Archive for September, 2008

Online technologies will transform how we make offline decisions

The true long-lasting impact of the Internet will be in how we harness the power of the web to transform how we make decisions in the physical world.

Instead of just focusing on how to improve our online lives, the far greater potential is in combining online and offline interactions. The power of the web is how we can structure, measure, analyze and evaluate the massive amount of data we continually create to guide our decisions in real-time.

In the physical world, however, we invariably face a lengthy lag between when data is created and when it is used in a decision. We constantly make decisions using incomplete information because we have no choice; collecting more data is often too expensive, too difficult or simply not possible. Thus we create heuristics, stereotypes, rules of thumb and other methods for making decisions in environments we do not fully understand.

While collecting more data does not always lead to better decisions, technology is creating unparalleled opportunities for us to use larger amounts of up-to-date data to make quicker decisions. We are beginning to see web technologies address larger opportunities in navigating the physical world using dynamic, real-time, structured data in addition to our more typical use of the web to access static, dated data (e.g. restaurants, bars and driving directions).

For example, SFpark is an experiment in using web technology to help people find open parking spaces throughout the city using sensors embedded in the pavement to detect used and vacant parking spaces. From The Economist:

The SFpark project will begin early in 2009 with a new network of pavement sensors in 6,000 of San Francisco’s metered parking spaces and 11,500 of its off-street car parks and garages. These sensors will detect when a space is taken and relay that information to a central database. From there, information about vacant parking spots will pass to drivers in several ways. The most basic will be through a network of road signs that will indicate areas with parking places. Eventually, however, officials want to provide web and mobile phone services that display the availability of parking block by block on a colour-coded map, much like the traffic maps now offered by Google.

… Tod Dykstra, [Streetline Networks'] chief executive, hopes eventually to create networks that monitor other bits of a city’s infrastructure too, including traffic flows, street lamps and water mains.

While it may seem like a relatively small matter, searching for parking spaces imposes huge costs on the infrastructure, environment and productivity of the city and its residents.

In a world where it is an order of magnitude easier, cheaper and faster to collect and process data for making decisions, how do our “rules of thumb” and traditional heuristics change?

What Startups can learn from Billy Beane

Billy Beane, the General Manager (GM) of the Oakland Athletics, explaining how to change the team during the regular season, from “Moneyball” by Michael Lewis (page 193-194 if you’re interested):

  • “No matter how successful you are, change is always good. There can never be a status quo. When you have money you can’t afford long-term solutions, only short-term ones. You have to always be upgrading. Otherwise you’re f**cked.”
  • “The day you say you have to do something, you’re screwed. Because you are going to make a bad deal. You can always recover from the player you didn’t sign. You may never recover from the player you signed at the wrong price.”
  • “Know exactly what every player in baseball is worth to you. You can put a dollar figure on it.”
  • “Know exactly who you want and go after him.” (Never mind who they say they want to trade.)
  • “Every deal you do will be publicly scrutinized by subjective opinion. If I’m [IBM CEO] Lou Gerstner, I’m not worried that every personnel decision I make is going to wind up on the front page of the business section. Not everyone believes that they know everything about the personal computer. But everyone who ever picked up a bat thinks he knows baseball. To do this well, you have to ignore the newspapers.”

Many observers, especially those in baseball, completely misinterpreted Moneyball. Moneyball was about strategy, not tactics: constantly measuring and re-evaluating tactics and alternatives, not about determining and defining the “winning tactic”.

Beane built (and continues to build) the Oakland Athletics by building a team focusing on under-valued players. What is under-valued, however, changes over time. In the mid-1990s, OBP (on-base percentage) and OPS (on-base percentage + slugging percentage) were under-valued statistics compared to more traditional stats such as RBIs, batting average and other counting stats. Beane was able to identify measurements of player performance that were under-valued by other teams (by player trade value and salary) compared to the contributions the player would have to the team’s performance (in wins and losses). Beane found, monitored and acquired players that fit his model and built teams that perennially out-performed their total salary expenditures; he created winning teams that were able to regularly compete against teams spending multiples more in salaries on players and teams. *

However, times changed. As other GMs picked up on the results, they started copying his tactics and bidding for the same players.

And over time, Beane adjusted. The metrics of value in the marketplace changed and Beane re-evaluated his tactics and identified new ways to value players and construct teams. He changed how he valued defense, draft picks, salary flexibility and pre-arbitration players; he changed how he constructed bullpens and timed acquiring and trading away players; and he made many other decisions that demonstrated his ability to measure, estimate and value player and team performance. It may not work every single year since many factors are difficult to model (e.g. freak injuries, personal matters), but over the long run Beane has demonstrated the ability to change tactics within an overarching strategy.

Lessons for all…

Also: more on GigaOm: What Startups Can Learn From Billy “Moneyball” Beane.

* Of course, while the A’s out-performed in the regular season, they regularly under-performed in short playoff series post-season baseball. The averages of the long-season simply did not always work out over the small sample sizes of five or seven games. Beane’s infamous retort: “My sh** doesn’t work in the playoffs.”

MORE: Financial Models for Entrepreneurs