Data analysis: From the baseball field to the plant floor

April 6, 2015
Thomas Wilk explores lessons from baseball on getting grounded with big data.

If you are trying to master all the data sources that are available to monitor plant assets and strengthen your maintenance and reliability programs, you’re not alone.

Try to guess the source of this quote: “The amount of data available to make your decision is growing significantly, and human capabilities are not growing significantly. The importance of assisting the human decision-maker with decision aid is important in any field, and especially in a field that you can argue is a big data field.”

The answer is Sig Mejdal, Director of Decision Sciences for the Houston Astros, who was hired in late 2011 with GM Jeff Luhnow to integrate predictive analytics into the Astros’ baseball operations, and help turn around a team that had gone 56-106 only six years after appearing in the World Series.

Mejdal and Luhnow acted fast, bringing on a medical risk manager and analyst, a mathematical modeler, and a senior technical architect to identify the KPIs that would drive better player and trade evaluations. This effort resulted in a private, built-from-scratch online database they called “Ground Control” that delivers instant access to a wide variety of player statistics and video, as well as enabling communications with other front offices around baseball.

Their work produced tangible results quickly: in 2013, the Astros farm system was ranked in the top tier of ESPN’s baseball farm system rankings after being near the bottom the previous year, and in 2014 they led the rankings list. One specific success story was helping then-struggling pitcher Collin McHugh improve by introducing more curve balls into his pitch routine, after observing that his curveballs rotated 33% more than that of the average pitcher, helping them move more in flight.

“We identified him as someone whose surface statistics might not indicate his true value,” says David Stearns, the team’s 29-year-old assistant general manager.

However, the Astros also learned firsthand about the risks of pulling big data together in a centralized location: “Ground Control” was breached in 2014, and documents stolen from the database were made public, including confidential conversations the Astros had with other clubs.

In a press conference held to address the breach, Luhnow said, “We’ve moved on from the age of pencil and paper to a computer age where information needs to be accessed quickly and logically and a lot of it is in digital form. We had security in place, and when you’re talking about criminals we just never know if we have enough. I think we were prepared as we could have been.”

The experience of the Astros captures both the high promise and the risk associated with big data. I hope this month’s cover story on cloud-based maintenance and reliability also sheds some light on similar factors to consider when evaluating the benefits that these technologies may offer your programs and practices.

Sponsored Recommendations

Limitations of MERV Ratings for Dust Collector Filters

Feb. 23, 2024
It can be complicated and confusing to select the safest and most efficient dust collector filters for your facility. For the HVAC industry, MERV ratings are king. But MERV ratings...

The Importance of Air-To-Cloth Ratio when Selecting Dust Collector Filters

Feb. 23, 2024
Selecting the right filter cartridges for your application can be complicated. There are a lot of things to evaluate and consider...like air-to-cloth ratio. When your filters ...

ASHRAE Standard 199 for Evaluating Dust Collection Systems

Feb. 23, 2024
This standard ensures dust collection systems are tested under real-world conditions, measuring a dust collector's emissions, pressure drop, and compressed air usage. Learn why...

Dust Collector Explosion Protection

Feb. 23, 2024
Combustible dust explosions are a serious risk, and an unprotected dust collection system can be a main cause. Learn what NFPA-compliant explosion protection you need to keep ...