http://www.wired.com/playbook/2012/12/data-mining-minor-league-stats/        at what point do  we become ...de-humanized ???                           

When Scouting the Minor Leagues, It’s Data That’s the Clincher

• By Jeff Beckham
• 12.13.12

The Visalia Oaks, now the Visalia Rawhide, in the dugout. A pair of stats-loving baseball fanatics say they’ve figured out a better way of predicting whether minor-league players will do well in The Show. Photo: Jim Merithew/Wired

A college pitcher with a knack for numbers and his statistics-loving coach have found a way to mine baseball statistics that could help big-league scouts and managers more accurately assess minor-league prospects and bring better hitters to The Show.
Major-league teams analyze reams of data when building and managing squads, a numbers-driven endeavor that’s been part of the game since the Brooklyn Dodgers hired the sport’s first full-time statistician in the 1940s. But while much work has been done on properly valuing major-leaguers, little has been done with minor-league hitters.
Predicting how minor-league players will perform in the majors is tricky because stats that matter in the Rookie league become less important as players advance to AA and AAA teams and into the pros. Hitters that keep their strikeouts down are the ones to watch in the lower leagues, but as they climb the ranks strikeouts become less of an issue so long as they’re hitting for power and getting on base regularly.

More on Stats in Sports

Guy Stevens and Gabe Chandler wanted to determine how you might assess how a minor-league player might hit in the majors before seeing him face a big-league pitcher. The answers could help teams decide who needs more time in the minors, who should go pro and who should get released.
These guys aren’t random number crunchers. Stevens is a right-handed reliever for the Pomona-Pitzer Colleges Sagehens who plays for Israel’s national team. He’s pursuing a double major in math and economics with an emphasis on statistics and hopes to get an analytics job with a big-league club. Chandler is a statistics professor and baseball coach. They’ve presented their findings in the Journal of Quantitative Analysis in Sports. (Stevens also shares his insights at Dorm Room GM.)
In trying to find a way to use minor-league statistics effectively, the pair focused on the 1999-2002 seasons, figuring that was long enough ago to see whether the players had gone on to successful pro careers. They focused on position players rather than pitchers.
“Pitchers are notoriously volatile, prone to injury or sudden lack of success,” they write. “We wonder here if hitters, less injury prone than pitchers, may be more predictable and therefore safer investments.”
Their research found that, in evaluating minor-league talent, teams placed too much emphasis on whether the player was selected early or later in the draft. The conventional wisdom held that players chosen early were superior prospects, even if the stats didn’t always bear that out. Those early-round draft picks tended to get the benefit of the doubt when it came to promotions, while late-round picks, even those with good stats, often were not promoted or even released.
A better approach, the researchers said, is to make a more objective assessment using the stats that matter at various levels: the ratio of walks to strikeouts for Rookie league hitters and OPS (on-base percentage plus slugging) at the higher levels. Using a technique called a “classification tree,” they looked for correlations between major league success and minor-league statistics of on-base percentage, home runs, and runs batted in. They found that you cannot accurately judge hitters by their early minor-league performances. Still, there were some warning signs.
“We found that high strikeout rates for high draft picks, which are almost certainly high-school draftees facing much better pitching, does not bode well for their careers,” said Chandler. If a player is overwhelmed almost immediately in Rookie ball, their chance for advancing to the majors is very slim.
The methods of evaluating players also caught their eye while researching the paper. Some clubs, like the Oakland A’s, put a premium on a player’s on-base percentage and have found success even with a limited payroll. Teams like the Pittsburgh Pirates choose to give more at-bats to players who they believe will be successful, and have found considerably less success, despite having high draft picks.
For now, the pair want to expand their study with more data. Stevens knows teams will always seek an edge in player development, and hopes his research might provide it. He’s already using some of it when he takes to the mound.
“Baseball is a huge part of my life, and I can’t always tell whether my analytical experience is changing my on-field approach, or if my playing experience is motivating my analytical approach,” Stevens said. “It’s a bit of a chicken-and-egg problem in that sense.”