How Things Have Changed
If there was one consistent theme in all of Bill Bavasi’s decision making, it was reliance on track record. He acquired players who had a track record and trusted that they’d eventually just play to the level of their past history. He hired managers who had track records of winning, and they hired coaches who had track records of winning. If you came to any of the USSM events we had with Bavasi, you heard him use the words track record quite a bit.
Well, if rumors are true, Don Wakamatsu’s bench coach is going to be the guy who was in charge of the A’s hitting last year, and his pitching coach is going to be the guy who was second in command for the Rangers pitching.
The A’s had the worst offense in baseball last year. We’re hiring their hitting coach.
The Rangers had the worst run prevention in baseball last year. We’re hiring their bullpen coach.
We really have no idea how these hires will work out, but one thing is for certain – the Mariners certainly aren’t hiring because of track record anymore. And that, in and of itself, is progress.
Results based analysis is outdated. Welcome to the 21st century, Seattle Mariners.
But it sure is a scary first step!
I’d still rather see Rick Peterson come in, since he has a track record that I think does mean something; but from some of the things I’ve read recently, I can understand Wakamatsu not wanting to have to handle such a dominant and established pitching coach — friends or not, that’s the sort of situation that could go bad in all sorts of ways.
I remember when we made Paul Molitor our hitting coach. Those were the days.
The Rangers led all of baseball in %FB last year (by about 3%). I’m worried this philosophy might carry over with the pitching coach, because we all know what happens when Felix throws a lot of fastballs to start games.
[Piniella]
For starters, I would suggest you can modify that to “there isn’t a significant difference between most coaches/managers in terms of immediate player performance, and most coaches/managers do not account for as many wins (or losses) as most observers may believe.”
However, how an organization chooses its coaches and managers may give you insight as to how it intends on handling personnel decisions in general. In this case, I think Dave is suggesting the choices may indicate that the organization is less beholden to past results, which can be misleading, and more interested in method and underlying justifications, which many believe can be more predictive at the macro level.
It’s not dogging experience, it’s appreciating indications that the organization is showing signs of moving away from an unsuccessful method for recognizing who has some idea what they’re doing at their job. Or at the least, becoming more flexible in how they make that assessment.
Whether their approach proves insightful or not is going to be the question, right? There’s a corollary to new(er) doctors v. those with decades of experience. You might think you would always prefer the latter, but in some cases that may mean outdated methods of diagnosis and treatment. Recognizing those cases is the tricky part.
Well, to respond to DaveValleDrinkNight’s comment, the reason we all are getting so excited about these hires is the overall approach. Maybe the manager only makes a difference for wins 90-93 that get you to the postseason. Fine. Maybe the coaches only make a difference for wins 87-89. Fine. But when the general manager and the guys he hires for scouting, minor league coaching and managing, player development, etc. begin using 21st century analysis, then the team gets to wins 81-86 and all those other things suddenly become relevant.
Well, at worst it’s a new way to fail!
Hypothesis A: Managers and Coaches don’t really effect Player performance at a measureable level.
Wrong. The statement is actually that there’s very little variation from one to another, so they don’t matter all that much because they’re all basically the same.
Hypothesis B: Experience in Coaching/Managing has little to no effect on the outcome of a teams performance.
Again, no. We’re saying that experience isn’t a substitute for knowledge or ability. Leaning on experience as a proxy for knowledge is dumb.
No one’s saying that the position of manager/coach just doesn’t matter and that they could put a clown in there and see no effect.
DVDN — you’re caricaturing the discussion. W/R/T managers, the point that has been made repeatedly is that in-game tactics and lineup construction don’t matter that much, because everyone does more or less the same things and the advantages/disadvantages are relatively small in most cases. No one around here has ever argued that pitching coaches don’t matter; indeed, given the amount of hair-pulling we’ve done over Felix el Rey throwing 80% fastballs in the first three innings of every — blinking — start, and why our pitching coaches are either a) encouraging him to do so or b) failing to convince him not to do so, and the harm said coaches are doing to him either way, I would say that your supposed “Hypothesis A” is in no wise an accurate description.
Way to be redundant . . . oh, well. In any case, I’d argue that the first rule for managers and coaches is “Do no harm”; the ones who can meet that requirement are the ones worth keeping, the ones who can’t need to be fired, and the ones who can do measurably more than that are rare gems.
I was pretty scared about learning that these guys were coming in, but hopefully they are good coaches and were just victims of not having the talent in their previous jobs…fingers crossed…
So now we’ll have at least two guys who have gone through Billy Beane’s OBP brainwashing program. This is excellent news. Plus, Ty Van Burkleo…what a great name.
Looks to me like DaveValleDrinkNight and Dave are really just saying the same thing in two different ways. Paste their two statements together with the word “because†between them and instead of:
Hypothesis A: Managers and Coaches don’t really effect Player performance at a measureable level.
Wrong. The statement is actually that there’s very little variation from one to another, so they don’t matter all that much because they’re all basically the same.
Hypothesis B: Experience in Coaching/Managing has little to no effect on the outcome of a teams performance.
Again, no. We’re saying that experience isn’t a substitute for knowledge or ability. Leaning on experience as a proxy for knowledge is dumb.
We get:
Hypothesis A: Managers and Coaches don’t really effect Player performance at a measureable level because there’s very little variation from one to another, so they don’t matter all that much because they’re all basically the same.
Hypothesis B: Experience in Coaching/Managing has little to no effect on the outcome of a team’s performance because experience isn’t a substitute for knowledge or ability. Leaning on experience as a proxy for knowledge is dumb.
Even JO-JO the incontinent Circus boy would agree that these statements fit together reasonably well and appear to agree with each other.
What do managers and coaches do to distinguish themselves? I think it would be interesting to have a matrix of “what winning managers do well” and “what bad managers do”, assuming they’re not inverse of each other, and then have a look at how much we value the people who adhere to the “winning” template.
From what I understand, managers:
a) know what to delegate and to whom (they know their personnel- coaches and players- and their own abilities enough to accomplish B-D)
b) they instill a sense of group cohesion, confidence, and motivation for players of different ages, ethnic and economic backgrounds;
c) they establish an expectation of professionalism [effort, repeatable display of basic skills…though i realize this “respect the game and play it right” is more a systemic product of our culture at large, and all developmental aspects baseball culture specifically. Compare and contrast Japanese and American approach to the game.]
d) finally, they manage lineups and in-game changes based on statistical performance, SABR-style, player fitness, chutzpah- things typically associated with baseball managers.
I think it’s interesting that we want our managers to be competent in an unusual combination of ways: they should be psychologists, analysts of people and statistics, and teachers. Is there a predictable route to “quality” team management, and if not, which of these (or other) skills matter more?
My last question is: does Zduriencik feel he knows what he’s getting by hiring somebody without experience? Where does Z fall in the spectrum of GM involvement in team management and philosophy? In other words, how much autonomy does he want Wakamatsu to have?
Look, I’m on board with you guys about the hiring of Z and the use of more Sabermatic principles in obtaining and developing players.
I think we all agree that this is the area the M’s are at their most deficient.
I also think that player development and obtaining the right players through free agency/trades needs to be heavily influenced by statistical analysis.
My point is that, Manager and Coaching positions do matter. Even if only as a means to smooth over egos and keep the Team moving forward.
I just don’t see cause to celebrate two coaches being hired from Clubs that are slightly less terrible than we are.
You do realize that your second statement, in absolutely no way, follows or is connected from the first, right?
You have to be able to elucidate how a coach can help a team win and show how these coaches differ. Don’t think you’re doing that at all.
I think you missed the point of Dave’s post.
We hired guys with recent histories of failure. This is great news, because it means that GMZ isn’t hiring coaches based solely on their recent history of success like Bavasi was.
The way Bavasi did it was bad. Really bad. That GMZ’s way is different would be wonderful news, and these hirings are strong evidence of that.
My point is that, Manager and Coaching positions do matter.
Okay, let me try this again.
Say you have 30 grocery stores within a mile of your house. You want to buy a turkey, and you want a really good turkey. All 30 grocery stores sell turkeys, and in fact, 25 of the 30 get their turkeys from the exact same supplier, while the other 5 use specialty suppliers of varying quality. Of course, where they’re getting their turkeys from isn’t public knowledge and they aren’t about to tell you.
So, now, you have to make a decision on which grocery store to buy a turkey from. How do you decide?
You pick one and hope it’s good, because 83% of them are all exactly the same, and you don’t really have any way of knowing ahead of time whether the other 17% are going to be great or terrible.
Our stance is “just buy a turkey and hope for the best.”
You think our stance is “who needs a turkey? Spam will do just as well.”
You need to understand our point before you can disagree with it.
So if we hired say….Leo Mazzone as the pitching coach would your first reaction be “Wow, why did we hire him? The Orioles pitching was just as bad as the Rangers when he coached!”
Determining people’s coaching quality based on past successes or failures is basically Result-Based analysis. What you’re really looking for is what philosophies they’re bringing to the table.
mmmmmmmmmmmmmmmmmmmmmmm…turkey…thanks for getting me ready for tomorrow Dave…
At first glance, this logic seems to suggest that we shouldn’t care what process the team uses to hire coaches, because you just grab anyone and cross your fingers. This appears to be what some posters here are arguing.
A more nuanced view is that a superior process might enable you to (going back to the turkey example) to identify grocery stores using low-quality suppliers. Then, by selecting from the remaining stores, you can be confident you’ll be getting either a good or excellent turkey.
And to augment the example, someone using purely results-based analysis would say “I’ll buy a turkey from Store A since I bought one from them 10 years ago and it was great!” An intuitively appealing idea, but one which could easily lead to sub-optimal results (eg. if supplier has changed, or you were just lucky to get a good turkey in a bad batch).
I believe Dave’s initial thought is spot on. Doing things differently is encouraging, mainly for what it says about Zduriencik and his autonomy within the organization.
Beyond that, I think these are the most pertinent comments:
Can anyone even imagine a metric that suggests the win value for a manager or coach? That would be really helpful, but I think instead what we’ve got are not much more than anecdotal evidence, personal recommendation and the ever-popular ‘gut feel’.
However, having said that, I can think of a few examples like Duncan, Mazzone and (I know I’m dating myself here) Ray Miller who seem to have generated results from pitchers which those same players failed to achieve in other organizations.
Not a guarantee, but something apparently more than random chance.
So, that leads to the second comment: if there are pitching coaches who achieve better results, have they done that through a ‘one size fits all’ philosophy (e.g., ‘establish fast ball’)…or do they tailor their advice to the individual skills (or lack thereof) of specific pitchers?
As for hitters, I’m struggling to think of one who seems to have had the same consistent effect for any team.
Our stance is “just buy a turkey and hope for the best.â€
I’d elaborate that slightly as “buy a turkey with no obvious defects,” like missing a drumstick or the fact that it’s been sitting out and is half-thawed. Which is why you’d see outrage about managers who show complete incompetence at, say, bullpen management, even though it’s difficult to identify a difference in quality otherwise.
Our stance is “just buy a turkey and hope for the best.â€
Ah, but then you can go to Culinary Communion, not very far from Safeco Field in Beacon Hill, and get some of their “house-cured, incredibly delicious bacon… made from 100% organically-raised Berkshire heirloom pork” to drape across your turkey while you roast it in a low oven – thus adding immeasurably to the flavor.
So… that’ll work for a turkey. I’m not sure how this applies to managers though.
It works the same for managers, but replace bacon with superior talent.
I’m not sure how this applies to managers though.
Well, from what I understand about teams like the A’s, it sounds rather like making sure your manager and coaches (the turkey) are supplemented by a consistent, statistically sound organizational philosophy (the bacon) to produce better results overall. Mmmm, sabermetrics and bacon.
baaaaaacon.
Baker tosses in some bloggy bits & pieces about the potential new coaching staffs this morning
Managers are probably better when wrapped in bacon, just like everything else (and say that as someone who doesn’t eat pork). They’d certainly smell better. And since smell supposedly has an effect on athletic performance, a bacon-scented manager sitting in the dugout all day might make the team play better. Or it might just make them eat more, and given the progression of Betancourt and the apparently already-sizable appetites of Felix and Carlos, that’s not such a good idea. And peppermint may be better anyway.
Wait, what were we talking about?
wonder if we could add Paula Deen as a special consultant…
Hmmm, genocidal dictators may be an exception to that rule.
This thread is starting to make me hungry.
It’s also important to make sure you’re buying a turkey, and not a turducken.
I’ll let you all puzzle out for yourselves what I mean by that.
It’s also important to make sure you’re buying a turkey, and not a turducken.
I don’t know – since one of the things Wakamatsu said is that despite not having a sabermetric background, he took the A’s job to pick up some of that – or scouting guy Zduriencik starting a stats-oriented research department – you could argue that these Mariners are actively looking for turduckens.
The Mariners finally turning away from results based analysis is the best thing to happen to this organization in years.
Even if all of these new hires end up being failures, its still a lot better than looking at only “proven veterans”.
Lets hope that this philosophy applies to player acquisition as well.
You know, I happen to dislike that term, “results based.” It’s basically just another term for empiricism, which is something we WANT to have.
What the Mariners did was STUPID results-based analysis, where they fixated on spurious correlations, which were not supported by OTHER results and evidence. It was knee-jerk analysis, seizing upon the biggest and most visible feature, just because it seemed easier to understand.
That’s true. What we denigrate as results based is really a complaint about a misuse of correlation.
We want to know how many line drives Adrian Beltre did hit, not how many line drives he can hit. So we are still measuring results. But we’re not measuring how many hits he got, because that isn’t actually related to his quality of play.
but layers are good– like an onion.
may I just say (1000 calories later) Burgermaster makes a great bacon cheeseburger
may I just say (1000 calories later) Burgermaster makes a great bacon cheeseburger
Oh, man. I miss Burgermaster.
Mmmmm…..bacon…..
If “Results based analysis is outdated”, where does statistical data that drives sabermetrics come from?
Does sabermetrics include such things as playing injured, contract issues, agent issues, marital issues, clubhouse issues…all things that can affect our day, just like their day?
I have read so much about this missing part of the analysis, however it seems to have its own qualifiers based on what one wants to gather….is this not statistics?
Just trying to understand..
To quote Warren Buffet:
If you give this coaching staff (or any coaching staff) poor players they are going to have poor results. It doesn’t matter how good Riggleman was as a manager, Bryan LaHair wasn’t going to be a .300/.400/.500 first baseman.
Hiring based on a good track record alone is a bad idea, but hiring someone with a lousy track record because you are comfortable with them is not an automatic upgrade. Nepotism isn’t something new and I don’t think it would be considered part of running a successful business in the 21st century.
Maybe these guys will be great hires, but I see no reason to have any reaction to this move (positive or negative).
Isn’t sabermetrics nothing more than glorified result based analysis? Or, put another way, result based analysis gone wild?
No, because sabermetrics seeks to use results to analyze process. Properly, the difference here is that “results-based analysis” merely draws conclusions from results, whereas a sabermetric approach attempts to analyze results.
“Analyze” meaning sabremetrics tries to reduce causes their component parts and tries to consider and rule out, if possible, alternative explanations. (In other words, looks deeper and looks truer).
The end of results-based analysis is good; we are no longer hunting for ponies in the pile of dung. But that’s as much as we do know. The great unknown is why the choices were in fact made.
I don’t think you can put up a metric for managers or coaches, because they don’t perform on the field. And their role is to fit the team they have, not to make the team fit them.
An example I’ve posted about before is George Karl. In his first couple years with the Sonics, he was a tremendous coach, because he is very good at working with players without a lot of ego and getting them all to play together and play defense. He is absolutely awful (witness his attempts at coaching Allen Iverson) at coaching a team with egomaniac superstars. Whereas Phil Jackson is good at the latter and not so good at the former.
Lou Piniella may have been the right manager to take over the Cubs after they fired Dusty Baker and signed Soriano; he was not the right manager to take the Rays where Joe Maddon took them. Joe Torre sucked with the Mets and Cardinals and won four titles in New York. There is no statistic for “times ran interference with the front office” to show what his actual impact might have been.
Management is all about managing individuals. The key to baseball management, which is quite different from football or basketball coaching, is that they play everyday, and a manager has to be able to motivate a diverse group of men to prepare to play that grueling schedule. Different groups of men take different kinds of motivation, and we’re just not going to be able to see it in some statistic.
That being said, the fact that the M’s have a lack of talent and some untradeable contracts doesn’t point to one kind of manager, pitching coach, bench coach, hitting coach or anything else. We’re playing by ear here, and frankly the only thing that will matter is the results.
[cough]Gary Payton[cough]
as far as I can tell, George does well with a team until his own ego rises and gets in the way
How much of that is ‘right manager’, how much ‘manager gets team at right time’, and how much is dependent on uncontrollable outside factors, such as how the other teams in a division play?
Was Torre’s success with the Yankees because he finally had a team with talent, or because he took what he learned from his failures (or both?) Was Maddon the right guy for Tampa because the players were finally ready?
In 2003, the fans & writers in Chicago hailed Baker for ‘getting the Cubs back to respectability’, apparently because they had made the playoffs.
The previous 5 years (which were apparently so horrible and embarrassing) included an appearance in the NLDS, and a 2001 season under Baylor where the Cubs finished with the same record as Baker’s 2003 club.
In 2001 that left them 3rd (5 games out) in their division … and helped get Baylor fired halfway into the next year. In 2003 it gave them the NL Central title — and if they’d been in any other division in baseball, would have been, at best, a 2nd place finish and not close to taking the Wild Card.