This may shock some people, but according to some of our advanced metrics here at Baseball Prospectus, Ted Williams was actually one of the greatest hitters ever to live. Really. Williams, of course, was the last human being to hit .400 in a season and was said to have 20/15 vision. After he retired from playing, Williams eventually accepted the position of manager with the then-Washington Senators, but famously was frustrated by the position. As the manager, one of his jobs was to help his players to get better, and who better to teach them to hit than one of the greatest hitters ever to live? It’s just that Williams, while he was gifted in the art of hitting himself, lacked the ability to understand that not everyone is Ted Williams when it comes to hitting.

We get to ask the question "What was wrong with Ted Williams?" about as often as my mother-in-law… oh, hi Mamochek! Didn’t see you over there. Not only that, but eventually I’m going to show that Ozzie Guillen was better than Ted Williams. Keep reading. (Except you, Mamocheck.)

The manager of a team has three (main) jobs. He is responsible for in-game strategy; he’s the official spokesman for the team both outside the clubhouse (with the media) and inside the clubhouse (keeping the players in line); and he’s the head coach. The first issue has been tackled by several people and the second isn’t really quantifiable. But what of the manager’s third job: teaching his players to be better. Are there some managers who excel in this area and some who, like Teddy Ballgame, just can’t seem to preach what they practiced?

It’s common to rate how good managers are at this sort of thing by looking at the results that his players produce for him. Of course, this isn’t fair to managers who have been given a collection of retreads, has-beens, and never-will-bes to coach, and overstates the abilities of managers who have the luxury of working with a lineup filled with All-Stars. How to start to disentangle this kudzu?

Some time ago, I looked at this very issue concerning a manager’s ability to influence his pitchers’ performances (at another website, which sadly has closed down…). Indeed, I did find differences between managers. Some were very good; Bobby Cox rated particularly high, as you might imagine. Some left a bit to be desired. What about the batters? For those of you who don’t like the methodological guts, please say to yourself "and then he hit a couple buttons and a miracle happened," and skip to the results.

Warning: Gory Methodological Details

Like my article on catcher/mentors a few weeks ago, I used an HLM-based analysis to look for "teaching effects." This sort of model is often used in "real" educational evaluations. I located all batters who fit the following criteria.


  1. He had more than 200 PA in the season in question (1990-2009);
  2. He played for the same team the whole season;
  3. That team had the same manager for the whole season;
  4. He played for more than one manager over the course of his career;
  5. His manager for the year had at least 25 qualifying player-seasons since 1990 that he had also shepherded.


This controls for a few nasty confounds, including small sample sizes (for both manager and player), cross-contamination of manager influence, and making sure that one manager isn’t overly blamed as being the cause for a bunch of guys who were awful and thus, never played with anyone else ever again.

I used the three "true" outcomes (strikeout rate, walk rate, HR rate, all per PA less intentional walks) to see if there were consistent effects to be found. (Any stat can be used, if one so desires.) I used these three because they are they are generally the most reliable stats in the batter’s toolbox. I controlled for league-year, home stadium, and age in the model (all as fixed effects), and then input the manager identity as a fixed effect as well. I used an AR(1) covariance matrix to control for within-subjects effects (i.e., player talent level). I set the comparison age to 27. So, these are numbers for what we would expect of the average 27 year old (adjusted for the 2009 National League), if he were assigned to each manager.

The results

I’ve been asked in the past to post the entire list when I do these types of articles, although the sheer size of that list makes it impractical here. If someone really wants to see the list, contact me backchannel; I will be happy to share.

Top Five at Teaching Walks
Manager              %
Ozzie Guillen      12.5
Clint Hurdle       12.1
Frank Robinson     11.8
Gene Lamont        11.6
Phil Garner        11.6

Bottom Five at Teaching Walks
Manager             %
Tom Kelly          7.1
Ron Gardenhire     7.1
Bobby Cox          7.4
Tommy Lasorda      8.6
Ned Yost           8.7

It's interesting that the two Twins managers were the most walk-phobic of the group (and Gardenhire was Kelly’s disciple). Guillen, for all his bluster, is apparently telling his team behind the scenes to work the count a bit and try for a walk. Let’s move on to strikeouts.

Top Five at Teaching Strikeout Avoidance
Manager              %
Felipe Alou        14.9
Ken Macha          15.0
Frank Robinson     15.1
Ned Yost           15.3
Mike Hargrove      15.7

Bottom Five at Teaching Strikeout Avoidance
Manager              %
John Gibbons       20.5
Cito Gaston        19.4
Fredi Gonzalez     19.2
Jim Fregosi        19.0
Terry Francona     18.7

Yost apparently doesn’t like his hitters to strike out or walk; instead, his philosophy seems to be "put the bat on the ball!" I checked to see if there was a correlation between each manager’s effect on walk rate and on strikeout rate. The result was a correlation of .001, or almost nothing… not even some random noise to push it up to .05 or thereabouts. So, being good at teaching walks does not mean being good at teaching to stay away from strikeouts.

Alou gets the high mark here in strikeout avoidance, which seems to match his reputation. Francona’s players are particularly bad with the strikeouts. I’ll have to tell him that when I look at his two World Series rings.

What About Power Numbers?


Top Five at Teaching Home Runs

Manager              %
Lloyd McClendon     5.5
Jim Tracy           5.0
Fredi Gonzalez      4.9
Ozzie Guillen       4.9
Joe Maddon          4.7

Bottom Five at Teaching Home Runs

Manager             %
Tom Kelly          2.4
Ron Gardenhire     2.5
Bobby Cox          2.8
Eric Wedge         2.9
John Gibbons       2.9

Kelly and Gardenhire are present again, paired together and both on the wrong side of a list: their presence coincides with depressed home run and walk rates. Now, that may mean that they are encouraging their players to have at-bats that end in singles or doubles or harmless fly balls to center, but clearly, the Twins’ philosophy has steered away from the longball but also away from the walk.

On the other hand, Guillen shows up on the list for increasing homer and walk rates; he’s also 10th out of 61 qualifying managers on increasing strikeout rate; if anything, his philosophy seems to push players toward a Three True Outcomes type of offense.

What is the Deal with Ozzie?

I lived in Chicago for six years (on the North Side … sorry Sox fans) and got to hear from the Blizzard of Ozz on a daily basis for most of that time. He’s never been known for keeping quiet when he has an opinion, which makes him at once entertaining (and how!) and unpredictable. (Others have been less kind in their adjectives.) But these numbers show something else about him. His players are very affected by his mere presence in the dugout. His coaching (or coaches) apparently makes his players better at hitting home runs and taking walks, although that also seems to lead to a higher strikeout rates. Some managers have very little impact on their players overall results, but like him or hate him, you know that you play for Ozzie Guillen.

What’s odd is that Ozzie Guillen was not a great hitter in his day. In fact, he holds the record for the most number of career games started in the ninth position in the batting order, and posted a career line of .264/.287/.338; Ted Williams he was not. Guillen rarely struck out (only 511 times in 7,133 career plate appearances), walked (239), or homered (28). Yet, his players now seem to be much more in line with everything that he was not when he was a player.

I didn’t run the numbers for Ted Williams in his managerial stint. However, I would bet that his numbers would reveal a man who had little effect on his players’ outcomes. Despite being one of the best hitters in the game’s history, Williams felt that he couldn’t impart much in the way of instruction to his hitters. Guillen, on the other hand, appears to have figured out a way to do exactly that and the proof is in the numbers. Guillen will never make the Hall of Fame as a hitter, but if the point of a manager is to have an effect on his players, then Ozzie is probably a much better manager than Ted Williams was.

Russell A. Carleton, the writer formerly known as 'Pizza Cutter,' is a contributor to Baseball Prospectus. He can be reached here.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
May I just say that, as a baseball fan, I love Ozzie Guillen? It's rare to have anyone in professional sports speak his mind so honestly, so often. Normally the mainstream media punishes people for this. Ozzie is more than entertaining; he's also perceptive. Articles like this will help keep him employed, so thank you for it.
My impression is that managers do relatively little in the way of instruction. Would these numbers correlate better to the hitting/pitching coaches working under the managers? When Kelly left the Twins, did Gardenhire keep the same staff?

Even if managers aren't directly responsible for player instruction, they should get some credit for setting the tone of the organization through the hiring of the coaching staff. Perhaps Ozzie is effective at selecting a staff?
I think that you're closer to the truth here. Much of the dirty work is probably the job of the specialty coach. However, I subscribe to the theory of management that says that Ozzie picked the coach and let him do his work, so Ozzie gets credit for that.
Then why not the GM that picked Ozzie of the owner who picked the GM?

More seriously, I think it would be pretty hard to separate out all of the factors (control for everything non-manager/coach related). Not that this isn't really interesting; it is.

Doesn't the fact the two Twins managers are grouped together suggest that there's something else Twins related going on there?

I guess when it comes right down to it, I don't find it credible that a manager could have an effect this large.
Can we see all of the lists, not just the top/bottom five? I think many of us scan through pieces like this looking for tidbits on our favorite team(s), and when there's no Dave Trembley our eyes gloss over and we start to nod off.
The full lists would have over-whelmed the texts. If you (or anyone else) would like a copy of the full list, send me an e-mail back channel and I will send it to you.
If I might make a suggestion, a reasonable compromise might be to put together a separate page with the full list or lists on it and then just drop a link to that into the main piece. There's no reason to treat a web piece like a newspaper piece (although even in the newspaper case there's precedent in the form of a sidebar for something like this).

Now, making this work with the BP content manager so that the list page doesn't show up in the article index might be an issue, but that's just a matter of programming.
Thanks for the entertaining read, but I think Brian Kopec's comment above is correct. The vast majority of teaching at the major league level is done by coaches, not managers. It would certainly be more difficult, but it would be more instructive to look at the effects hitting and pitching coaches have, rather than managers.
How are you controlling for Park Effects? Couple of the outliers in particular (Cox and Guillen) would seem to be impacted by Park Effects.
I entered home park as a fixed effect. Perhaps it isn't clearing up all the noise.
Russell: perhaps it would be possible to post the list of current managers?
How I envision stepping up to the plate while playing for Ozzie: "Wait for your pitch, now, wait for it, wait for your pitch... Hey now why the *** did you swing at that ****'s **** garbage ****, you dumb *** **** ****!?! I tell you every **** practice to wait for your **** pitch and you **** swing when I tell you not to ***** swing!!! You're gonna run tonight until your **** turns into a **** ***** ****!!!!"
Which is to say that repetition is the mother of learning.
In education, HLM models are used to disentangle school effects from teacher effects as well as teacher effects from student effects. Here, your "manager" (teacher) analysis is really just "team effect" (school), right?

Given that Kelley and Gardenhire have similar stats, one might wonder whether managers have any effect at all - perhaps it is all, as Brian suggests, the coaches or some other aspect of each organization (including organizational culture). Look at the recent history of the Angels - the dramatic increase in walk rate happened without a change of managers; it seems to have been an "Abreu" effect rather than a "Scioscia" effect.
On the first part, I think that's a fair statement in that the manager should be treated like the principal of the school (to extend the metaphor).

On the second part, when I did my pitchers study (need to get those archives back on line somewhere!), I ran into some similar problems. I specifically sought to correct for them in this study (some of those overly-stringent inclusion criteria were aimed at that). The Angels may walk more because of an Abreu effect, and that's fine. It won't show up as a Scioscia effect and if it isn't, it shouldn't.
The Williams bad manager/Guillen good manager make intuitive sense to me. Williams cannt teach 'talent' that he had in spades. How do you explain something you can just 'do'.

Ozzie on the other hand probably had to sweat blood to get his hitting up to the level it was at. He probably worked with every hitting coach he had, trying every gimmick in the book to find a way to hold his own at the MLB level.

It intuitively makes sense that the person who had to teach himself the hard way, would be better at showing people how it got there vs the guy who 'just did it'.
Those who can't do teach, and those who can't teach manage baseball teams.
Having been a professor, it's more than just that. Understanding the underlying process is necessary, but not sufficient for either doing something or teaching it. Williams was able to physically execute it. Guillen is apparently able to explain it well. There are some people who can both physically execute and explain well, but those skills are probably independent of one another.
As I recall from reading the great baseball book, The Science of Hitting, Williams was very effective at improving his players with home runs and walks.

Look at Frank Howard when Williams took over
I think it is an incorrect statement that Ted Williams could not teach hitting. Many of the greatest hitters in the game (Tony Gwynn & Mike Schmidt come to mind) gave Williams credit for helping them understand the art of hitting. His book, the Science of Hitting, is on the bookshelves of most college and professional hitting coaches, I suspect.

For me, there are a lot of uncontrolled variables that make it very difficult to measure...though I appreciate the effort, I think this kind of analysis is a bit of a stretch. By the time players get to the big leagues, they are what they are. If you have a team of no OBP, no hit guys -- well, that's what you are going to get...even if you are the greatest hitter who ever lived.
Its also farly unfair to say that Teddy was a bad Manager. in 1969 he took over a last place (10th out of 10) ball club and led them to an 86-76 season, increasing their wins by 20, and increasing their runs scored by 170 while only slightly improving their run prevention. Unfortunately the message didn't stick, and his teams run scoring dropped all 4 years as a manager to a low of 461 runs score in 1972, their first in Arlington.

In the 1968 off season, turnover was limited to Eddie Brinkman and Lee Maye replacing Ron Hansen and Ed Stroud, with Eddie Brinkman playing a significant amount in 68 and Stroud playing a significant amount in 69.

The overlap:
Name 1968 1969
Casanova .196/.210/.252 .216/.254/.282
Epstien .234/.338/.366 .278/.414/.551
Allen .241/.301/.343 .247/.337/.389
Brinkman .187/.259/.202 .266/.328/.325
McMullen .248/.326/.382 .272/.349/.425
Howard .274/.338/.552 .296/.402/.574
Unser .230/.282/.277 .286/.349/.382
Stroud .239/.284/.376 .252/.353/.393

Sure 69 was a better offensive environment than 68, but not one of these 9 returns had a lower ANYTHING. 27 out of 27 stats raised.
I agree with Irgreen that this is navigating through some very thick fog, to borrow a Jamesian phrase. I'm not sure this line of research is worth the effort. (That doesn't mean I don't appreciate the article or the effort. In fact, it is just the opposite. I love that Pizza, er, Russell took a risk.)

I disagree with Irgreen that major league players 'are what they are.' Surely there is some benefit to coaching, even at the major league level. Examples abound.
It's possible that Williams was better when he had a chance to sit down, think things all the way through and write it out, but couldn't explain things on the spot in a coaching situation. It's two different skill sets.
Howard was a beast before Teddy got to him:

36 homers in 1967, and 44!!! in 1968.

However, the big change for Howard was patience. 54 walks in 1968 against 141 strikeouts. In 1969, Howard walked 102 times and struck out only 96. (while clubbing 48 homers.)
Ed Brinkman saw a big jump in his batting stats in 1969 (admittedly from really bad to mediocre). I was a fan back then, and I'm almost certain he credited that to Williams. Mike Epstein and Del Unser also saw significant improvement, if only for a short time. Maybe Williams had some immediate impact that gradually wore off.
Gibbons worst at teaching strikeout avoidance and worst at teaching home runs? That's an interesting accomplishment.
Sorry Russ, I'm not buying it. I think roster construction has a lot bigger effect on TTO's than managerial coaching. I think you have to look at the preferences of the GM and the club philosophies on baseball more than the coaching style of the manager.
I'm going to have to agree with Peter here. Just thinking back, the White Sox of recent years seem to have more walk/power players (Thome, Konerko) than the Twins (Piranhas...) Unless you've accounted for that in the statistical description and I just didn't understand it.
The model specifically corrects for that. That's the beauty of HLM. For example, if I end up managing a team of walk-happy guys who have been walk-happy elsewhere, then the model will not credit me with increasing their walk totals.
Russ do you have a mathematics degree? Where does one learn these analytical tools if he's sick of being shown up like this?
Peter, my degree is in psychology. I learned all my stats through the research requirements in the program. If you do want to learn, take a class. This sort of model is an advanced model and you have to work your way up to it, but it's only knowledge.
I would like to see this study done using EqA or wOBA or linear weights. The 3 true outcomes note the shape of a players production but not overall value. It is presumed that a hitter who hits more homeruns and walks more is more valuable, but this is not necessarily true.
What you've said is true and might make a good follow up. For what it's worth, I was more interesting in looking at the shape of the production for my own nefarious purposes. I was thinking of looking at some one-number outcomes, but it got cut for space.
"I used an AR(1) covariance matrix to control for within-subjects effects (i.e., player talent level)."

I'd like to believe your conclusions and learn something from this article, but you're just talking over my head. Can you actually explain and justify to the average reader what your process was, or do I need to come back after I've gotten an advanced degrees with an emphasis in statistical analysis?

I'll add that I find this problem a lot with BP. I'm smart, I'm educated (I actually tutored college-level stats for cripes sakes), and I'm been reading this website for 6 years. Don't make me take 4 trips to wikipedia or your glossary to get through an article.
Luke, I hear ya and I struggle with how to balance out how to present exotic methods. Part of the problem is that if I went all the way into explaining AR(1), then my articles would become the Wikipedia entry and that would overwhelm the baseball end of things. To that end, I try to simply state what I do. There are people with experience in this area who will understand the short-hand, but alas there are people who won't (not through any fault... like you, they probably just haven't learned this particular method.)

I regret that it sometimes gets into "trust me, I'm an expert."

The short version of AR(1) goes like this: I'm using a technique called mixed linear modeling. What that does is that it allows me to run a regression that incorporates both repeated measures (multiple seasons of data from the same player) but also look at factors that are outside the player that might influence the outcome (manager, park, etc.) Now, when we're dealing with within-subjects effects like that, we need to specify to the program what sort of covariance matrix that the program should use. AR(1) is a covariance matrix that allows for the fact that if we know what Player X did in 2004, we should have a pretty good idea what he did in 2005. The performances are no doubt correlated. AR(1) specifically says that we would expect certain elements in the covariance matrix to be correlated, because they come from the same person, and that the model should count this as within-subjects (individual person) variance, and not mistake it for variance associated with the other factors under consideration.
I keep waiting for Dusty Baker to show up here, but he never does! I'm mystified!

Actually, I did think he might show up in this post. David Gassko did a nice study in one of the THT annuals showing effects of managers on getting players to play above or below a marcel-like projection. Dusty ranked in the top 10. A lot of that was Barry Bonds, of course, but even if you remove him Dusty did pretty well. Here's a link to a summary I did of it.