March 26, 2012
Prospectus Hit and Run
Rattling SABRs in the Desert, Part II
Up until my appearance on Friday's “Clubhouse Confidential” panel, my experience at the SABR Analytics Conference might have been largely interchangeable with that of many an attendee. I had seen a good selection of the presentations but hardly a comprehensive one while renewing old friendships and building a few new ones, both around the conference proper and elsewhere in the greater Phoenix area. Basically, I blended into the woodwork. After the panel, not so much.
For the panel, I joined fellow “Clubhouse” consultants Vince Gennaro (SABR president, author of Diamond Dollars: The Economics of Winning in Baseball, and organizer of the Analytics conference), Rob Neyer (lead baseball writer at SB Nation, formerly of ESPN), and Dave Cameron (editor-in-chief at FanGraphs, blogger at USS Mariner, and alumnus of Baseball Prospectus). Whereas many of the first two days' panels had been focused on specific areas of data collection, ours, as moderated by Vince, was a general baseball discussion not unlike the Q&As the BP authors participate in during the annual bookstore tour. We kicked around a variety of topics, offering what I hope was a diversity of viewpoints. Sample topics:
Ours was a playful discussion, and the four of us exchanged our share of jibes along the way before taking questions from the audience. Audio for the panel will (hopefully) be up at some point in the near future here.
When the conference was initially announced, there was hope that Clubhouse host Brian Kenny would be a part of it, but alas, his schedule did not permit. Kenny did not let the occasion go unnoted on his show, however; they ran one segment devoted to Bill James, John Thorn, Pete Palmer, and the history of sabermetrics and another with an interview of Vince. The first one featured a great quote from Palmer and Thorn's The Hidden Game of Baseball that is worth sharing:
Baseball statistics are not the instrument of vivisection, taking the life out of the game in order to examine it; rather, statistics are themselves the vital part of baseball, the only tangible and imperishable remains of contests played yesterday or a hundred years ago. Baseball may be loved without statistics, but it cannot be understood without them.
As the conference spilled into the Hilton Phoenix East/Mesa lobby following the “Clubhouse” panel, I soon found myself summoned to the bar to join Thorn for a beer, an offer I simply couldn't refuse. I had never met MLB's official historian, but I was thoroughly entertained by his panel earlier in the afternoon, and had been a fan of his work for years (sheepishly, I admitted I still owed him a review of Baseball in the Garden of Eden: a Secret History of the Early Game, which I started last spring but was forced to put aside amid my own March madness), and simply getting the chance to talk baseball with him was a thrill. Soon we were joined by Gary Gillette, and for some reason, our lively discussion steered toward the myths and facts surrounding the Black Sox scandal, as well as my own work for the forthcoming Extra Innings on steroids, the home run boom, and evidence that the baseball itself changed. I admitted I was somewhat nervous as to the reception of my chapter. Gillette comforted me by saying, "You wouldn't be the first person to wander into the pasture of steroids and step into a cow patty," which cracked us all up even if it didn't quite put my stomach at ease.
Eventually, we were joined by Prospectus colleague Geoff Young, as well as Hanging Sliders' Wendy Thurm, and the conversation continued until the dinner bell rang. As the learned scholars peeled off to tend to their own plans, Wendy, Geoff and I, joined by the Score's Dustin Parkes (briefly a BP contributor) and his girlfriend Miranda, headed out for authentic Mexican food at a place called Los Dos Molinos—a name that couldn't help but draw a few laughs—where we loaded up on sangria, and some of the most outstanding green and red salsas I can recall. Aside from bemoaning the weaknesses of NL West offenses (Wendy is a Giants fan, Geoff a Padres fan, and I'm a Dodgers fan), the conversation veered away from baseball and into beer, road trips, Canada (Dustin's from Toronto), significant others, and other sports. Dustin's own "Top Ten Things I Learned At SABR Analytics" column is definitely worth a read even if it did have embarrassingly nice things to say about the rest of our dinner crew.
Fast-forward to Saturday morning. Though I once again failed to make the day's first panel at 8:45 a.m.—"Scouting and Analytics," with the Cubs' Joe Bohringer, the Padres' A.J. Hinch, and Corrine Vitolo of SmartKage (read Corey Brock's write-up here)—I did luck into an impromptu breakfast with Thorn when we bumped into each other at the elevator, and we resumed our discussion from the night before. He left me with two take-home points for my writing: first, to retain the esoteric influences one brings in beyond baseball, and second—an outgrowth of my reaction to a piece he wrote last year for Bleacher Report—to not lose sight of the story amid the numbers. They’re both worthy goals for this writer, and generally applicable to anyone who tends to wade deeply into the numbers.
We adjourned in time to catch "The Fielding Bible III: An Evaluation of the Williams Shift," a presentation by John Dewan and Ben Jedlovec. The owner of Baseball Info Solutions, Dewan has done much to advance the discussion of fielding metrics in recent years with his organization's systematic review and evaluation of every defensive play. In the opening panel of the conference, Dewan reiterated a statement he made in his new book's introduction: while we might be able to measure 80-90 percent of offense, and 75 percent of pitching, we're perhaps only in the 60-70 percent range with regards to fielding—but that's up from perhaps 10 percent a decade ago. One could quibble with the percentages he's attached, but there's little question we've still got further to go with regards to defense than with other areas.
Harkening back to a 1946 game when Indians manager Lou Boudreau ordered all four of his infielders to the right side of second base to combat Ted Williams (who ended up walking), Dewan's presentation focused on infield shifts where three infielders move to one side of second base in an effort to stack the deck against an extreme pull hitter. With his team of stringers reviewing video of every game, he now tracks the number of such shifts each team employs, and the results of those shifts.
According to his data, teams employed shifts around 3,800 times in 37,000 plate appearances, roughly one percent of the time. The Rays led the majors with 437 shifts in 2010-2011, about three times the normal rate. In 2010, they were fifth in the majors with 41 Defensive Runs Saved, while last year they led the majors with 85 DRS, perhaps not a coincidence given their propensity for the shift; on the other hand, the number-two team in terms of shifts during that span, the Indians, did so 278 times and still ended up with −14 Defensive Runs Saved.
The NL leader in shifts last year was the Brewers, with 170 under new manager Ron Roenicke, up from 22 under Ken Macha the season before, a strategy which helped their infield defense (Prince Fielder, Rickie Weeks, Yuniesky Betancourt, and Casey McGehee, all ranked among the worst at their positions in 2010) improve by 56 runs relative to the season before. Oddly, both Rays manager Joe Maddon and the Brewers' Roenicke are protégés of the Angels' Mike Scioscia, yet the Angels aren't a team that employs many shifts.
Of the hitters who faced shifts most often, David Ortiz (486 times) and Ryan Howard (461) led the majors during that span, followed by Carlos Pena, Adam Dunn, Prince Fielder, Jim Thome, Adrian Gonzalez, Brian McCann, and Jack Cust—not coincidentally, all lefties, and generally slow-footed ones at that. That group saw its batting average against the shift on grounders and short liners decrease by 30 points relative to when there was no shift on, .208 to .238. Only two of those players saw their averages increase against the shift, Pena and McCann; not coincidentally, those two were the ones who bunted most often against the shift, with Pena doing so for a hit 11 times, and McCann doing so four times. The other eight hitters saw their batting averages against the shift dip from .245 to .194, a 51-point drop.
Dewan presented an avalanche of data in his presentation—most of it reproduced from the book—showing breakdowns of how often pitches were pulled based upon pitch type and location. Space doesn't permit me to walk through all of his findings here, but his take-home point was that the shift generally works from the defense's standpoint, and that it should be employed more often against extreme pull hitters of either hand, not only most of the aforementioned lefties (and some others, such as Alex Gordon) but also righties such as Dan Uggla, against whom there were just 12 shifts in the two-year period. His system identified 48 lefties and 56 righties against whom the shift should be deployed, though as he noted, frequent targets Ortiz and Fielder are actually not among them, since the rates at which they pull grounders to the right side of the infield have fallen below the 80 percent break-even point. Meanwhile, his findings suggest that hitters facing the shift frequently would do well to drop down an occasional bunt; not only will they collect hits, but they'll also reduce the frequency of shifts against them.
Through Dewan's publicist, I had arranged to interview him at some point following his presentation, but instead of what I envisioned to be a potentially dry one-on-one session, he sat down at our table for lunch and answered questions from Geoff, Wendy, Craig Minami, and myself as fast as we could come up with them—not only on the shift presentation but on various other aspects of his book, including catcher defense (BIS measures blocked pitches now, for example). While I'm sure that some of his findings in the new book will fall under scrutiny of those who spend more time analyzing defensive data—our own Colin Wyers, for one—I found the discussion fascinating and can't wait to dig further into the book.
Between Dewan's presentation and our lunch sit-down, Fox Sports' Ken Rosenthal interviewed Indians president Mark Shapiro for around half an hour. The Indians have become one of the game's more progressive organizations as far as sabermetrics goes; a few years ago, they hired BP's own Keith Woolner (whom I met during the weekend) to be their Director of Baseball Analytics, and under him in some capacity or another are fellow BP alums Jason Paré and Russell Carleton. Shapiro noted that when John Hart was the team's general manager, the computer in his office was never on ("I'll turn it on when I can touch the screen," Hart had joked); obviously, things have come a long way.
Continuing the ongoing topic of defensive metrics, Shapiro said, "We look at conventional statistics that are available to everybody, and we've got our own proprietary statistics. Where we feel we're at as far as objective measurement of defense today is somewhat around the equivalent of using batting average for offense. It has some value, certainly very limited value, but we factor it in. It's an area where we feel like subjective scouting information is still most important and we combine it all and make evaluations based on that."
Later, during the audience Q&A session (right around the 40-minute mark of the audio clip), I asked Shapiro about the makeup of the Indians' rotation—a topic that had come up for discussion during the “Clubhouse” panel and again as I chatted with Keith. "Yesterday on the Clubhouse Confidential panel we were kicking around the idea of groundball rates among pitchers and the way that that has an influence upon team-building," I explained. "You've got a fairly ground ball-heavy staff, and especially the rotation… How much of that is by design, and how much of that is happenstance in terms of injuries and trades and attrition?"
"I'd like to say it's all by design," said Shapiro, "But it was more Derek Lowe at $4 million represented the best value for a starting pitcher and those innings… It has not been by design. If it was by design, we'd have better infield defense," he conceded as the audience chuckled, surprised at his candor. "We certainly hope to address our infield defense, and Casey Kotchman was a big part of that… his range is not exceptional but the number of errors he's made in his career is less than we made as a team last year at first base. His ability to pick balls out of the dirt—that doesn't get objectively measured—and pick up guys in that way is exceptional. So I think him, and if you go with [Jack] Hannahan—and we'll have to weigh that decision between Hannahan and [Lonnie] Chisenhall at third base—all of a sudden you've got two among the best in the game at first and third. And hopefully that makes up for a very young guy at second [Jason Kipnis] and a guy that we know is not going to be good, hopefully can be average at short [Asdrubal Cabrera]. He can hit."
It's not often you hear a club executive talk so frankly about his All-Star shortstop's defensive shortcomings; rarely has Yankees general manager Brian Cashman discussed those of Derek Jeter in such terms—but then, the Yankee captain is a sacred cow with five rings and 3,000 hits, and Cashman knows that such remarks would fuel a feeding frenzy. Shapiro was willing to level with a more sophisticated audience, though to be fair, Cabrera's defensive numbers are all over the map depending upon which system you use:
After lunch, the schedule offered one more panel, "Digital World Meets Baseball," with Jeff Bennett of ESPN, Rob Shaw of Bloomberg Sports, and Ryan Zander of SportVision. Much of the discussion, moderated by ESPN's Michele Steele, had to do with the way advanced metrics are used in broadcasting, particularly PITCHf/x type visualization systems and the use of run expectancy/win expectancy-type data connected to a given base-out situation. The point that I took from it is that the deployment of such advances, while increasingly feasible from a technology standpoint, and growing in popularity, still has much to do with the level of comfort the on-air talent has with calling upon such systems. As the Joe Morgans are swept aside and younger announcers take the booth, we'll see more of them.
That was it for the conference proper, aside from the finals of the student case competition, in which teams of undergraduates and graduates representing various schools were presented with a hypothetical situation regarding the Nationals, who are several games back at the All-Star break, around .500 but with some mitigating circumstances—they've preserved Stephen Strasburg's limited innings for later in the season, and their record in the just-completed interleague play was drastically below that of their intraleague record; they have the option to white-flag it with a trade Edwin Jackson for a package of prospects, to go for broke by acquiring Brandon Phillips for a package of players, or… I'm not sure what else, as I never did get the specifics nailed down. Each team had to show their reasoning as to what they would do, and a panel of judges chose finalists, then threw them another curveball related to a Scott Boras-driven trade scenario. The winners were four students from the University of Chicago's Booth School of Business.
I sat that out, but before peeling off from the conference, I had a brief but wonderful conversation with Richard Cramer, who along with Thorn and Gillette had been part of the "Retrospective Look at Baseball Analysis" panel the day before. Back in the early ‘80s, Cramer, who designed pharmaceuticals via computer as his day job, founded STATS Inc. STATS' baby was the Edge 1.000 analytical system, which could show spray charts for different batters, different pitchers, different pitches, and different counts (see Alan Schwarz's fine book, The Numbers Game, for more on the software and on Cramer's career). The A's, the first team that bought the system, used it for broadcasts but couldn't persuade manager Billy Martin that he needed it, while the White Sox, who also bought it, found an open-minded young manager named La Russa who ate it up with a spoon. Cramer told me that thirty-some years later, he still has a working version of Edge running on an Apple II, and said he enjoyed my work on JAWS and the Hall of Fame, and on “Clubhouse Confidential.” I have to admit, I got a little verklempt.
The rest of the weekend would be… manlier. That evening would feature our Beer, BBQ & Baseball bash at the Surprise residence of our own Jason Parks, with fellow BPers Kevin Goldstein, Derek Carty, Joe Hamrahi, Daniel Rathman, and Geoff Young—provider of several wonderful beers from San Diego—on hand, as well as ESPN's Jorge Arangure, MLB's Cory Schwartz, SABR's Gary Gillette, Lone Star Dugout's Jason Cole, and DIPS creator Voros McCracken, who has turned his attention to European soccer after being chewed up and spit out by the Red Sox. The food, beer, and conversation were great, but I chose not to partake in the FIFA 12 tournament, while some of the participants in said tournament could scarcely be bothered to take their hands off their joysticks.
I had hoped to get to one more ballgame, but Sunday's forecast for rain led to the pre-emptive cancellation of some games, and the mid-game scuttling of others, including the Dodgers-Giants tilt that I had hoped to attend. Thus the highlight of my afternoon was a trip across the boulevard to In-N-Out Burger, (where I may have gone back for seconds in all of the confusion), finally joined Kevin, Eric Stephen and Dave Young of True Blue LA, Phil Lacovara of AZSnakepit and his fiancée, Jenna, and one other Snakepit-related couple who had their infant son with them. The food was delicious and the conversation lively—particularly with Kevin fielding questions on Dodger and Diamondback prospects—but soon I reached the glazed last-night-of-winter-meetings feeling that I know too well. It was time to go home.
In all, the SABR Analytics Conference was an unqualified success that brought the baseball industry, researchers, media and fans together for one great weekend in Arizona. It's not hard to envision this becoming an annual staple on the baseball calendar, with participants building the conference into their Cactus League schedules. On a personal level, I was honored to be part of the inaugural event and to get so many opportunities to talk baseball with people I hold in such high esteem. I was touched by how many people remarked upon my work on “Clubhouse,” underscoring my own belief that I'm not up there simply representing myself, but a grand tradition of intellectual inquiry of which I'm proud to be a part. A sincere thank you Vince Gennaro, Joe Hamrahi, and everybody else who helped make my weekend in Arizona so utterly amazing.
Images courtesy of SABR.org.