I am, up to a point, a fan of the “Moneyball” approach to scouting talent. It removes certain biases from the evaluation, which is always good. On a more basic level it opens our eyes to the fact that “undervalued talent” is out there and we need to change the way we define talent, in order to spot that talent.
However to transplant this to the world of youth employment there is a danger to lose the creativity that sabermetrics is trying to spot in sports.
Like human nature tends to make us do, if there is a test we train to pass the test, not always to master the skill. If we know what stats are important, we focus on those.
We at GapJumpers, and I am biased here, favour open ended problems to help companies spot great talent based on verifiable ability. The catch however is that it relies on data and on the eye of an expert.
We should measure, but always remember that the goal is impact, not just efficiency of hiring, for the way we measure impacts the way we educate.
Much like Sabermetrics and scouts together help the Red Sox win the world series (that and a big bankroll, because a pure Sabermetrics team like the Oakland A’s still has not won the Series) technology and human (expert) judging and training need to work together.