Tag Archives: baseball

The Moneyball Approach to Cyber

It took a while for me to understand baseball: I didn’t get it until someone pointed out that I was watching the game when I should have been watching the season.

Much of the game’s strategy snapped into focus — and the differentiation between game-day action and long-term success illustrates key lessons that information security executives need to learn.

Love it or hate it, Moneyball is part of the game now. Moneyball and sabermetrics-applying sophisticated statistical analysis to baseball records-helps teams avoid overspending on showy all-arounders and focus instead on key metrics, however unusual, to build a successful team.

Information security should follow the same strategy. (And most chief information security officers (CISOs) probably feel more kinship with the cash-strapped Oakland Athletics, pioneers of Moneyball, than with the flush New York Yankees.) CISOs will see that, as in baseball, relying on a few stars to carry the team is a short-sighted and potentially costly plan.

In his 2014 Black Hat keynote, computer security analyst Dan Geer declared the end of the era of information security generalists. It can be hard to measure the contributions of specialists. We understand the easy metrics intuitively-the “batting averages” of information security. But it is the hard and subtle metrics that really teach us something new. Getting these metrics will require automation and thoughtful changes to existing sources of unstructured data: processes performed manually can’t keep pace with business needs.

Security & Privacy Weekly News Roundup: Stay informed of key patterns and trends

Alongside the outmoded concept of star all-arounders, we also should toss the concept of clutch players. Statistically, they don’t exist, and seeking them out in a technical organization is asking to be deceived; individual heroics are dramatic but not sustainable. An organization’s long-term success won’t be seen in the individual who burns the midnight oil to deploy the patch of the week, but in the one who quietly solves the problems around reliable, rolling deployments.

CISOs should also listen to the refrain of baseball commentators: “fundamentals.” A team that cannot execute basic, everyday maneuvers flawlessly is not prepared to get fancy. There’s no point in deploying a shiny intrusion-detection system or hiring an expensive, full-contact “red team” unless operations can convince you that every last default password has been changed.

Finally, we can take one more lesson from the game: Every so often, be sure to stand up and stretch.

Moneyball and the Art of Workers' Comp Medical Management

Recently, I watched “Moneyball,” the movie, for the third or fourth time. The story is compelling, as is the book by the same name that preceded it.1

“Moneyball” is based on the concept called Sabermetrics, defined as “the search for objective knowledge about baseball.” The central premise of “Moneyball” is that the collective wisdom of baseball insiders, including players, managers, coaches, and scouts over the past century, is subjective and flawed. The book argues that the Oakland Athletics general manager, Billy Beane, took advantage of analytic, evidenced-based measures of player performance to field a team that could compete successfully against far-richer teams in Major League Baseball. During the 2002 season, the Oakland A's won enough games to make the playoffs in spite of a meager salary budget and “inferior” players.

Even though the two industries are diametrically dissimilar, distinct parallels can be drawn between baseball and workers’ compensation medical management.

Similar Resistance to Analytics

One similarity is the resistance to adopting analytics as a knowledge tool. Baseball insiders and managers opposed Beane’s analytics, sometimes vehemently. Long-held beliefs among baseball insiders promoted measures of performance such as stolen bases and batting averages. Beane’s metrics debunked the old methods, revealing unrecognized strengths in lesser-known, more affordable players.

Similarly, workers’ compensation leaders have relied on traditional medical provider networks and personal preferences to select medical doctors. If doctors are in a network and offer a discount on medical services, all is good. Yet, industry research has shown that not all doctors are equal. Doctors and other medical providers who understand and acknowledge the nuances of workers’ compensation drive better outcomes. It’s a matter of finding those doctors.

Finding Best Performers

The purpose of “Moneyball” Sabermetrics is the same as workers’ compensation medical metrics—to find the best performers for the job. The way to do that in baseball is to analyze the data defining actual performance in terms of outcome—games won. In workers’ comp, the data must be scrutinized to find doctors who drive positive claim outcomes. In both cases, a variety of metrics are used to support the most effective decisions.

Performance Indicators

As in baseball, the goal in medical management is to apply objective information to decision-making using evidenced-based measures of performance. For both industries, cost is a factor. However, in workers’ compensation, the cost of medical care must be tempered by other factors:  What is the duration of medical treatment? What is the return-to-work rate associated with individual doctors? What providers are associated with litigated claims?

As in baseball, the list of indicators for performance analysis is long. However, the sources of data differ significantly.

The Data Challenge

In baseball, all the data necessary for analysis is neatly packaged. Statistics are gathered while the game is in progress. In workers’ comp, the data that informs medical management resides in disparate systems and must be gathered and integrated in a logical manner.

Essential data lives in bill review systems, claims adjudication systems and pharmacy (PBM) systems and can also be found in utilization review systems, peer review systems, and medical case management systems. The data must be integrated at the claim level to portray the most comprehensive historic and current status of the claim. Data derived from only one or two sources omits critical factors and can distort the actual status or outcome of the claim.

Once the data has been integrated around individual claims, meaningful analysis can begin. Indicators of performance can be analyzed with new conclusions drawn about the course of treatment and medical provider performance. Moreover, concurrently monitoring the updated claim data leads to appropriate and timely decisions.

Data Positioned as a Work-in-Progress Tool

In baseball, the data is used as a work-in-progress information tool. Decisions about the best use of players are made daily, sometimes hourly. Workers’ compensation medical management can do the same. Systems designed to monitor claim details and progress can alert the appropriate persons when events or conditions portend complexity and cost.

Industry Status

Analytics in baseball is not exclusive to the “Moneyball” Oakland Athletics. All of Major League Baseball now relies heavily on its use. Unfortunately, there are still only a few visionary Billy Beanes in workers’ compensation medical management. Yet, applying analytics for cost and quality control is simple and affordable and can be adopted quickly by all.

1Lewis. M. Moneyball: The Art of Winning an Unfair Game 2003. The film “Moneyball”, starring, Brad Pitt was released in 2011.