In Search Of: Sabean-metrics

The Giants recently promoted Bobby Evans to VP of Baseball Operations and Jeremy Shelley to Senior Director of Baseball Operations/Pro Scouting. Henry Schulman has a good account of this here, and Joan Ryan has a nice personal interest story on Shelley here.

Part of the Plan

Clearly, this is part of Neukom's strategy this off-season, to dispel the notion that the Giants are stuck in the dinosaur age in terms of the use of sabermetrics. He mentioned it in the first Clubhouse meeting with season ticket holders, refering to a "mathematician" who had been doing this for the Giants for 15 years now, and given that Shelley has been with the Giants for 15 years and what Ryan wrote about him, he must have been the guy Neukom was refering to; however, Ryan noted that Shelley was a Finance major at Santa Clara University.

And it is not like the Giants didn't leak some of that usage previously. I recall references to stats from Felipe and Sabean 3-5 years ago that were sabermetric principles, like OBP. The main unknown was how much did they use sabermetrics, were they tyros or very experienced. Based on what I got from Schulman's post, he has some depth.

For instance, Schulman noted that the Giants quoted how Renteria defense was good going to his left. I haven't checked this, but this clearly looks like a reference to Baseball Musing's Probabilistic Model of Range, which is based on the most advanced statistical methodology for studying defense, UZR, which Mitchel Lichtman created and Baseball Musing's David Pinto continued on-line, when Lichtman worked for the Cardinals and couldn't do it publicly anymore.

OK, it does appear to be PMR: http://www.baseballmusings.com/cgi-bin/DisplayCharts.py?PlayerID=1178&fpos=6&year=2008 If you look at his dotted line in the graph, which is the difference between his predicted and actual outs, he is much better going to his left (i.e. towards 2B) than he is going to his right. He is also below average in terms of balls that goes right where the shortstop spot is (look at the peak, the green predicted line is above his red actual). He is also bad with liners. And that jibes with his low 95.24 ratio for 2008.

Here is his historical ratios:

2006: 101.12 (but only 92.01 for turning DP)
2007: 98.87
2008: 95.24

Here are the UZR:

2006: 2.0
2007: -1.7
2008: 0.9

They don't quite match up in 2008, but was pretty good for the previous two years.

So Shelley seems to be relatively knowledgeable, PMR is a bit advanced, you have to seek it out plus then need to understand all that.

Schulman noted that at an early partners meeting, once Neukom had took over the Giants, he had Shelley come to the boardroom and explain all the analysis that he does. He also noted that Neukom is "keen on alternative statistics."

Giants Sabermetrician

I know that some Sabean-Naysayers would say that's an oxymoron, so I thought that would be a good title. Joan Ryan notes that his closeted status up to now have been to keep the Giants secret weapon on the down-low. I've seen his name in the media book for years and all they ever list him as was head of their information systems, though he was named Director of Baseball Operations previously before this promotion.

Ryan describes what he does:

In short, Shelley decodes baseball statistics. He finds meaning in numbers. He excavates databases like an archaeologist at a dig - except Shelley uses his unearthed artifacts to piece together a picture of the future instead of the past.

He projects how players are likely to perform over a season, or over the course of a contract. He and his team figure out before each season, for example, how many runs the Giants are likely to score and how many they are likely to give up. They go through each guy in the lineup. They add and subtract. They move decimal points around. They substitute this player for that one and recalculate everything. They use arcane formulas they have developed over the years and that they share strictly on a need-to-know basis, and, really, almost no one outside baseball operations needs to know.

"I'm a little uncomfortable saying anything more than that,'' Shelley tells me. "What I
can say is that analysis in baseball has gotten so sophisticated and it's changing all the time. I read, read, read - every baseball web site, every publication, everything.

"And every day, it seems, I see some new stat or method of analysis out there. We're always asking ourselves Does this make sense? Does this help us?''

Ryan's account is also nice because he's living the dream, and his start with the Giants shows that he was in the right place at the right time, much like his predecessor, Ralph Nelson, who rose from being the team's statistician to nearly GM (he eventually helped put together the Arizona Diamondbacks; one of my friends in high school, Jim Nelson, wrote him to see if there was a way he could get in, but just got a nice perfunctory letter in return).

Go Giants!