Friday, April 04, 2008

A 2008 Billy Beane Lesson in Adaptation
It's Not 'Change', It's How You Change Changes  

You can never step on
the same home plate twice - Heraclitus

The last skill a manager can master, and only if the manager has mastered all the others, is managing change. That's Home Plate in the Management by Baseball model, and one of the widely-acknowledged masters of it is Oakland A's GM Billy Beane.

Beane's front office team have done it again -- found new ways to apply change management and planning to push organizational achievement. It doesn't look as revolutionary as Moneyball did, but it requires the application of new techniques.

If your tank is already full on Moneyball and competitive change management, skip down to the next header, otherwise, the next bit of background will be useful.

Beane's front office techniques, and those of his innovative staff, were publicized through Michael Lewis' engaging, and influential beyond baseball, book Moneyball. It became influential because the Athletics and Beane gave Lewis an uncommon level of access (teams sometimes give full access to a writer, for example the Cardinals' kimono-flashing to gee-whiz-meister Buzz Bissinger for his book Three Nights  in August, but Bissinger doesn't bring the weight of knowledge and insight to a subject Michael Lewis does) and Beane and his workgroup were willing to reveal to the world all their "tricks". Those "tricks" weren't really tricks at all; they were proven techniques imported from other several endeavors and then given a baseball context.

Beane & his workgroup were fearless about sharing that with the world because...

  • It's really hard for management, even very good management, to import techniques from other lines of work,
  • It's uncommon that managers are able to synthesize theirs own industry-specific methods into imported techniques in a context-sensitive way (it's more common that practices, when imported, are taken virtually whole-hog), so they tend to miss opportunities to yield more, or sometimes they fail entirely, and,
  • Lewis didn't reveal (or perhaps didn't know) the key technique that made the change methods so powerful.

That undocumented technique was one that was close to explicit to Moneyball readers who "do" innovation for a living (like Joe Ely or Dave Pollard) or who work for Toyota, or those who know the work of Deming, or just about anyone reasonably experienced who has a management position in baseball.

That undocumented approach is that you change the changes on an on-going basis. Mutation, adaptation, testing tweaks and big jumps, are all constant. So Change is not a moment you can bring about with a plan and schedule (though you'll need plenty of both of those), change is an ongoing process. You Observe, Measure, Analyse and Adapt. 

And it's not one of those tidy-but-egregiously-insightless Powerpoint slides illustrating a cycle that everyone uses because it's de rigeur to include it in the "deck"; you have to do all of those actions simultaneously, every day.

So the Athletics front office were glad to reveal their methods because they knew by the time Moneyball hit readers' eyes, they would already be doing their details differently. It's not just that Beane, et.al. didn't have to lose a competitive edge by laying it all out, they might actually gain a competitive edge by having slow boats chasing methods they had already discarded for more evolved ones...a double whammy.

So while the Alderson-Beane-DePodesta-and-others front office brought surprisingly competitive baseball to a small-market franchise owned by bottom-line humping real-estate guys, the string of success appeared to snap in 2007. Last year was the first sub-.500 season since the one Moneyball was being written:

2007 AL West 76-86 (.469) 3
2006 AL West 93-69 (.574) DIV 1
2005 AL West 88-74 (.543) 2
2004 AL West 91-71 (.562) 2
2003 AL West 96-66 (.593) DIV 1
2002 AL West 103-59 (.636) DIV 1
2001 AL West 102-60 (.630) WC  2
2000 AL West 91-70 (.565) DIV 1
1999 AL West 87-75 (.537) 2

Oakland Athletics Team History & Encyclopedia - Baseball-Reference.com

Note, please,  neither won/lost record for 2006 and 2007 are fully indicative of the team's ability. In 2006, the A's had a supernatural power over the Seattle Mariners. The M's, otherwise an above .500 team, went cherry-pie time against the A's, and Oakland's pride went 17-2 against Seattle. Had they gone 10-9 instead, they wouldn't have won the division - they would have been in 3rd place behind the M's. In 2007, Seattle worked on countering the specific tactics the A's had used to beat them down in 2006 and turned the tables, beating Oakland 14 out of 19 games. Had the split been a more even 9-10 in favor of Seattle, Oakland's record would have been 80-82. Not a flustercluck, just a disappointment.

The real (not popularly assumed) idea of the Moneyball approach was that the A's front office could analyze the market every year and take advantage of an evolving set of overlays and underlays to be one step ahead of competitors that were bound to exist within the areas the resource-constrained franchise could afford to invest in.

Like all comparative advantages, the Alderson/Beane/DePodesta design for ongoing analysis/action worked. Until it didn't. Their approach started under-performing relative to recent past performance because:

  • The feedback loop in Baseball is tough: you have to keep change rolling every day or take a chance of losing difference-making edges, and at the same time,

  • The resource-thin model allows for fewer mistakes, and at the same time the lack of resources allows less time for staff to tend to Big Picture stuff.

I'll get back to the results of that die-namic duo in a few paragraphs.

It didn't put a (functionally) ~.500 team on their field last year because it "failed". It underachieved last year relative to previous years because the panoply of techniques the A's had resources to pursue had been largely neutralized by like-minded analysts working in other front offices.

According to several articles (one from a San Francisco newspaper I can find no on-line link to), the Oakland front office realized they had fallen behind in scouting Latin America, an activity that had been a chosen casualty in previous tight times. So, an act of changing change to changing back. And, articles claimed, the A's are investing in proprietary medical research into preventing player injuries or shortening recovery time (the Colorado Rockies started investing in medical research a few years ago. That research appears to have played a part in the team's ability to balance its home-road performance, and enhance its late-season endurance, precursors, I think, to greater success). So, an act of changing by launching resources at the non-standard.

John Donovan of Sports Illustrated put together the best summary I can find on-line (found thanks to BTF):

The new course of action was charted last October, shortly after the A's had concluded their worst season since 1998. Beane sat down with Forst and others and conducted a top-to-bottom organizational review, Oakland's first in years. A new strength coach was brought in to examine why players were having such a huge problem with injuries. The A's also hired five new area scouts, adding to a staff that had become one of the smallest in the league, and re-sectioned the country to get better coverage. They increased their scouting budget in Latin America and elsewhere internationally, and have earmarked more money for signing bonuses.

The change in scouting has been especially noteworthy, considering the criticism the team has faced for its mediocre drafts over the past several years. Haren was traded to the Diamondbacks for six prospects last month; Baseball Prospectus immediately ranked four of them among the top seven players in Oakland's organization.

The cold, inescapable fact is that if resources are limited enough, gravity will pull even the most determinedly innovative organisation down towards the mean average.

The Athletics' approach mutated every year, but always within the same paradigm. Competitors adapted not slowly, but surely, to the A's paradigm (if not the twists the A's were introducing each year) to the point where in a zero-sum system, the yield for the A's diminished.

What should they have done to avoid that?

Hold top-to-bottom organizational reviews annually or more frequently.

Why didn't they?

Because on an extreme resource diet, managers have to make a lot of decisions about what to leave in & what to leave out, and the seduction of current success (exhibit 1: that team standings chart, previously) is to think that strategic, big-picture stuff might be less inexpendible than their more (apparently) urgent needs are.

Even the Yankees & Red Sox can't do everything their bright staffs can think up. Managers make choices, balancing present and future, importance and urgency.

So when the basic A's paradigm was being neutralized, the A's front office was squeezing out the resources they had to other needs. You can function with fewer scouts in the field than competitors if you can make their methods more effective/efficient. If your resources are tight enough, you squeeze out slack (getting leaner by eventually cutting muscle when you run out of fat). In a changing environment, you don't always know what will turn out to be muscle. In addition, the top of the organization has been passionately engaged in a set of real estate negotiations and designs around the franchise's new ballpark, and those non-baseball concerns are sucking up attention and (already limited) resources.

So, to have a top-to-bottom review in a slack-stripped system is a significant cost that you pull from each of the system contributors' time buckets. More scouts would have meant more slack would have made such a thing easier. Fewer scouts mean less slack, less tolerance for normal error, less time for not-here-and-now activities. And as long as the team was winning playoff spots, this seemed tolerable

The reality of limited resources is that, yes, it is, more often than not, the mother of invention, but it equally may be the Uncle Ernie of change management, cascading what is a tough situation into a tougher one.

There's no guarantee that pulling a piece from the pre-Moneyball paradigm will revolutionize the outcome. There's no guarantee that the non-standard medical research will make for a newer paradigm to replace Moneyball -- and, in fact, I suspect that would be very unlikely. But until the staff comes up with something new to break both the A's current and competitors' models, more change was necessary, and these both seem like reasonable additions (and the medical research has to be great brain candy

This page is powered by Blogger. Isn't yours?

free website counter