04/19/24
October 17, 2007
A Scientific Hitter in the Computer Age: David Ortiz of the Red Sox does not look like a computer nerd, but he is.
Cool stuff. Love the picture of him and Manny looking like they could eat Lugo for breakfast.
posted by qbert72 at 04:59 PM on October 17, 2007
I always enjoy these articles, because- as something of a computer geek- I know I'd be like Schilling and Ortiz, using the latest technology to get every edge I could. Quite frankly, I'm amazed there is even one player who doesn't do this. You're professionals, for pete's sake! :) Two thoughts: 1) The article graphic has a mistake: it says "Power is down, averages are up", but this is incorrect. Ortiz has improved his HR totals and his SLG percentage every single full season he's played, and his OPS as well in almost every year. The real exception is that this year his "power" is down. However, without the benefit of Ramirez, and playing with injuries... he still posted a .620 slugging percentage, second only to last year's .636 in his career. As Theo Epstein noted, his performance this year is underrated, as he was only one point behind A-Rod for the MLB lead in OPS (that's one more single or walk over the course of the year). He was third in SLG overall, and first in OBP. The "power is down" comes from hitting 35 HR instead of 54- but that 35 was 7th in the Majors this year and 3rd in the AL, while his 52 doubles was only 2 behind Ordonez for the major-league-lead. The point being that he had a fantastic year, and hasn't really lost any power while boosting his average and on-base percentage. If he's 100% healthy next year... wow, he's going to put up some gaudy numbers. 2) Why don't more teams do this for their players? And what are the major league rules about technology, and when/where you can use it? I.e., coaches and managers still seem to use paper and pen when looking at the numbers in-game, or having their spiral binders of situational stats... but why couldn't a relatively small investment put a team in a position to access the power of computing resources to get them this information, and more, at the touch of a finger? They could: a) Store video information of every pitch of every game in the season from every player in the league, easily called up in sortable format. Say you want to see every two-strike pitch the pitcher has thrown this game, or how he's pitched you in the past, and see the percentages on what he likes to go to for his out pitch when you are at 1-2, 2-2, and 3-2. The hitters can before the game and also during the game see how a pitcher behaves, look for tipoffs or tendencies, or just sit dead-red on the highest percentage pitch in a count situation to improve their chances. The pitchers could similarly watch how the plate is being called from that behind-the-umpire camera view, or look at how hitters have been working him as a pitcher- what they're flailing at, and what they're jumping out of their shoes to hit. b) Track actions and use data mining algorithms to discover tendencies of players- both on their team and the opposing team- to date, as well as in-game. Deep analysis could occur in off-hours, producing useful reports on tendencies and trends, and how a pitcher has been beaten in the past and what worked, and what didn't. c) Have this info available in the dugout at a moment's notice. Why shouldn't every player who's waiting to go on deck not be able to watch a quick summary of how he's faired in his at-bats against the current pitcher, and to say watch video of all the instances of a particular pitch in a row. That would help the hitter know how the breaking stuff is breaking, etc. I imagine the manager and his coaches hunkered over a laptop as the game progresses, and hitters stopping by a video terminal as they wait to go to the on-deck circle, refreshing what has happened in past at-bats. See, if these tactics collectively made every player into a Rain Man- meets- Tony Gwynn thinker with little effort from them, and that resulted in just 1 out of every 33 outs made by your players turning into a hit or a walk instead... that would be BIG AND HUGE!!! It would literally turn a decent .278 hitter into a dangerous .300 hitter, and overall the benefit to the team in runs scored due to on-base percentage and slugging percentage increases, would be immense (the Yankees score 968 runs this year on a team BA of .290). The first team to fully utilize these strategies would start putting up 1000-run seasons and romping to the playoffs with ease. For example, the .275 hitting Mets finished at 88-74, with a RS/RA of 804/750 (their expected Pythagorean record was 87-75). What if they'd hit in the .290's, with a team OPS more like the Yankees? At 900 runs, they would have finished at 96-66, and god forbid should they have scored the same 968 runs the Yankees did by virtue of upping their team OPS by around .054, from .775 to .829? They would have won 101 games this year. The numbers are fuzzy, but I find it hard to believe that having that kind of real-time data and guidance to your hitters could turn just one at-bat a month into a hit or a walk. And that seemingly tiny difference, across all your players for the whole season... could be the difference between missing the playoffs by one game, and having the best record in the majors. So... is there some rule about using technology in the dugout? And if not... what the hell is wrong with these teams?
posted by hincandenza at 05:10 PM on October 17, 2007
Excellent link. Though provoking comments too. Wow its SpoFi in October! *ahem* Hal: Clubs do use technology in much of the ways you describe. Here is an ESPN article from last season that not only describes how the Rockies started using video iPods to review film on the go, it also details why a few players, managers, and front office types don't. If you want a test case as to why teams generally don't use technology, read every LA Times piece on former Dodgers GM Paul DePodesta tenure. Pay special attention to the columns by Bill Plaschke, you would think the man's body was occupied by the spirit of Joe Morgan.
posted by lilnemo at 05:42 PM on October 17, 2007
Store video information of every pitch of every game in the season from every player in the league, easily called up in sortable format. About ten years ago I got the opportunity to see the Chicago Bulls IT operation behind the scenes. They wouldn't show me everything, but they told me that had a video operation that allowed them to pull up something like "every shot Jordan missed in the 4th quarter from outside 10 feet with a taller player guarding him." I don't think they used that during the game, but perhaps at halftime. Of course, basketball is much more contextual than pitcher-battle duels, but it's awesome nonetheless. They viewed their video operation as an important weapon and were very careful not show the actual technology.
posted by drumdance at 06:23 PM on October 17, 2007
Last night, I saw the Sox hit the first back-to-back-to-back homers in ALCS history (that's what the announcer said). Manny hit the second of those three. So this article came at a most opportune time. Having said that ... Go Indians! Also, I wanna know: Mac or PC?
posted by worldcup2002 at 01:09 AM on October 18, 2007
OH, I meant David, not Manny (altho Manny was one of the back-to-back-to-backers). Anyways, that'll show me to comment on a baseball post!
posted by worldcup2002 at 01:43 AM on October 18, 2007
lilnemo: Clubs do use technology in much of the ways you describe. Here is an ESPN article from last season that not only describes how the Rockies started using video iPods to review film on the go, it also details why a few players, managers, and front office types don't. If you want a test case as to why teams generally don't use technology, read every LA Times piece on former Dodgers GM Paul DePodesta tenure. Pay special attention to the columns by Bill Plaschke, you would think the man's body was occupied by the spirit of Joe Morgan.Cool link, lilnemo. And yeah, there's no good reason not to use the technology; every edge counts, and paying some savvy computer folk- oh, people like me for example- to build you the computing power and data storage costs less than a utility infielder making the major league minimum. But just as Joe Morgan is a complete moron and a tool, there are still plenty of people in the top rungs of baseball's infrastructure who think "OBP" is a fancy pants techno-term that no real man would use. Given how- and drumdance offers evidence to support this- other sports have realized the power of video and data analysis for years, it's even crazier that as of 2006, the idea of having all the video of key pitches/plays/at-bats was "novel" to so many players in MLB.
posted by hincandenza at 02:56 AM on October 18, 2007
But just as Joe Morgan is a complete moron and a tool, there are still plenty of people in the top rungs of baseball's infrastructure who think "OBP" is a fancy pants techno-term that no real man would use. Really? I don't think this is too pervasive anymore. I mean, On Base Percentage is hardly cutting edge. Maybe they're uncomfortable with Win Shares, ERA+ or the like, but certainly not OBP.
posted by WeedyMcSmokey at 10:13 AM on October 18, 2007
Weedy- yes, *now* they are. But that's only because progress is happening. Before, it was stats like OBP; now it's things like VORP, etc. The point is that the Morgan types, and his front-office compadres, are resistent to these "new" ideas, until they are well-established. And in doing so, they miss a golden opportunity to be cutting edge and successful. I may have misspoke in saying OBP was a good example, but the point is other sabremetric measures are as ignored as OBP once was in favor of HR, RBI, and average. Billy Beane was successful for so long because other teams weren't doing the sabremetric work his organization was doing. Now that they likely are, Oakland doesn't have the same success. Extensive data mining and video analysis could be the next thing that a smaller market cutting edge team uses to gain an edge against the bigger payrolls. Maybe that team is Colorado...
posted by hincandenza at 04:06 PM on October 18, 2007
Billy Beane was successful for so long because other teams weren't doing the sabremetric work his organization was doing. Lew Wolff spoke last week to a breakfast group of which I'm on the board. Having Bean and the A's soccer-fanatic staff backing up John Doyle as the new Earthquakes get ready to start play in MLS next season should be different.
posted by billsaysthis at 07:32 PM on October 18, 2007
As Theo Epstein noted, his performance this year is underrated, as he was only one point behind A-Rod for the MLB lead in OPS (that's one more single or walk over the course of the year). And that's pretty amazing. One point. He had an amazing season, and as you said, without a mashing manny to protect him most of the year. As far as the home runs, I think it goes back to the knee problem and not being able to get into his normal hitting crouch. Standing more upright would seem to equal more line drives. I wish ortiz played enough games at 1st to see exactly how much the video work adds to his stats.
posted by justgary at 02:39 AM on October 24, 2007
I misread that as "A Scientific Hitler"...