Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Baseball Primer Newsblog > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Tuesday, May 20, 2014

Arizona Diamondbacks’ CBO Tony La Russa clarifies his stance on Sabermetrics

Just for mendacity. Targets only the gray area.

La Russa, himself, has been critical of Sabermetrics in the past, but acknowledges their place in today’s game.

“My opinion is that it’s a valuable tool, but mostly a tool to help you identify talent and then prepare the talent,” La Russa told Burns and Gambo Monday on Arizona Sports 98.7 FM. “I think the biggest problem I see is there are teams that have gone way overboard and they are really interfering with the way the managers and coaches conduct strategy during the game by running the analytics and forcing them into it.

“I would definitely say I think that’s a bad idea, so I think my attitude, and I’ve talked to Derrick (Hall) and Ken (Kendrick) about it, is there’s an important place for it, but if you don’t stop and recognize the tradition and history of how you play and how you get coached and managed, and the contributions you can make, a computer analysis is never going to be able to replace that.”

La Russa went on to recall George Kissell, who spent nearly 70 years in the Cardinals organization as a player, manager, coach, instructor and scout.

“If we allow the metrics to do the exaggerated claims that they want to do, guys like Kissell are a relic and irrelevant and that’s a mistake,” he said. “What you do is you take that basic premise about the way human beings compete and how they need to be coached and managed, especially during a game as conditions change, and you don’t mess with it.

“You help prepare using the analytics, but you don’t let them exaggerate. In fact, the teams that do are making a big mistake.”

Repoz Posted: May 20, 2014 at 07:04 AM | 28 comment(s) Login to Bookmark
  Tags: d-back

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. Shooty Survived the Shutdown of '14! Posted: May 20, 2014 at 07:56 AM (#4709981)
In fact, the teams that do are making a big mistake.

Which teams are these? I want names! (Otherwise it's impossible to know of TLR is making a good point or is just full of shit.) Really, though, I've never heard of a team that forces strategy on a manager based on computer generated probabilities. Sounds like Tony is engaging in a straw man to me.
   2. salvomania Posted: May 20, 2014 at 08:38 AM (#4709994)
LaRussa suggests it's a zero-sum game, and if you use analytics then you have to get rid of the George Kissells.

It should be pretty clear that better, more-detailed information about "what is happening" allows people to make better-informed decisions, with a caveat that they need to be able to interpret and understand the information.
   3. Hal Chase School of Professionalism Posted: May 20, 2014 at 08:48 AM (#4709999)
I've never heard of a team that forces strategy on a manager based on computer generated probabilities. Sounds like Tony is engaging in a straw man to me.


Sounds like Tony believes that Michael Lewis' version of Art Howe's role is the standard manager on a saber-friendly team.

Billy Beane should have never wrote that book.
   4. Harveys Wallbangers Posted: May 20, 2014 at 08:56 AM (#4710005)
i will just remind the younger posters that once upon a time larussa would proudly have his photo taken with a computer with the related story of how data was helping him better position his team to win.

over time tony assumed the persona of curmudgeon and luddite.

   5. Into the Void Posted: May 20, 2014 at 08:57 AM (#4710006)
The A's and Rays (and on the opposite spectrum, the Astros and Rockies) are generally seen as the most sabermetric focused teams right? So Larussa is suggesting they would have had more success had they been less so?
   6. Jose Can Still Seabiscuit Posted: May 20, 2014 at 09:00 AM (#4710010)
Really, though, I've never heard of a team that forces strategy on a manager based on computer generated probabilities. Sounds like Tony is engaging in a straw man to me.


I think most teams at the very least provide their manager with information based on probabilities. I think you just need to read the nightly omni-chatters to see that managers don't slavishly stick to that information though. The Red Sox almost certainly give Farrell this sort of information and he still makes some pretty perplexing moves.

The bigger issue is that a guy like Kissell isn't being taken out of the equation. Teams make decisions on non-scientific factors all the time. Getting back to the Sox they won the World Series last year playing Jonny Gomes in left field at a time when the analytical community was screaming to get Nava in there. It's not an either/or situation. Good managers should be able to blend statistical info and eyeball tests.

The other issue is do we really want our team to be making decisions on what an 85 year old guy tells them to do. I'm sure Kissell or Johnny Pesky or Yogi Berra has forgotten more baseball than I'll ever know but the world moves forward. You'd think someone as innovative as LaRussa was would understand that. I'm willing to bet you could find some articles from 1988 of old school managers saying LaRussa's bullpen management was bunk.
   7. RoyalsRetro (AG#1F) Posted: May 20, 2014 at 09:23 AM (#4710017)
i will just remind the younger posters that once upon a time larussa would proudly have his photo taken with a computer with the related story of how data was helping him better position his team to win.


Exactly. Read "Men at Work". TLR is hailed as an innovative genius with his index cards and binders full of stats. He even uses a computer!
   8. Shooty Survived the Shutdown of '14! Posted: May 20, 2014 at 09:37 AM (#4710025)
It is ironic now that the first openly sabrmetric friendly gm in the modern era--Sandy Alderson--hired LaRussa. At the time it was a great partnership, though. I suppose over the years LaRussa began to believe his own press.
   9. PreservedFish Posted: May 20, 2014 at 09:52 AM (#4710035)
Sounds like LaRussa might benefit from reading the proceedings of a certain roundtable discussion.
   10. I am the Can Posted: May 20, 2014 at 10:20 AM (#4710067)
“If you don’t stop and recognize the tradition and history of how you play and how you get coached and managed, and the contributions you can make, a computer analysis is never going to be able to replace that.”


Usually, when otherwise-intelligent people have a well-reasoned and heartfelt opinion on something, they can string together a few clauses in a way that is both grammatical and (reasonably) logical. I listened to the full interview and while LaRussa was very coherent and engaged during most of it, it really did sound during this segment like he was just spitting out words like "tradition," "history," and "strategy." Not sure what you ought to conclude from that, but my feeling was that this is Tony playing luddite and curmudgeon, as Harvey says, rather than actually having any real thoughts of substance on this topic.

I have plenty of respect for Tony LaRussa, but it really rubs me the wrong way when I get the sense that people are b-s-ing.
   11. SoSHially Unacceptable Posted: May 20, 2014 at 10:33 AM (#4710079)
over time tony assumed the persona of curmudgeon and luddite.


Maybe somebody in St. Loo inadvertently passed down a communique from MGL when the ol' stathead was working his magic there. By the fourth "biggest idiot ever," TLR may have had a change of heart about the whole numbers thing.

   12. Squash Posted: May 20, 2014 at 10:45 AM (#4710088)
i will just remind the younger posters that once upon a time larussa would proudly have his photo taken with a computer with the related story of how data was helping him better position his team to win.

And at that time was openly mocked for it in the media as being a young whippersnapper who relied too much on his charts and data.
   13. Harveys Wallbangers Posted: May 20, 2014 at 10:56 AM (#4710095)
squash

larussa was not mocked. larussa has always been held in high esteem. at no point did anyone in the media write anything of glowing reports of larussa and his tactics/approach/etc
   14. RMc's desperate, often sordid world Posted: May 20, 2014 at 11:03 AM (#4710100)
   15. What did Billy Ripken have against ElRoy Face? Posted: May 20, 2014 at 11:40 AM (#4710122)
Usually, when otherwise-intelligent people have a well-reasoned and heartfelt opinion on something, they can string together a few clauses in a way that is both grammatical and (reasonably) logical. I listened to the full interview and while LaRussa was very coherent and engaged during most of it, it really did sound during this segment like he was just spitting out words like "tradition," "history," and "strategy." Not sure what you ought to conclude from that, but my feeling was that this is Tony playing luddite and curmudgeon, as Harvey says, rather than actually having any real thoughts of substance on this topic.


Indeed. Doesn't it, once you take out all the word salad, basically come down to "analytics are great as long as teams don't actually try to, you know, put them into practice?"
   16. Squash Posted: May 20, 2014 at 01:08 PM (#4710195)
larussa was not mocked. larussa has always been held in high esteem. at no point did anyone in the media write anything of glowing reports of larussa and his tactics/approach/etc

Not true. When he was younger in his White Sox/A's years there were many comments about his overreliance on his charts and such. I guess it comes down to how you perceive the word "mocked" and how much guys were taken to task then in the media versus now. He wasn't hit over the head with a stick/run out of town like guys like DePodesta were, but there were definitely comments about how he should focus less on his charts and numbers (and more on what I don't think anyone defined - merely that he was too number-y).
   17. cardsfanboy Posted: May 20, 2014 at 01:21 PM (#4710204)

Maybe somebody in St. Loo inadvertently passed down a communique from MGL when the ol' stathead was working his magic there. By the fourth "biggest idiot ever," TLR may have had a change of heart about the whole numbers thing.


I'm fairly certain that has some to do with it. I do not think that TLR appreciated MGL and that he more or less forced(or made his input feel unwelcome) MGL out of St Louis organization. I think that is why MGL held a grudge so strong that he was able to write the truly idiotic article in which he accused the Cardinals of making the worst managerial decision ever, in keeping Carpenter in to pitch the ninth against the Phillies in the 1-0 playoff game.

But TLR does hold some very high bias against analytical analysis. During the baseball writers dinner, when they were giving out the gold glove awards, with the Rawlings representative present, he went out of his way to bag on their new method of choosing the gold glove. Saying that he understands it's 40% analytics and 60% the managers vote(or whatever the percentage is), and that they should make it 5% and 95%.
   18. cardsfanboy Posted: May 20, 2014 at 01:23 PM (#4710205)
Indeed. Doesn't it, once you take out all the word salad, basically come down to "analytics are great as long as teams don't actually try to, you know, put them into practice?"


I thought it was analytics are great at evaluating and getting talent but it should stop there, when it comes to developing the players and in the dugout decision process go with the old school methodology.
   19. Harveys Wallbangers Posted: May 20, 2014 at 01:50 PM (#4710217)
squash

if you are going to use the comments from hawk harrelson we will have to disagree

   20. Harveys Wallbangers Posted: May 20, 2014 at 01:50 PM (#4710218)
squash

if you are going to use the comments from hawk harrelson we will have to disagree

   21. What did Billy Ripken have against ElRoy Face? Posted: May 20, 2014 at 02:05 PM (#4710224)
if you are going to use the comments from hawk harrelson we will have to disagree


That's a pretty good rule for life right there.
   22. Walt Davis Posted: May 20, 2014 at 09:38 PM (#4710486)
Strawman aside, I don't see LaRussa saying anything other than "you can't let computers make the final decisions." If you want, you can infer "George Kissell can do a better job of figuring out which 5th round pick has what it takes and which one doesn't."

I've used this comp several times. Sabermetrics is (primarily) nothing but actuarial science. On average, young male drivers get into an accident T% more often. On average a 5th round pick returns value X. On average, the last 3 years' performance tells you what next year's performance will be like, +/- Y. On average, a 35-year-old will decline by Z%. On average, a steal attempt here is a good/bad idea. Pitch/fx and such things are offering the potential to move to a finer level than that so the future may be different (but, as you do so, samples become small and particularistic).

Teams have to make their decisions at a micro-level. Of the 100 guys they're choosing from, which one should be their 5th round pick. Sabermetrics isn't at a point to inform that in any reliable way, at best it gives you that list of 100 guys or maybe narrows it down to 75. Possibly George Kissell can't reliably tell you which to pick either but the Cards track record in development has been good for decades so I ain't saying he couldn't.

Should you attempt a steal? Should you bunt? That's a mix of a ton of factors -- the specific combination of score, inning, pitcher, catcher, fielders, batter, runner, next batter, who's in the pen, park, weather conditions, field conditions, positioning of the 3B and 1B. Again, maybe LaRussa can't really help you there either but he's got as good one of the better track records of success in history.

Sabermetrics was lucky in a sense. The pitcher-batter matchup is pretty close to a true experiment. One PA is pretty independent of the next. Over the course of a season (and especially 3), any given batter/pitcher will have faced a sample of pitchers/batters that is close to a "representative" sample. You've got to adjust for the park effects but otherwise the analysis is pretty simple. Even batting order doesn't have much of an effect. And the outcomes of pitcher-batter constitute, what, 90-95% of what's important to victory.

The rest of baseball though gets closer to the complexity of analyzing basketball or one of the other interdependent sports. And it's not that hard for a human batter/coach to make the decision -- look how far back Aramis Ramirez is playing. They don't know his Rfield but they know ARam is not a strong 3B, they know it's a tough pitcher on the mound, they know the batter is a decent bunter and the guy on base is a good runner and it ain't Trout in the on-deck circle.

At some point, with real-time field/fx data (or whatever that new system is called) and enough data on most of the individuals involved, sabermetrics may be able to say that when the 3B fielding range/distance from home ratio is less than X then bunt. That still may not help much given the 3B may have changed his position by the time the batter's brain chip tells him to bunt.

   23. shoewizard Posted: May 21, 2014 at 08:20 PM (#4711174)
Maybe somebody in St. Loo inadvertently passed down a communique from MGL when the ol' stathead was working his magic there. By the fourth "biggest idiot ever," TLR may have had a change of heart about the whole numbers thing.


I'm pretty sure this has A LOT to do with TLR's current stance. But while the rest of us are secretly looking at porn on our computers, TLR is probably secretly looking at Fangraphs.
   24. greenback calls it soccer Posted: May 22, 2014 at 12:30 AM (#4711292)
Joe Garagiola mocked the computer managers, a pretty clear dig at La Russa, at a time when Garagiola was somehow the most visible baseball commentator on the planet.

Earlier this week I heard a quote from Douglas Adams, which went something like this:

You will regard technology that existed by your 15th birthday as perfectly normal.
You will accept technology that was created between your 15th and 35th birthday, because it is new and exciting.
You will reject technology that was created after your 35th birthday as counter to nature.

The ages are a bit different for baseball field managers, but Tony La Russa got old. You can avoid this fate, if you are cursed with self-awareness, but considering how tight he wore his uniform into his sixties, I am guessing La Russa does not suffer from that malady. So note cards with batter-pitcher history are acceptable, even if they are computer-generated, but God help you if you look at a WPA chart to make an IBB decision.

Preservation of the roles of George Kissells is important to La Russa. This is very similar to a discussion of Old Time Baseball Men in Bissinger's book Three Nights in August, only the argument there was the MBAs gone wild couldn't love the game way the OTBM did. I see La Russa groping for the argument that Walt raised, but he simply isn't articulate enough to express it. It is a shame he didn't get a job in a bigger media market, for the Bobby Valentine-class hilarity that could ensue.
   25. villageidiom Posted: May 22, 2014 at 09:03 AM (#4711344)
Sounds like LaRussa might benefit from reading the proceedings of a certain roundtable discussion.
Did they ever do that? Perhaps someone can provide a link.
   26. cardsfanboy Posted: May 22, 2014 at 12:28 PM (#4711445)
God help you if you look at a WPA chart to make an IBB decision.


God shouldn't help you on that, common sense should help you on that, unless you have wpa chart made up based upon your team's makeup, it would make no sense to look at it for any decision making process. WPA (and run expectency..etc) is all based upon average players following the current batter. It's completely oblivious to the the handedness of the batter/pitcher matchup, the quality of the pitcher on the mound, the quality of the hitter up to bat, the defensive alignment etc.

Mind you, there is nothing wrong with more information, but TLR would manage circles around an MGL controlled manager.

   27. Ron J2 Posted: May 22, 2014 at 12:43 PM (#4711459)
#26 Well the thing is that you can look at a WPA chart for a very low offensive context (say the NL 1968) and one for a much higher offensive context (AL at the height of sillyball) and that'll give you a good idea of the impact of offensive context on any given tactical decision.

In general the choices are pretty stark.

Now I agree that a LaRussa managed team would be a heavy favorite over one managed by MGL, but color me doubtful that it'd be because of tactical decisions.

EDIT: The first article that discusses what we'd now call WPA is in the Hidden Game of Baseball and they touch on the extreme cases. As in, break even point sharply changes in the late inning of close games. You're less concerned with the chance of blowing a big inning and a single run is huge.
   28. cardsfanboy Posted: May 22, 2014 at 01:05 PM (#4711472)
Now I agree that a LaRussa managed team would be a heavy favorite over one managed by MGL, but color me doubtful that it'd be because of tactical decisions.


MGL controlled manager would immediately replace the pitcher, no matter how well he is pitching or the quality of the team he is facing, after the 27th batter. That alone makes an MGL controlled manager the biggest idiot to walk the planet.

In general the choices are pretty stark.


Not really, the choices that a manager is going to make which is up for debate aren't usually that much of a difference. And when you factor in the host of information that a real field manager has, vs a generic chart, as long as the manager isn't an idiot, I'll take his knowledge of the game going forward. Even Tango and MGL has talked about how you have to play against the odds in order to keep weapons at your disposal(if you never attempt a steal or sacrifice, the opposition knows it, and it improves their defensive setup and pitch selection)

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
TedBerg
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogOTP - July 2014: Republicans Lose To Democrats For Sixth Straight Year In Congressional Baseball Game
(3993 - 3:51am, Aug 01)
Last: Dr. Vaux

NewsblogAthletics, Twins Swap Tommy Milone, Sam Fuld
(23 - 3:45am, Aug 01)
Last: Dr. Vaux

NewsblogMarlins acquire Jarred Cosart from Astros in six-player deal
(7 - 3:16am, Aug 01)
Last: KT's Pot Arb

NewsblogOT: The Soccer Thread July, 2014
(557 - 3:04am, Aug 01)
Last: Swedish Chef

NewsblogA's Acquire Lester, Gomes For Cespedes
(128 - 3:00am, Aug 01)
Last: PreservedFish

NewsblogJULY 31 2014 OMNICHATTER/TRADE DEADLINE CHATTER
(363 - 2:09am, Aug 01)
Last: Crispix reaches boiling point with lackluster play

NewsblogTigers To Acquire David Price
(72 - 1:58am, Aug 01)
Last: if nature called, ladodger34 would listen

NewsblogCardinals Acquire John Lackey
(93 - 1:57am, Aug 01)
Last: bjhanke

NewsblogMinnesota Twins sign Kurt Suzuki to two-year contract extension
(7 - 1:19am, Aug 01)
Last: boteman is not here 'til October

NewsblogYankees land infielders Stephen Drew, Martin Prado at Deadline
(20 - 1:07am, Aug 01)
Last: shoewizard

NewsblogMariners acquire outfielder Chris Denorfia from Padres
(4 - 1:04am, Aug 01)
Last: PreservedFish

NewsblogOT: Monthly NBA Thread- July 2014
(1074 - 1:02am, Aug 01)
Last: Squash

NewsblogCliff Lee Re-Injures Elbow
(18 - 12:55am, Aug 01)
Last: Joyful Calculus Instructor

NewsblogCubs deal Emilio Bonifacio, James Russell to Braves
(10 - 12:38am, Aug 01)
Last: Jacob

NewsblogBrewers acquire outfielder Gerardo Parra from D-backs
(5 - 12:27am, Aug 01)
Last: KT's Pot Arb

Page rendered in 0.2674 seconds
52 querie(s) executed