Friday, January 6, 2012

Light Bulb: MLB Front Offices and the Power of Information

Lately, I've come to a number of realizations, not all of them baseball related but all of them relevant. First, the quality (such as it is) of the writing on this blog is directly related to what I read. If I'm not being exposed to new ideas and information, much of which I can apply to baseball and the Reds, then my writing tends to stagnate. Second, and as an example of the first point, over the past week a number of writings have combined to help reshape my view of Major League Baseball front offices, their role and how they operate.  

In part, my revised view is a happy accident relating to the collision between (1) a very good book ("The Powers That Be" by David Halberstam) that I'm reading about the information distribution business (aka: media) and how it shapes all facets of American life and (2) the revolution happening in the NL Central this offseason. It has taken a surprisingly long time for the new wave of baseball thinking to reach the Central, but this offseason has seen it hit with tsunami like force. For all intents and purposes, the wave started on the West Coast where the A's were looking for a way to compete. It spread gradually in a few directions, but the next big step was when it was carried across the country and landed in the uber-competitive AL East where the Boston Red Sox were looking to change decades upon decades of futility. Now, after penetrating both coasts, the new wave of thinking and operating is coming to an NL Central team near you. Both the Astros and, most noticeably, the Cubs have changed GMs and are beginning to completely overhaul their front offices and their overall methods of operating. That makes for interesting times in the NL Central and challenging times in the future for the Reds.

In light of the efforts of the Northside regime to, once again, overcome decades and decades of futility, I've become more interested in how and why these new front offices are able to effectively change the culture. We're talking about cultures so flawed and consistently ineffective that fans and media created "Curse" narratives, Bambinos and Billygoats, to explain their lack of success. Facing that level of consistent futility, it can be safely assumed that there was something greater at work than the individual talents of the GM. The problem had to be systemic, as countless GMs were rotated in and out of the top posts in these organizations and yet the futility continued. So, the problems clearly ran deeper than just the individual talents of one person.

That was the state of my thinking when I read the following from David Halberstam:

"What it lacked, (Ben) Bradlee and his friend Phil Geyelin sometimes agreed, was cruising speed. The New York Times had cruising speed, the Times was not a product of one or two men's talent and brilliance, it was the sum of its many parts, often at the expense of the individual talent. The Times was special because of its awesome and often stifling structure, it could carry weaker reporters and raise them to the general level of the product just as frequently as it pulled more talented reporters down to that same level. The Times was on a plateau, a moderately high one, all by itself; the (Washington) Post was a series of peaks and valleys."

Even though the blurb discusses the issue in the context of newspapers, it applies equally to organizations of all types. To build a consistently successful organization requires implementation of an effective structure. In Major League Baseball, an effective structure is one that supports quality decision making by consistently increasing the probability of making a successful decision: Cruising speed. A less structured organization may be able to generate success in spurts, but any type of consistent, sustained success requires a structure. It's the system vs. the individual distinction, basically the difference between an organization built on the individual talents of a GM and an organization built on a structure that lends itself to a systematic way of operating.

With that in mind, it becomes much clearer what the new regimes in Chicago and Houston are attempting to do. Basically, harness the power of all the organizational assets into an effective and efficient system for the collection and distribution of data. Over the years, unsuccessful organizations have either had no structure at all or, even worse, an ineffective structure that has been ingrained into the organizational fabric by each and every decision made. The efforts of these front offices to install new structures became even more clear when I stumbled across this story, discussing Jeff Luhnow's efforts to rebuild Houston's front office, on

HOUSTON -- Jeff Luhnow has long been a proponent of sabermetrics, so it shouldn't come as a surprise that one of his first duties as Astros general manager was to bring in someone to help sort through the massive amount of data that's available. 

Luhnow announced Tuesday he had hired Sig Mejdal from the Cardinals, where he had worked with Luhnow for the previous six years and was most recently the team's director of amateur draft analytics. Mejdal's title with the Astros will be director of decision sciences. 

The Astros also announced they had hired Stephanie Wilka as coordinator of amateur scouting, a position previously held by Mike Burns before he took a job as an area scout with the Blue Jays.

Mejdal was involved with modeling, analysis and data-driven decision-making throughout all levels of the Cardinals organization and was a key contributor in the Draft decision processes that led to more Major League players than any other organization during that timeframe. 

"It's a capacity that he was involved in and helping us build when I was with the Cardinals, and he had a lot skills and passion in that area," said Luhnow, who was hired away from the Cardinals last month. "He's going to be a key member of the front office." 

Luhnow described Mejdal's new role as a systematic method of combining all the information you can collect on players, whether it's using previous performance information, health and medical information and opinions of scouts who have laid eyes on players. 

"It's really a systemic approach to combining all the information into a decision-making tool to assist people, whether it's the scouting director or farm director or general manager, in making decisions," Luhnow said. "There is a pretty significant component with it that's data-driven and based off of using stats from the past to project future performance. It's definitely a critical component." 

Wilka, meanwhile, has previously worked in the PR department of the Red Sox and served as executive director of the philanthropic arm of the Dodgers. She has an undergraduate degree from Harvard and a law degree from Pepperdine. 

"She has a passion about baseball and should be a good addition," Luhnow said.
Mejdal earned two engineering degrees at the University of California at Davis and later completed advanced degrees in operations research and cognitive psychology/human factors. He has also worked at Lockheed Martin in California and for NASA. 

Mejdal has been active in baseball statistics and analytics since earning his membership in The Society for Baseball Research (SABR) while in grade school.

The article lays out the approach that the Astros wish to implement and the importance they place on information of all types and from all sources. The long running debate of sabermetrics vs. traditional scouting has overlooked the key point, namely that ALL valid information is of value to the organization. The Astros' decision making is data-driven and ensuring both the quality and volume of that data is paramount to the success of such a system.

To me, there are also inferences to be drawn from the article based on both the people hired and their respective job titles. It's clear that Luhnow's prime objective in the early going is improving the information gathering and distribution abilities of the organization. After all, a decision maker is only as good as the information on which he relies.

Oddly enough, I first heard of Sig Mejdal from the book Fantasyland, which detailed a writer's attempt to win his fantasy baseball league by hiring a team that blended the traditional scouting viewpoint with statistical analysis, the latter viewpoint strongly represented by statistical analyst Sig Mejdal. But, one of things that strikes me as particularly interesting in this blurb is his job title: Director of Decision Sciences. A job title which seems to both (1) indicate the importance the organization places on data and (2) set Mejdal's individual responsibilities as everything right up to the decision point. 

In a data-driven system like this, the information flows in an upward direction from all points in the organization. In essence, the organization is a pyramid with all information flowing from the base up to the top, where the decision makers reside. If you generate a consistent, reliable flow of information, then you improve the ability and effectiveness of the decision maker. And, the longer this flow of information runs through these defined channels, the more and more ingrained they become in the organizational fabric (forget Billygoats, this is the real curse of inept baseball organizations). Further, the organization can have sustained success because, to a certain extent, once the decision maker installs this system he himself becomes more fungible. If all the information for making the proper decision is readily at hand for the decision maker, then the specific abilities of that decision maker become somewhat less important. The decision is driven by the process, not the individual. At that point, you have established a measure of organizational intelligence, which supersedes, and should outlast, any single individual. 

As for the second hire, Stephanie Wilka, at first it seemed an odd one. Admittedly, I don't know anything about Ms. Wilka, so I'm only going off what is written in this article. At first, she didn't strike me as being qualified (though she could certainly be an undercover scouting genius), but upon further review it's patently obvious that it was my own understanding of the job qualifications that was wrong. If someone is hired as the Coordinator of Amateur Scouting despite having no background in scouting (bringing instead public relations and ??philanthropy?? experience to the table), then it seems likely that her responsibilities are something other than rating players on the 20-80 scale. In actuality, I believe her job is ensuring the proper flow of information up the ranks. She's not there to scout, but rather to manage those who do, ensuring both the quality and flow of the scouting information coming from her underlings. In short, the emphasis in her job title should be front-loaded, as she is more "Coordinator" than "Amateur Scout". It's important to have those who generate the information, but it's equally important to have those who verify the quality of the information and facilitate its distribution. Again, this hiring seems to be an organizational effort to ensure that good information flows up to the top of the pyramid for consideration by the decision makers. 

Finally, once the information is gathered and arrives at the pinnacle of the pyramid, it has to be properly compiled and presented to the decision maker for it to be useful. So, it's not enough just to collect and compile the information, but it has to be readily sortable and accessible for the decision maker. And, not surprisingly, many organizations have implemented proprietary computer programs to manage the data generated by so many different sources.

For example, the recent Sports Illustrated Moneyball issue had an article that discussed "Carmine", the Boston Red Sox proprietary data management program: 

Now I can click on the screen... ." Epstein says as he executes a few keystrokes, "... and call up ... there's [Jacoby] Ellsbury."

A trove of information pops up about the Boston centerfielder. It includes some proprietary statistical data, including Boston's in-house defensive metrics (the popular ones cannot be trusted, especially in one-year samples) and an overall empirical valuation that combines offense, defense and baserunning metrics. Carmine automatically updates the numbers, including projections, every day for every player in professional baseball.
Carmine also guards less quantifiable data, including the first report filed on Ellsbury in 2003, two years before he was drafted, as well as eight follow-up reports chock-full of anecdotes culled from interviews with his coaches, trainers, college SID, opposing coaches, summer league coach and others. There is a story about a foot injury "no one knew about" that explained a brief slump in the Cape Cod League in '04. There is the story of the day in '05 when two cross-checkers worked out Ellsbury in a San Diego gym because of rain: The 21-year-old picked up a stray basketball and threw down a monster dunk, confirming their reports on his athleticism. All those notations—which a decade ago would have consumed just two or three sentences under "makeup" in most teams' player files—are separate from the actual nuts-and-bolts scouting report on his skills.

Each spring the typical area scout for the Red Sox will follow about 50 players on his watch list. Cross-checkers will see about 100 players each. They all file background reports with Carmine every time they see a player. "That's a lot of information," Epstein says. "But that's where you get the edge. You're not going to have the most success with the most obvious, readily available information."

In short, I have gained a new found appreciation for the power of information and the importance of those mechanisms, institutions, and entities that distribute said information. It is becoming more and more apparent that in the modern game, knowledge truly IS power. As fans, it's fun for us to think of GMs as guys who wheel-and-deal to make winning trades solely on the basis of their individual brilliance and charisma. In reality, the role of the GM is to design and construct an organization that supports decision making by increasing the probability of making correct decisions. Think of the massive advantage to be reaped by an organization structured in such a way that the probability of making a successful decision is increased by 15% on each and every decision (i.e. trade, draft pick, free agent signing, etc) to be made. Over the life of an organization, the payoff would be massive. Accordingly, much of the organization is designed solely for the purpose of collecting and generating valid information, ensuring the proper flow of that information up to the proper decision maker, and enabling that decision maker to access it in a usable format. Further, establishing that type of structure is what generates long-term, sustainable success, as it doesn't rely on individuals, but rather on the collective power of the organization.

It remains to be seen to what degree the Reds utilize such a data-driven system, but the competition is certainly hitting the ground running. Make no mistake, competition is coming to the NL Central. Ultimately, competition leads to one of two outcomes: 1) it drives an organization to improve efficiency and effectiveness in order to succeed or 2) it destroys organizations that fail to adapt and evolve to meet the challenge. It remains to be seen which path the Reds will follow, but the competition has arrived on the shores of the NL Central and it's here to stay. The Reds need to be ready to meet the challenge. And, they need to start now.

1 comment:

  1. Here's a prime example of organizations built on an effective structure vs. individual brilliance:

    1. New England Patriots vs. Indianapolis Colts

    If you compare what happened to the Patriots when Tom Brady blew out his knee with what happened to the Colts when Peyton Manning went down with a neck injury, you can see a clear difference between the two organizations.

    When Brady went down, the Patriots simply plugged in Matt Cassel and proceeded to go 11-5 and win the AFC East. Cassel looked so good that the Chiefs were convinced to make a big investment in him as their quarterback of the future. Of course, the Cassel didn't get to bring the Patriot structure with him to K.C. and has gone down in flames as a result.

    On the flip side, when Manning went down the Colts turned the reins over to Curtis Painter, Dan Orlovsky, and Kerry Collins with disastrous results, going 2-14 and landing the first overall pick in the following draft. Coincidentally, the 1st overall pick landed them the likely individual brilliance of Andrew Luck.

    The obvious difference was that the Patriots had an effective structure. A structure built by Bill Belichick and others, which included effective and unified front office and coaching operations. A structure that allowed for success despite the loss of the best quarterback in football. The Patriot structure effectively elevated Matt Cassel into an above average quarterback and allowed the organization continued success. Since leaving that structure, Cassel has proven that he is far from above average.

    As for the Colts, they relied almost entirely on the individual brilliance of Peyton Manning for their success. Many long believed that Manning acted as his own offensive coordinator and covered up the deficiencies of the Colt coaching staff with his play. There really was no effective organizational structure, which became painfully clear when Peyton went down with injury. They had no one in the wings to provide that level of individual brilliance to carry the organization. And, they paid a heavy price for it.

    While you can build success on individual brilliance or organizational structure, there is inherently more risk and less longevity to success built on individual brilliance.