I can see how analytics help would be useful during game prep but he's absolutely right that they shouldn't be openly second-guessing the manager during a game. If they're going to make the in-game decisions then hire them as the manager. The analytics folks should report TO the manager, just like the coaches.

It could be seen as egotistical but I can see his point. Right or not, doing something for 20 or 30 years and then having 27 year olds who have never done it before come in and tell you how to do things better (based on percentages of only what can be tracked and ignoring what can't- in other words, ignoring the things you have observed having done that very thing for 20 or 30 years), is not something most people will react favorably to.

Part of using analytics intelligently is communicating intelligent analytics throughout the organization. If you only crunch numbers in little rooms near the GMs office then throw them at the manager every day you aren't going to get very far, even if the manager buys in. You need to develop analytic recommendations with the managers buyin and questioning, and answer his every question.

Then the players and coaches have to unlearn some of the strategies and tactics they've been taught in the minors, colleges, high school and even little league. You need to present good sound analytic reasoning to players as soon as they enter the organization at every level, from Rookie ball to MLB free agents so they can be more understanding and supportive of their managers use of them.

I love the image of an analytics guy going into Billy Martin's office to tell him how to manage. Hilarity (and violence) would ensue.

I can see how analytics help would be useful during game prep but he's absolutely right that they shouldn't be openly second-guessing the manager during a game.

I don't know. Perhaps a "Harry and Alex" should have been in the Cubs dugout during the 2016 World Series to second-guess Maddon over his planned usage of Aroldis Chapman?

Wow, Joe Maddon likes it when Joe Maddon is (seen as) the most crucial part of a team's success! Who'd've thought?

He uses Billy Martin as an example, and isn't Martin sort of infamous for grinding his pitchers' arms into a fine paste, especially when he felt his job was in danger? That's probably an extreme case, and I suppose it's possible at the time that his front office more or less approved of that, especially if they weren't on the hook for expensive long-term contracts and felt their development was good enough to replace whatever he might break pursuing short term goals. But he seems like the sort of guy baseball operations would probably have liked to put the brakes on, and the only way they could do it at the time was firing him when he was a problem and re-hiring him when he looked like he might have been useful.

Somewhere in between, it sounds like this won't be an issue if you've actually got a decent, coherent organization where everyone is pulling in the same direction. I understand Maddon's desire to just be left alone to do his job without having to worry about what's going on in other parts of the organization, but I don't know if the way teams are constructed now, with half-billion dollar contracts, a triple-A shuttle, more attention to injuries, and the like, makes that possible. It seems like the best thing for a team would be making sure that everybody in the organization is of the same mind, which would both mean the front office isn't meddling so much and, conversely, the manager can say that they hired him because he knows the details of this job better, so let him handle that.

I don't know if there's necessarily a spot for someone like Maddon in that situation, for better or worse. But it's probably going to be a bigger problem in an organization like the Angels where there doesn't seem to be that sort of coherence, with Moreno hiring Maddon and his baseball operations guys for different reasons and probably not worrying a lot about how those pieces are going to fit together.

I love the image of an analytics guy going into Billy Martin's office to tell him how to manage. Hilarity (and violence) would ensue.

That's a funny image. But wasn't he one of the most meddled-with managers of that era? Plus all the firings. He seems like a bad example of someone left alone to do his job.

Joe maddon telling 29 teams in baseball, he's no longer interested in managing their teams and putting all his hopes on Jerry Reinsdorf when TLR retires due to a heart murmur.

From Fire Joe Morgan in 2005 to Fire Joe Maddon in 2022. Who would've thunk it?

It's a shame that one of the many analytics guys bothering Maddon didn't convince him to bat Trout in front of Ohtani.

The podcast should be available for those who aren't Athletic subscribers. Starkville (Jayson Stark/Doug Glanville podcast) is one of the better baseball podcasts, with the interviews usually being very good. Effectively Wild is my other favorite, and in their case, I prefer their banter over the interviews.

There are a few (close enough to) true-isms in what he says and, without being in the dugout, we can't know the mix that was actually achieved:

1. It is good to have a final voice of authority, probably especially so in sports. Somebody has to be in the position of "here's what we do" without half the staff looking over at somebody else for approval or thinking "but the other guy said to do Y and I agree so that's what I'm going to do."

2. Sounds like Maddon, along with all of the rest of us, likes "information" when it confirms the decision they wanted/planned to make; doesn't like it when the "information" thinks it knows more than they do.

3. Everybody is much more comfortable with information that flows up to them to help them make a decision, especially if they have the authority to discard/ignore the info they don't want. "Information" that flows down is really orders. (The reaction to information that flows sideways is probably the most interesting and is maybe the bigger key to success.)

Maddon seems to think #1 was violated. To the extent #2 was an issue, that's likely on Maddon ... or at least he should have a good reason for ignoring the info he ignores.

For #3, to the extent that the tactical plan is laid out with the manager essentially required to follow it -- i.e. A gets the day off, B starts at 2B over C, Z starts at C over Y, use D for the first 72-75 pitches, then if in the lead EFG, otherwise HEF or HKL if the defict is more than 3, under no circumstance use M or N. Here's the updated data for the positioning algorithm, make sure you check it before every pitch because it can vary based on count. If you have any questions, just click on our 24/7 online help chat icon. -- then the manager no longer has a tactical role, no longer has a strategic role and their role is to clearly communicate decisions made above them, get "buy in" and deal with any complaints without involving upper managerment.

That's clearly not a role you need a Joe Maddon for and it's not a role that a Joe Maddon would enjoy. I struggled with the other side of that in the last few years of my career where my role was being limited to just providing the info and having no real input to the questions being asked or the interpretation of the results. I retired.

If the manager always has to ask someone else's permission to make a decision then why have him at all? Eliminate the middle-man and put the analytics guy in the manager's job.

Shildt got fired presumably for not wanting to be that rule of manager. Marmol clearly is running out lineups that are drawn out by the analytics team.

I don't know how much difference it all makes. There is a perfect way to construct a lineup but the marginal differences are so small, it's hard to believe it really matters.

Shildt had rigid bullpen roles. Oli is more mix and Match.

2022 Redbirds May be a game or two better but most of that is attributable to better players.4

Shildt had rigid bullpen roles. Oli is more mix and Match.

Rigid bullpen roles is idiotic in any and all forms of managing. The one thing that Whiney did well was when he had a bullpen by committee, and one of TLR's early weakness's was bullpen roles. Oli isn't just following the analytics, he's still doing his own thing, but where Shildt might let analytics figure into 25% of his decision, Oli is letting it figure into 50% of his decision.

I cannot imagine that there is any freaking analytic system out there that would have Bader batting behind Molina, nor Edman, even against rhp batting ninth. If there is, it needs to be overhauled.

The analytics folks should report TO the manager, just like the coaches.

I'd have doubts about this. At some level an organization has to guard itself against a manager prioritizing short-term victories over mid-range plans, especially with pitcher arms but also with development of younger, unproven players. To the degree that your GM is the person with that perspective, the stats people need to report to them.

And I don't think it's wrong for a GM to say hey now it's time for us to see if this younger guy can play.

A manager has some incentives to say let's burn the arms and let's run with the old guys that an organization has to counter.

There are of course two senses of "analytics" in this discussion. One would be stuff like "studies have shown that you shouldn't bat a scrappy low-OBP shortstop leadoff and then bunt him to second when he occasionally gets on base." The other would be more like what Walt's describing: "We have solved baseball and we demand that you take this supposedly optimal path through in-game strategy."

The latter involves fallacies, I think. In-game decisions involve too many variables and too much gamesmanship for there to be a calculable correct decision for every situation. So you appoint somebody as strategist. He may be bad at it, he may overestimate his genius, he may be conceited, but when he's good he probably adds value.

A manager has some incentives to say let's burn the arms and let's run with the old guys that an organization has to counter.

but enough about Joe Girardi.....

:)

The latter involves fallacies, I think. In-game decisions involve too many variables and too much gamesmanship for there to be a calculable correct decision for every situation. So you appoint somebody as strategist. He may be bad at it, he may overestimate his genius, he may be conceited, but when he's good he probably adds value.

This feels right to me. Also having your analysts be clear on when the margins between options are so tiny that it probably doesn't make a difference so may as well follow the manager's hunch.

That's a funny image. But wasn't he one of the most meddled-with managers of that era? Plus all the firings. He seems like a bad example of someone left alone to do his job.

I don't think Billy's firings had much if anything to do with ownership meddling with his managing. It was all about drunkeness and brawling. Beating up a marshmellow salesman, fighting with Reggie in the dugout, getting his arm broken by Ed Whitson, etc., etc.

The problem for me lays in the fact that any good analytics are going to be probabilistic. And most decision makers have no use for probabilistic information. That incidentally includes whoever's evaluating the manager and his use of the analytics.

Further, most of the time the information is going to be nuanced -- and people pretty much always hate nuance.

The simplest example I can give is about bunting. Going back to Pete Palmer's work people have been saying, the numbers say bunting is bad.

But that's not what they actually show when you really look. What they in fact show is that base for an out with a position player is nearly always a bad idea.

However if the batter is actually good at bunting and has a substantial chance of reaching (whether by error or beating it out -- so quality of defense also enters into things) then bunting is at least OK. And there are a handful of players where bunting will be a good idea almost regardless of game situation. And that doesn't make for an easily digested bit of info. Multiply by damned near anything.

All that to say that good analytics rarely uses words like demonstrates or proves, but weaker words like indicates and suggests.

And he chose another odd example. Earl Weaver famously did his own analytics. Not very complex ones but he absolutely knew how all of his players did in any given matchup. (which seems to be the level of info Maddon was getting)

Just pointing out the bunting issue, Mgl I believe (or maybe Tango...someone) pointed out that the threat of a possible bunt alone helps. If the other team knows you are never going to bunt, then in bunt situations they can play normal, and you are taking a tool out of your bag.

There are all types of variables involved in most decision making in baseball, a vast majority of them are still going to result in the "obvious" call. But there are still enough that can't be pre-predicted that you need a field general to make the call.

22. So earl was the master of knowing SSS data because mostly that's what everyone has. Albert Pujols trendlines are just coming into focus - but there aren't many that are.

My favorite "stat" is how a player hits against a certain team or how a pitcher throws against a certain team. It's usually over multiple seasons and is always talked about as if the info is gold. "Paul Dejong is a Mets killer." Is my favorite. I bet not one Mets pitcher that has given up a big hit to Dejong is still there.

I think there is a small subset of SSS data that can be useful but a lot of baseball comes down to how everyone is feeling on a certain day. Maddon is saying "just give me the three things that matter and then get out of my office". That's not an unreasonable ask

#16 ... that's what I was trying to get at with "sideways" information. The top-down model has problems, the bottom-up model has problems. In some ideal scientific sense, the information should be independent of hierarchy -- the data is what it is, the analysis and conclusions are as close as we know to do to "it is what it is." So the sideways approach of getting the nerds and the manager/coaches and the GM together and forming a collective conclusion may be ideal ... and virtually impossible to achieve but Albert's closing in on 700 and I wasn't expecting that to happen either.

If "value added" is still corp-speak for "value added" then the manager should be adding insight on the interactive human bits that the analytics can't really get at. I'm not sure that's Maddon's forte really but he seemed to haave his moments with the Cubs. But if the manager's input on that is no longer welcome, I don't know where else it comes from.

The problem for me lays in the fact that any good analytics are going to be probabilistic. And most decision makers have no use for probabilistic information. That incidentally includes whoever's evaluating the manager and his use of the analytics.

Further, most of the time the information is going to be nuanced -- and people pretty much always hate nuance.

This is frequently true and frequently a source of frustration in my career ... and not one I was particularly good at finding a solution for. What you tend to find is a strong desire to reduce all that complexity to a single number -- "why don't we just add these 5 things together?" -- or at most a handful of "key performance indicators" (followed 6 months later by "why don't we just add these 3 KPI together?" From that point forward, those are the only numbers anybody will want to see. It can also lead to some challenging discussions around the differences between causality and measurement although those might not arise much in this context.

#21 and #23 ... yes (was it in The Book?). But my conclusion from MGL/Tango's discussion was that there are so many factors involved that you might as well "go with your gut" or it was so obvious whether you should or shouldn't bunt that there wasn't much point. Things like "sure, with a good bunter who has speed but not much power and a poor defensive 3B is playing too far back and the game is close and a good contact hitter is on-deck and the pitcher doesn't miss many bats then it's a really good idea to bunt" -- no ####, if they aren't playing you for a bunt even though you've got a good bunter at the plate, bunting is a good idea. And the two times a season those factors come together doesn't give you a large enough real-game sample to know if the guy at the plate actually is a good bunter.

Which is back to Ron's point but requiring empathy with the decision-maker.

"This guy's a good bunter."
"What does that mean?"
"There's an 80% chance (+/- 20%) he'll bunt the next strike fair and only a 5% chance he'll offer at a pitch outside of the zone for a strike. Further, conditional on a fair bunt, he'll put it in a spot where the play is 'difficult' 70% of the time (+/- 20%)."
"So 70% of the time, he'll do the job ... "
"That's the conditional probability. So he's got a 80% chance of bunting the next strike fair then a 70% chance the play will be difficult so overall a 56% chance of a good bunt on the next pitch ... conditional on that pitch being a strike."
"OK so a 56% chance of a hit?"
"Well no, a 56% chance of a 'difficult' play for the defense. In this case a 'difficult' play means that an average 3B will get a force out 20% of the time, another 50% of the time the sac will be successful, 25% of the time it will be a hit and 5% of the time there will be a throwing error resulting in an extra base."
"So let me see ... 56% of the time difficult times 50% successful sac is 28% overall plus another 58% times 30% (or 17.4%) that the batter will reach base so about 45% chance this works as well or better than I hope."
"Yep ... against the 20% of strikes he doesn't get the bunt down (which includes 5% pop-out bunt) plus the extra 5% of balls he turns into strikes plus the 1% chance of a DP plus the 35% chance it's just a ball. Meanwhile if we let him swing away ... Of course any outcome that changes the count means we need to recalculate everything."
"Whew, anything else?"
"We need to take into account the defensive quality of the 1B and 3B."
"Oh yeah, who's at third?"
"Nolan Arenado."
"You coulda told me that 10 minutes ago and saved me the headache."
"Do you want my printouts on whether you should pinch-hit?"

Understandably, it is quite useful if you can use an algorithm that reduces that to bunt/don't bunt. And the first line of that algorithm might as well be "if Nolan Arenado is playing 3B then GOTO NO."

"if Nolan Arenado is playing 3B then GOTO NO."

I lol. (and yes Nolan has had a fantastic defensive season this year, that is absurd--- I really hope someone compiles a video at the end of the season of just Arenado's plays, but does it right so that context of each at bat mattes, not just the physicality of the plays---note: I think there are probably about 30 or more players in baseball who the same could be done, make them about 10 minutes or less and I'm watching)

Maddon's point only stretches so far. Weaver, Martin, and Mauch all had very distinctive ideas about the specific factors that won and lost baseball games. Insofar as they had the power to do it, they molded their rosters around those ideas. Some were "right" and some "wrong" from the perspective of modern analytics, but they were decisive in articulating what they wanted.

If those managers had "analytics guys" coming after them, they'd (assuming Martin was sober and in a good mood, lol) hear them out and then explain, "OK, but I'm making Decision X based on factors P, Q, and R, which aren't fully incorporated into your model."

I've observed Maddon's whole career. While he's an engaging character and media presence, I've never got the sense that his teams embodied a specific style of play akin to the managers he mentioned.

Maddon's point only stretches so far. Weaver, Martin, and Mauch all had very distinctive ideas about the specific factors that won and lost baseball games. Insofar as they had the power to do it, they molded their rosters around those ideas. Some were "right" and some "wrong" from the perspective of modern analytics, but they were decisive in articulating what they wanted.

My favorite punching bag, Whiney Herzog also had a philosophy that worked, when it worked, and failed spectacularly when it failed, but at the same time was often mis-represented by the press. The press presented it as a put the ball in play and let speed take over type of thing, pitch to contact and let defense work for you, when it was a bit more complicated, his teams won when they led the league in obp, obp is an analytic way at running a team. And with his speed, and multiple switch hitters, he compensated for power and bench weakness's etc.

Maddon's point only stretches so far. Weaver, Martin, and Mauch all had very distinctive ideas about the specific factors that won and lost baseball games.

Well, "pitching, defense, and three-run homers" is pretty much unassailable, but it's pretty trite. It's basically be good at everything.

I've been so sick of this pretentious, insufferable prick for years.

All these analytics and nobody bunts with the 3B shifted halfway to the first base dugout.

I don't think Billy's firings had much if anything to do with ownership meddling with his managing.

Early in the second half of 1978 George DID demand that Billy use a “new regular batting order” that included Thurman Munson and his sore knees in right field, Reggie at DH, Gary Thomasson in left field, Mike Heath at catcher and Chicken Stanley at shortstop; the lineup was used for a couple of games. Martin was fired shortly thereafter, although for unrelated reasons (calling Reggie and George “one’s a liar, and the other was convicted”, among other things).

And he chose another odd example. Earl Weaver famously did his own analytics.

Maddon's point has basically nothing to do with "analytics" per se - or rather, "analytics" is just the MacGuffin. His bigger point is about the diminished role of the manager in general, and analytics just happens to be the means by which that is happening.

Whenever somebody complains like this you have to question how much of this is actually true and how much of this is the person deflecting blame.

#22 Yeah, but he was smart enough (generally) to in effect ask why. Didn't much matter in his best teams when the question was basically sorting out the playing time for Buford, Blair, Frank Robinson, Powell and (after he developed) Rettenmund (increasingly giving Powell days off against tough lefties) but trying to pick the 4 best from 5 very good options. More or less straight platoon at catcher and trying to give the infielders the odd day off when it was convenient. (Belanger hit in both 1969 and 71 and that made his job easier)

When he was working with more marginal talents later he tried to extrapolate from the small sample size. To get limited players playing time when they had the best chance of success. So it wasn't just that a player had minimal success against one pitcher, but what was he like against similar.

I think it moved the needle, even if there's a limit as to how far.

He also tried to work Hendricks and Cuellar together as much as possible. Cuellar was far more comfortable with Hendricks catching but ... well Cuellar was going every 4th day and Hendricks rarely played against LHP.

#29 Yeah take the talent of the 1971 team (not the best record but the easiest to manage) and everybody's going to come up with a broadly similar usage pattern. Most managers would love having to deal with the problem of finding playing time for Buford, Blair, F. Robinson, Rettenmund and Powell in a non DH league. Great D. Durable pitchers who throw quality strikes and keep the ball in the park.

And Walt's #25 is exactly right. The manager would be perfectly willing to use anything which simplified decisions. But they want things that don't require them to weigh odds quickly while under stress. Things will become hard and fast rules and very little in analytics lends itself to that.

Whenever somebody complains like this you have to question how much of this is actually true and how much of this is the person deflecting blame.

I'd go a step further. Maddon is burning bridges. Just because it's obvious who he's describing, what's the point in bad-mouthing your former colleagues by name?

One of the first solid pieces of advice I received while in my first gig after undergrad was to be gracious to a soon-to-be former employer because the next one would like to know you'll behave honorably when you move on from his or her office.

Maddon is burning bridges.

But he's sucking up to the press. Is his next job in the booth?

The Rays and Cubs with Maddon had a top down approach and had great success. Maddon moved to an organization that was not strong and his teams struggled and he struggled. Kind of lends support to the idea that at the very least Maddon is fungible.

But he's sucking up to the press. Is his next job in the booth?

Or as an analyst. Heck, Maddon is probably having a studio constructed inside his Hazelton home as we speak.

Maddon, like him or not, has an enviable record across a few orgs. I don't think he's burning bridges as much as describing a phenomenon that he thinks is hurting the sport. What's interesting about the "GM in the locker room"comments is that teams seem not to want to talk about it or even acknowledge it. The cardinals fired Shildt because of philosophical differences but the likely story is he didn't like the front office meddling in his decision-making. Kershaw gets yanked in a perfect game, Snell gets yanked in a WS when he was lights out. Dave roberts says he only is one voice in the room in lineup construction. Are the GMs real making more of these decisions in-game and we just don't know it ?

I think managers should be just one voice in the room on pre game decisions.

For in game it really depends on the decision. Like the manager opts not to use Bob as a PH because when he went to give Bob the nod he saw Bob was high as a kite. Great decision. Outside of things you literally have to be in very very close proximity to in order to make a proper decision there's no need for it to be Manager is a Dictator type setup. Plenty of teams over the years will have someone else run the pitching staff and there's no reason why a GM or someone he designates other than the manager can decide when and who steals bases and when a hitter bunts.

I don't think managers should be just a voice. Look at
The maddon/Pujols/Minasian process. If I remember it maddon told
Pujols he'd be in lineup the next day and then the GM told maddon no and then Pujols
Got all mad so they released him.

The manager needs to have the authority. What's a player to do now - instead of having a discussion with manager the player has to schedule time with the Gm to talk about playing time. Everyone in the room learned that day that maddon wasn't making decisions anymore. That undermines their ability to lead. There always needs to be a clear leader.

Everyone in the room learned that day that maddon wasn't making decisions anymore. That undermines their ability to lead. There always needs to be a clear leader.

why? Whats your basis for saying that? Its baseball, not following Monty to the beaches of Normandy. Does it really matter if the guy coming to the plate has no respect for Maddon? All that matters is that he's facing a RHP that he does well against, he can swing away with no one holding the runner, etc.

What was Bob using?

The position of any manager - actually, in sports or not - requires three primary skills:

(1) The ability to evaluate players and put them in a position where they will best succeed.
(2) The ability to make in-game, real-time decisions.
(3) Carry the respect of your locker room so that your roster will be motivated to play hard for you.

Rarely is someone excellent at all three.
Analytics have very little influence on #3. And one could argue that analytics insight regarding #1 is far from comprehensive.

there's no need for it to be Manager is a Dictator type setup

Delegation is reasonable as long as the lines of responsibility are clear (as they are in principle, say, in the NFL). But the rhetoric of baseball is still that the manager takes all responsibility (the occasional Bill-Veeck fan voting or College of Coaches aside). So you can see why somebody who is going to get blamed for everything might also want actual control over the decisions he'll get blamed for.

(3) Carry the respect of your locker room so that your roster will be motivated to play hard for you.

If this was important and Duke's theory in no. 45 is true than apparently MLB has spent the last few years of the analytics revolution deliberately weakening the ability of their roster to play hard.

The Maddon/Pujols situation seems to suggest Maddon overstepped.

There's a book on the 2016 Cubs that talks at length and has Maddon himself talk a ton about everyone knowing their place, their role, and their strengths and weaknesses.

It seems Maddon the last time around didn't know these things for himself.

Re 47

You don't need A manager to do 1 and 2 and as mentioned in football you have a bunch of people doing 1 and 2 for various parts of the team. You really don't even need A manager for 3 as well. You basically need a manager/HC because tradition and the rules kind of demand one.

The most common phrase used when discussing why a manager was let go is "he lost the locker room "

Now, this could be yet another baseball syllogism of sorts or it could be very important. I suspect it is very important

in MLB, it's more "clubhouse" than "locker room."

the Phillies correctly figured out months ago that Joe Girardi had lost the clubhouse, and they were wise to fire him. doesn't mean he is an awful manager, but apparently the vibe in the 'house' was that Girardi was too beholden to old/over-the-hill veterans - and that was off-putting to some players.

I've never got the sense that his teams embodied a specific style of play akin to the managers he mentioned.

I think I disagree. I'd love Dag's opinion as he certainly has a better idea than I which historicl managers might be said to have a "specific style" and how much that style has to differ from the run of the mill to matter. Plus (shocking!) I didn't follow Maddon that closely in Tampa.

The most distinctive element I think was that he was more than willing to move players around the field. It wasn't positionless baseball but it was "role-diverse" baseball. In Tampa he probably had to do that out of necessity but with the Cubs he could have easily trotted out Baez, Bryant, Zobrist, Contreras, Happ at the same positions day after day. In the 7th game of his career, our supposedly defensively challenged young 3B Kris Bryant found himself starting in CF. He didn't create the IF/OF player but Zobrist pretty much perfected it. Javy played all over the IF, Schwarber's bat found a home in LF, so did Contreras's and, while not his wisest decision, Chris Coghlan's bat briefly found a home at 2B once. When Heyward's bat failed and Fowler left, Heyward started playing a lot more CF. The ideal type of a Maddon team would be 9 Zobrists and 3 Contrerases with Maddon planning a game with one Contreras behind the plate, one at third and one in CF just to see if it worked. I don't recall him really doing anything distinctive on the pitching side other than never fully trusting Hendricks' stuff.

We don't get to see it really and maybe it disappeared or was a fluke, but he seemed to value keeping the players loose. Rizzo's in a slump, let him bat leadoff. Rizzo again doing a little faux-macho thing with Pedro Strop on the mound in a close, important game -- no worries. I recall he did the Phil Jackson thing of buying books for each player to read on a long road trip (may have only tried that once). And, even for managers of the time, he was more open to analytics than most -- granted maybe only as long as he was still recognized as the expert in the room.

Not that there's much competition but I certainly consider him the best, most distinctive** Cub manager of my lifetime.

** in a good way. Lee Elia, Bruce Kimm and Jim Essian were also distinctive.

Well, "pitching, defense, and three-run homers" is pretty much unassailable, but it's pretty trite.

You forgot "and when you have a HoF SS who can play every inning for 18 straight years, you don't need to carry a backup SS."

On Twitter, Tango shared his perspectives on this article/Maddon's ideas based on his time working with clubs.

You forgot "and when you have a HoF SS who can play every inning for 18 straight years, you don't need to carry a backup SS."

Earl Weaver only got a couple full seasons of Ripken, and only one in which he was a full-time shortstop.

#57 Weaver's the guy who had the insight to look past the images that said "3B" for Ripken and realize that he could be a plus SS defensively despite being much larger and less agile than a stereotypical SS.

And this is an interesting example of his independence. The Orioles had traded a pretty fair 3B to open up the position for Ripken and Weaver chose to play Ripken at SS.

56. What a terrible medium for his response. Couldn't just write a blogpost ? Management 101 - if you can't explain something to your boss, you've already lost the argument

Wasn't Ripken's dad in the organization and close to Earl at the time?

#60 Yes.

#57 Weaver's the guy who had the insight to look past the images that said "3B" for Ripken and realize that he could be a plus SS defensively despite being much larger and less agile than a stereotypical SS.

Absolutely, and he gets a great deal of deserved credit for that. It just didn't directly benefit his record as a manager as much as Walt seemed to be implying in 55.

#29 Yeah take the talent of the 1971 team (not the best record but the easiest to manage) and everybody's going to come up with a broadly similar usage pattern. Most managers would love having to deal with the problem of finding playing time for Buford, Blair, F. Robinson, Rettenmund and Powell in a non DH league.

Weaver also routinely had a great bench. The 1971 O's had a bench whose RC/27 outs was about 33% better than average -- not better than the average bench -- better than overall league-average offensive performance. It's one of the best benches ever. Overall, he had the most offensively productive bench in the league (defined by RC/27 outs) seven times. That's finding the right role for his role players a whole buncha times.

why? Whats your basis for saying that? Its baseball, not following Monty to the beaches of Normandy. Does it really matter if the guy coming to the plate has no respect for Maddon?

Guess it all depends on if managers are solely managers of the game or if they are also managers of men. There's 100+ years of info that front offices, players, and managers themselves think the managing men thing matters a lot. I've been to a few SABR conferences where GMs are asking what they look for in a manager. Usually, they say something about how you need a broad philosophical agreement on in-game tactical decisions but that you can't worry about micromanaging that. If teams are veering to micromanaging, that just sounds like a terrible idea. I mean, there's always going to be bad decisions. You can't fixate on each one. Beyond that, GMs generally spend more time talking about the managing of men. One GM (Mark Shapiro, then of CLE, I think) said the three key things he's looking for are: 1) self-awareness (so the guy comes off as he can), 2) communication (so his intended message gets through), and 3) prioritization (there's always 100 fires to put out at any time - focus on the right ones). I don't mean to say that Shapiro is some unquestionable savant - but this is the sort of things GMs say matters to them with managers.

I mean, in many ways a clubhouse/dugout is a workspace like any other. It helps if there's someone running the place who is competenant at handling the day-to-day BS to the employees can foucs on doing their dang jobs. And if the BS steams over - yeah, that can affect the workplace. Just like it would any other workplace.

All that said, you know one guy who thinks the manager's primary job is in-game strategy? Joe Maddon. I heard him asking the question of the manager's most important role once in a radio interview, and he said handling the bullpen. The host asked if it's more important than handling players - and yes, Maddon said. It is.

I think I disagree. I'd love Dag's opinion as he certainly has a better idea than I which historicl managers might be said to have a "specific style" and how much that style has to differ from the run of the mill to matter. Plus (shocking!) I didn't follow Maddon that closely in Tampa.

Thanks. I don't have many/any brilliant insights into a particular style of Maddon. I thought he was great in 2015-16. His decline began at his peak: the last two games of the 2016 WOrld Series when every pitching move he made was questioned in real time and nearly all blew up in his face immediately.

But he's been a great manager overall. He's got a nice track record dealing with kids from his days in Tampa. He prides himself on maintaing the exact same disposition every day - and I assume he's really good at that too. He was really into analytics when he started out, though he's clearly less comfortable with that in the StatCast era of Big Data Baseball (this was true his final years in Chicago, too).

Beyond that, the era of Big Data is just the latest chapter in a long trend of managers losing authority. Early managers were often barely more than a travelling secretary. They made sure the bills got paid and things worked out day-to-day. Often, in-game tactics were left to a team captain - but there wasn't much in-game tactics back then (we're takling 1880s and stuff here). Then early tactics began: defensive alignemnts, bunting, etc. Older team captains couldn't play. By the 1890s, the manager came into his own. He was THE BOSS. He answered only to the owner - and sometimes became owner (Mack, Comiskey, Griffith, heck Jimmy McAleer became a part owner). The early 20th century was the peak for managers.

Then came the post of GM in the 1920s. They developed an early modern farm system. Managers still had plenty of authority - and large amounts of control over the roster - but they gradually lost control of the roster by mid-century. They were clearly below GM on the totem pole. But they still had authority over their players - so the rise of free agency limited them a bit more. (Not as much as it limited the front office's power, but some manager power was tied to their links to the front office). Another downturn came in the turn of the millenium. Ever since there was a post called GM, managers could go in and out of it. Whitey Herzog was manager/GM in the 1980s St. Louis. Bobby Cox was GM. Heck, Jim Frey was Cubs GM at one point. After the 1990s, you basically never see that (I'm not sure you have seen it at all in the 21st century).

Now enter Big Data. For the first time ever, the front office can say that their data is better than what the manager can see with his eyes when it comes to the little things taking place on the field. That removes one area managers always had an advantage.

Is Maddon burning bridges? I see it that he thinks if this is where the job is headed, he isn't interested in going with it.

is decline began at his peak: the last two games of the 2016 WOrld Series when every pitching move he made was questioned in real time and nearly all blew up in his face immediately.

Maddon is a good talker. has he ever explained WTH he was thinking in those games?

for that matter, has Francona?

had no dog in that hunt, but it was the worst-managed pair of games I have ever seen - regular or postseason. and of course it ends with perhaps the worst hitter ever to end a World Series at the plate. that was almost poetic, amid the carnage.

#63 Part of the reason for his successful bench is that he was good at putting limited players in a situation where they could succeed. I remember he had a player who (in Weaver's words) couldn't hit anything with spin. He also had at least one guy who couldn't hit a plus fastball. Didn't ask them to do what they couldn't do.

And because he ran a small pitching staff, the back end of his bench could be made up of some very limited players who could cover off each other. I remember he once said that the 25th man was a really tough decision for him. If he picked player X what situation would he not have a good solution for.

Or to put it another way, I think it's fair to say he judged players by what they could do -- who gave him the most plus situations. That's actually pretty unusual. It's far more common to judge players on what they can't do.

Maddon is a good talker. has he ever explained WTH he was thinking in those games?

He's gone over it. He admits to making one mistake: He should've had an idea who to warm up for the ninth in Game 6, after the Cubs lead expanded from five runs to seven runs. He and Bosio dithered for seeminlgy ever, and they had to have Chapman go out to start the ninth (where he walked a guy). He stands by all the other decisions he made. Mostly, it's him saying that he wanted his best arms out there for the most inning. Also, didn't sound like he trusted Hendricks as much as I think he should've.

The one slight thing I'll say in his defense is that the Cubs bullpen was their weak spot. For four months they're only reliable guys were Strop and Rondon - but both imploded in the last third of the season (Strop would recover next year, but Rondon never did). So there was Chapman, late season call-up Edwards, and swingman Montgomery. (Which makes his quick hook on Hendricks all the more questoinable but nevermind that).

I think he was badly burnt by a Tampa playoff game against the Red Sox where Boston overcame a big late deficit to win (may have even come back to win the entire series - I forget and am too lazy to look it up). I think that scarred him mentally. It threw off his risk analsyis. Look at what he did in Game Six of 2016: with a five run lead and two inning to play, he put his closer in the game Mind you, that closer recorded an eight-out save in Game Five (which was absolutely defensible: he was the big arm, it was a one-run lead, and it was a possible elimination game for the Cubs). But even afterwards, Maddon still defending using Chapman in the 8th inning of Game Six.

Really, I'd be willing to defend dang near everything Maddon did in Game Seven if he hadn't used Chapman in Game Six.

There's 100+ years of info that front offices, players, and managers themselves think the managing men thing matters a lot.

RIght and if so called "respect" is so important than why does the last 20 year trend of analytics is seemingly putting percentage baseball over that respect? That's what Im arguing. Apparently its more important for most teams to position a 2bmen in RF and gain some tiny advantage than to let ball players feel that Maddon is calling all the shots?

Or take the LAD. Do the dodger players have less respect for Roberts because he pulled Rich HIll in game 4 because of analytics when Hill was cruising? And do the dodger players have less respect for him?

SO I guess the first question is: Do players have less respect for a manager when he's being told what to do?

2: If they do lose respect does it make any difference to how they play? the analytics people probably just say "who cares what player X thinks? Just go up there and hit the ###### ball."

I dont see any evidence that they do.