiny new bullpen didn't carry them, but baseball analysts everywhere will barely take notice.

The Enemy

Of course, the main factor in the Phillies' rise to the top is that the Mets are almost certainly going to win fewer games this year. The Mets experienced a surplus of breakout seasons last year, with better-than-expected seasons from Jose Reyes, Carlos Beltran, Jose Valentin, Paul Lo Duca, Endy Chavez, and basically the entire pitching staff sans Pedro and Victor Zambrano. This year, there is a badly injured Pedro who may be expected to contribute 60 innings at less than his normal effectiveness, no Chad Bradford, no Darren Oliver, and significant regression from all last season's breakouts.

This doesn't mean the Mets are doomed. They still have the NL's best top-to-bottom lineup (assuming Lastings Milledge is playing every day) and a very good bullpen. But the rotation is counting on two fortysomethings at the top and three guys who have recent ERAs in the vicinity of 6.00. It shouldn't be considered a huge surprise if the Mets spend October planning tee times rather than deciding what brand of champagne to stock the locker room with.

Saturday, February 03, 2007

The 2007 White Sox

Last winter, the White Sox boldly retooled their defending champion roster, trading for Jim Thome and Javier Vazquez after breaking the bank (by the standards of the time) to re-sign Paul Konerko. I was an outspoken critic of these moves, not only because the Sox gave up a lot of talent--including top prospect Chris Young--but also because they weren't the kind of team that would really benefit from the additions.

Even after the Southsiders penciled three big names into their 2006 lineup card, they were projected for just 82 wins by both PECOTA and Diamond Mind simulations. Why was a 99-win roster plus two big-name additions suddenly about to regress to .500? Because the 2005 White Sox were a highly aberrant team, winning on the strength of career years for virtually every member of the pitching staff, plus outperforming their Pythagorean record by 8 games. When every pitcher outperforms expectations, the staff as a whole is expected to regress the following year, which they did.

The White Sox still won 90 games last year, primarily because they stayed remarkably healthy, but also owing to a few overachievers, this time on the other side of the ball: Joe Crede, Jim Thome, Paul Konerko and especially Jermaine Dye all handily beat their preseason expectations. As with the pitchers the year before, it is likely that each will see their value drop significantly in 2007. To wit:


Year Dye Thome Crede Konerko
2003 .514 .958 .741 .704
2004 .793 .977 .717 .894
2005 .845 .712 .757 .909
2006 1.007 1.014 .829 .932

Career .825 .974 .763 .849

2007 .879 .906 .795 .885

The 2007 Line represents each player's PECOTA forecast. What the chart demonstrates, other than Jim Thome's career-long studliness, is that the middle of the Pale Hose order is due for a correction. What about the pitching? Will they bounce back to 2005's level?


Year Buehrle Vazquez Contreras Garland
2004 3.89 4.91 5.50 4.89
2005 3.12 4.42 3.61 3.50
2006 4.99 4.84 4.27 4.51

Career 3.83 4.34 4.28 4.44

2007 4.83 4.50 4.96 4.77

The ERA forecasts may look brutal, but PECOTA usually appears pessimistic for pitchers; Johan Santana is the only starter projected for a 3.25 or lower ERA who will begin 2007 on an active roster. Furthermore, all except Vazquez are finesse pitchers, and pitchers who cannot rack up strikeouts are much more prone to regression. Notice, however, that three of the four are projected to have a substantially lower ERA than in 2004, which isn't that far ago in baseball time. The common theme, though, is that the magic of 2005 isn't coming back.

A third factor towards a probable 2007 regression is the team's health. Last year, the Sox racked up a total of 15 DL days, all from Jose Contreras missing two starts. That's a remarkable record, one that puts the team on the short list of healthiest teams ever. Perhaps some of this is the result of good training techniques, but it has more to do with simple good fortune, which has no predictive value. There's little reason to believe the Sox will be much healthier than an average team in 2007.

There's really nothing here that someone who passed Stats 100 can't follow. Two years running, the Sox have had several aberrant individual performances, plus good health and good karma, which has led to 189 wins, but doesn't bode well for the team's future. My predictions model, based on 2007 PECOTA forecasts adjusted for playing time, portends a 74-win season for a team that has no superstar talent--a .900 OPS at DH or a corner spot in the AL is hardly a superstar--plus a lineup with several offensive zeros and a very tough schedule.

Of course, that's not the view that the public is taking. Rather, they believe that the 2007 Sox will combine 2005's pitching success with 2006's hitting success and jump back to the top of the division, ignoring both common mathematical sense and the current zip codes of the 2005 bullpen that was the championship team's biggest strength.

Yesterday, MLB.com previewed the AL Central. Not surprisingly, Jim Molony picks the White Sox to win the division, although he also expects the Tigers to be better than they were in 2006 and still somehow finish second, the Indians to have an "excellent chance to make the playoffs" if their bullpen succeeds, the fourth-place Twins to still contend for the division crown, and the fifth-place Royals to win 74 games. Apparently he feels the Central teams will win about 70% of their games outside the division.

Molony's delusion may begin with his assertion that Sox GM Ken Williams "worked to improve his staff this winter." Um, what? If I'm banking on winning a World Series right now, trading Freddy Garcia and Brandon McCarthy for Gavin Floyd, Gio Gonzalez, and John Danks is not the first step I'd take.

This is not to say I don't like these trades; I think Williams won both of them. However, I do think that he is operating from a different angle. Williams is presumably a smart enough guy to know that his team is not talented enough to be a leading contender for a title in 2007, though of course he won't come out and say it. Without the need to load up for '07, he can make trades that maximize his long-term return. Giving up one year of Freddy Garcia for many potentially cheap and useful years from young pitchers is the opposite of the moves Williams has made for years, trading promising young players for Garcia, Todd Ritchie, Vazquez, Thome, Carl Everett, and others. These trades resulted in some short-term gains, but deprived the Sox of years of cheap service time from Chris Young, Kip Wells, Josh Fogg, Jon Rauch, Gary Majewski, Luis Vizcaino, and Gio Gonzalez before he was re-acquired.

Though the Sox won a championship, they didn't do it through expensive trade acquisitions, but rather with homegrown talent and deft signings. Garcia and Everett played bit parts in the World Series run, while Crede, Garland, Aaron Rowand, Juan Uribe, Tadahito Iguchi, Bobby Jenks, Dustin Hermanson and others formed the core of a team while earning a relative pittance in salaries.

These trades don't make the Sox better off in 2007, but they should return a profit in every year beyond that, and that's what really counts. If Williams can flip Buehrle and Vazquez for good returns at the deadline, the seeds will be planted for a solid future.