CommentsDEEGAN ON LA-The convergence of statistics and baseball dating back even to Branch Rickey’s time nearly a century ago has, in recent years, been exemplified by the Oakland A’s former general manager Billy Beane, who started a revolution in “America’s favorite sport” with an analytical system that became known as “Moneyball.”
Ever since then, the decisions to use sinkers, sliders, breaking balls, changeups or fast balls, how to bat at what pitch, how to use baserunners offensively, when to use defensive fielding “shifts” and other arcane baseball moves have relied on stats that are parsed and analyzed to suggest the best possible results, and are often sent by the experts down to the dugout in real time for the next play. Algorithms have joined the teams as the “tenth man,” sharing space with fans, the historic additional player.
Technically, this sort of empirical baseball analysis is called “sabermetrics”, and includes among its activities the number-crunching research of “in-game activity to answer specific questions.”
Is it possible for stats to be used to help mitigate the homeless problem? This is the thesis of the Los Angeles Roundtable, a former county research department, now an independent research organization that is proposing to use two datasets to predict if someone is on the “bubble” of becoming homeless.
It recently published a report identifying two vulnerable bubbles: unemployed workers and youth leaving the foster care system. They say their predictive screening tool can identify the 1% of unemployed workers most likely to become persistently homeless with 81% accuracy, and 72% accuracy for youth leaving the foster care system.
Of the two categories, the youth are the most critical because of their tenderness and short life experience. Some of the accelerating factors for youth becoming homeless include that they are often solitary, not part of a family unit, that they may have already experienced homelessness, have been in the foster care system, or have been incarcerated. In 2012, California Assembly Bill 12 extended foster care services from 18 to 21 years of age. This has improved, but not eliminated, the problem of youth homelessness.
The other category, the unemployed, includes workers who lose their jobs, specifically low-wage, male African Americans, a group the study identifies as “high risk.” Their employment barriers include substance abuse and involvement in the criminal justice system. One-third are working while homeless before losing their jobs. Some have physical or mental disabilities. Job training, education and housing are offered as solutions that will ease this problem.
County officials are declining to use the report’s recommendations to utilize County resources to offset homelessness among these two populations because, it alleges, the Roundtable misused County records in preparing it.
Stats can make skeptics of us: who has not been fooled by believing a poll or a research report? Mark Twain, in less sophisticated data-mining times, liked to say, "There are three kinds of lies: lies, damned lies and statistics.” Yet, in addition to Moneyball, economists use “prediction markets” to forecast future events, gamblers learn the odds that will help them to win, and card players understand the predictability of where the cards may be. Probability is in every facet of life, starting with the morning report of the weather predicted for the day.
How data is created, manipulated and then interpreted is always a challenge. Having a county agency reject data-driven recommendations, for whatever reason, is a blow no matter how well-intentioned the project is.
It’s now up to the County and the Roundtable, in the best tradition of baseball, to slug it out and do what they can to create a home run for these two homeless subsets.
(Tim Deegan is a civic activist whose DEEGAN ON LA weekly column about city planning, new urbanism, the environment, and the homeless appear in CityWatch. Tim can be reached at [email protected].) Edited for CityWatch by Linda Abrams.
