Happy Thanksgiving! Regularly Scheduled Articles Will Resume Monday, December 1
December 30, 2005
Looking Back...and Looking Ahead
I have a collection of lines in a file like "Vazquez trade nice for Sox" and "Glaus deal?!?!", fragments of columns never written. I've just never been able to get caught up this December, what with travel to the winter meetings and now, spending the holidays back east in an apartment barely wired for the last century, much less this one. ("Dear everyone in my mother's building: Could SOMEONE have a Wi-Fi Internet connection installed? Love, Joe.")
I intend to do a stack dump of those notes, because I've missed covering some significant moves--and have the e-mails to prove it--but it'll have to happen next week. It's just been very difficult to find blocks of time and access to information, and putting those two together has been a virtual impossibility. Come Monday, I'll be back amidst my technology--if on the wrong coast--again, and can get caught up on all the things I wanted to be writing over the past couple of weeks.
With today being the last business day of 2005, I wanted to write about some of the things I learned over the past year. There's no way to go through an entire year of covering the game without picking up new things, and incorporating those lessons into my analysis is a key part of what I do. In the same way that I like to insist that people within the game need to learn from the new information available to them, it's important that I, as a performance analyst, continue to re-evaluate my premises and learn about the game.
I don't know that any new lessons jump out at me as I look back at 2005, but a number of things I'd learned in recent seasons took on additional prominence. Most notable among these is the value of defense, as applied by the Chicago White Sox in their run to a championship. The Sox had a so-so offense and a good pitching staff, but what turned them into a title team was a defense that never quit, turning batted balls into outs at an amazing rate. Even more than the '02 Angels or '03 Marlins, the 2005 White Sox were a team that played defense better than it did anything else, and by shutting down opposing teams with range, limited run scoring and built a championship-caliber team.
The lesson here is that defense can be a weapon. To apply that lesson, though, we'll have to figure out how to better incorporate defensive evaluation on a team level to prognostication. When the Sox traded for Scott Podsednik, perhaps I should have considered the effect he would have on their outfield defense, giving them two center fielders. Looking ahead to 2006, does the swap of Troy Glaus for Orlando Hudson have a significant effect on the run prevention of two teams? Does adding Hudson, probably the best defensive second baseman in baseball, make Brandon Webb a much better pitcher? Does upgrading two defensive spots make the Diamondbacks a threat in the NL West?
Defense is still a challenge for performance analysts, but the White Sox reinforce the lessons of the early 21st century: that defense isn't an afterthought, that it can have a massive impact on the course of a season. We have to find a way to not be surprised by the next White Sox, or for that matter, the next team that disappoints because it doesn't play good defense.
If the White Sox followed in the footsteps of the Angels and Mariners, the Astros retraced their own path by starting 15-30 before turning their season around, winning the NL wild card and advancing to the World Series. In the AL, the A's and Indians overcame horrific starts to insert themselves into the playoff picture before falling short in the season's last week.
I'm not sure what we can do about this, but it has serious implications for anyone trying to write about baseball. At what point in a season can you evaluate a baseball team if any one of them is seemingly capable of being both the worst team in the league and the best at different times during the season? I wrote off the Astros in August of 2004 and the summer of 2005, only to see them go on terrific runs that ended deep in October. The A's and Indians were among the game's worst offensive teams well into May of last year, only to turn things around and become run-scoring machines for most of the summer.
At what point in a season does performance become real? "Sample size" has long been a catchphrase of performance analysts, although it's one we largely use in considering individuals. What kind of sample size do we need, especially in this era of imbalanced schedules, to properly evaluate an entire team? The last couple of seasons have shown us that even four months might not be enough time to get a handle of what a team can do, and certainly that two or three months' performance can be misleading. How seriously can we take the standings on Memorial Day, or July 4, with the memory of the last two years fresh in our minds? Is even an entire season sufficient to reach conclusions, given what we know about the gaps between win-loss record and the underlying performance?
As with defense, I'm unsure how exactly to incorporate this into my work. I mean, I don't think I can just take the first two months of the season--or more--off each year, but it's a challenge to make salient observations when you see just how misleading 20, 30, 40 games of baseball can be.
This is, to a certain extent, a business issue as much as a baseball one. How does a performance analyst who writes about the game for a living produce entertaining, substantial content before Memorial Day?
With both of these notions, I think what I'm getting at is the notion of uncertainty. Even taking into account all of the available information, baseball has a large element of unpredictability, from the career paths of players to the way in which collections perform together--defensive synergies, rather than "chemistry"--to the variability of play within a season. I've long argued that analysis weighted with excessive caveats is of little utility (and certainly makes for a boring read), but the more I watch the game, the more I see the need to temper my own statements, especially forward-looking ones, with the idea that we really don't know what's going to happen in the short or medium term.
While I love that as a fan, I don't entirely know what to do with it as a writer or analyst. Figuring that out may be the biggest challenge I face in 2006 and beyond.
There are other things that pop into my head as I look back at 2005. What can we say about players who peak very early? I got to thinking about this with Glaus, who was an MVP candidate at 23 and who hasn't come close to that kind of season since. Is there a good way to identify players who are simply having an outlier season, as opposed to those showing true development? What are the markers, and how do we fold them into our evaluations?
Frustration was a problem, as I sometimes felt like the work that analysts have been doing for two decades hasn't made a dent in the industry. When I see five-year contracts for mid-rotation starters with injury histories, or the persistent dealing of actual value for perceived value, or the continued focus on perceived personal characteristics--too often shorthand for "quotability"--rather than baseball performance, I want to throw up. Well, my hands, anyway. If anything, it feels like baseball has taken a step backwards in the past couple of years, with a backlash against the work of, and people behind, performance analysis. Accusations of arrogance are largely monodirectional, although the tone of outsiders has mellowed over the years even as the attacks on informed outsiders have become even more vicious.
My frustration has likely carried over into this space, especially in the latter part of the season. This was a hard year, personally and professionally, and I know that affected my output. Setting aside the personal stuff, it's hard to keep analysis fresh while staying true to certain principles, hard to write both for people who own Baseball Prospectus 2006 and those who just got a gift subscription to the Web site this week. I've been writing regularly about baseball for six years now, and while that's a fraction of the tenure someone like Peter Gammons has, it's long enough to force you to self-evaluate in an effort to avoid stagnation and self-parody.
In 2006, I want to keep learning about baseball in as many ways as I can. I want to apply the lessons to my work, to be a better writer and analyst. I want to make my writing entertaining for long-time readers and accessible for the burgeoning crop of new ones. I want to continue questioning everything I read and hear, whether from Nate Silver or Joe Morgan, and be open to the idea that the things I know are wrong.
To all of you who read BP online, who listen to the radio show and buy the books and tell your friends about the work we do, thank you. I hope you've had a wonderful holiday season, and that you'll go on to a happy, healthy and successful 2006.