PDA

View Full Version : MLBCenter Article


barzilla
06-28-2004, 01:06 AM
I wanted to forward everyone the article I'm writing for MLBcenter.com. Princess, I have numbers on Terry Francona if you want it.

How do we rate managers?
By Scott Barzilla

In the past few decades statistics in baseball have come a long way. In particular, rating hitters and pitchers is a lot easier now than it was thirty years ago. In fact, in the last decade defensive metrics have also come a long way. Even though we still have a ways to go we have a pretty good idea. However, managers are a lot more difficult to rate. How much credit or blame do we give managers for their team’s performance?

The problem with baseball is that most of the action is one on one, so strategy and coaching is not nearly as important as it is in the other sports. Players develop their craft in the minors for three to five years before they even come to the big leagues. Usually, a change in managers, pitching coaches, or hitting instructors does not change how a hitter hits or a pitcher pitches. So, what does a manager control? Managers are usually judged on how the shape the lineup, when they use subs, and how they handle the pitching staff.

Contrary to popular belief, most of the credit or blame for how a team performs lies with the general manager. The talent or lack of talent on the team determines how good they will do. This is another major difference between baseball and the other sports. There is not one manager in baseball that is also mostly responsible for player personnel. In football and basketball there are a good number of coaches that are also responsible for player personnel. It used to be that way in baseball with the likes of John McGraw and Connie Mack. Managers these days don’t have that kind of clout.

So, if most of the credit goes to the general manager then how much goes to the manager? When sabermetrics was gaining a foothold, the Pythagorean method was developed. Most teams should match their expected record based on the number of runs they give up and the number of runs they score. No one has come up with a good reason why some teams overachieve and others don’t. When there is no reason we usually assume it is attributable to randomness, but we do notice that good teams tend to overachieve and bad teams tend to underachieve.

Yet, the question with the Pythagorean method is whether is can be used to evaluate managers. Of course, if we have learned anything in sabermetrics it is that no single statistic can explain everything. So, like when we evaluate hitters, pitchers, and fielders, when we evaluate managers we have to use a variety of measures. In our analysis today we will see their record, winning percentage, average finish in standings, post-season appearances/successes, and their Pythagorean record. It isn’t perfect, but it may be a start. We will include the twelve managers at the big league level that have managed at least five complete seasons.

1. Bobby Cox

Record: 1906-1465 (.565)
Standing: 2.3
Postseason: 12 division championships, 5 pennants, 1 World Series title
Pythagorean: +40
Per Season: +1.90

2. Joe Torre

Record: 1680-1509 (.527)
Standing: 2.5
Postseason: 8 division championships, 6 pennants, 4 World Series titles
Pythagorean: +23
Per Season: +1.21

3. Dusty Baker

Record: 928-789 (.540)
Standing: 2.1
Postseason: 3 Division Championships, 1 pennant
Pythagorean: +18
Per Season: +1.64

4. Felipe Alou

Record: 791-778 (.504)
Standing: 3.0
Postseason: 2 Division Championships
Pythagorean: +15
Per Season: +1.67

5. Frank Robinson

Record: 846-909 (.482)
Standing: 4.0
Postseason: N/A
Pythagorean: +13
Per Season: +1.30

6. Tony La Russa

Record: 2009-1789 (.529)
Standing: 2.8
Postseason: 9 Division Championships, 3 Pennants, 1 World Series Title
Pythagorean: +11
Per Season: +0.48

7. Art Howe

Record: 1058-1046 (.503)
Standing: 3.3
Postseason: 2 Division Championships
Pythagorean: +10
Per Season: +0.77

8. Jack McKeon

Record: 845-782 (.516)
Standing: 2.7
Postseason: 1 Pennant, 1 World Series Title
Pythagorean: +7
Per Season: +1.00

9. Bruce Bochy

Record: 694-746 (.482)
Standing: 3.6
Postseason: 2 Division Championships, 1 Pennant
Pythagorean: +6
Per Season: +0.67

10. Buck Showalter

Record: 634-595 (.516)
Standing: 2.8
Postseason: 2 Division Championships
Pythagorean: -1
Per Season: -0.11

11. Lou Pinella

Record: 1382-1234 (.528)
Standing: 2.6
Postseason: 4 Division Championships, 1 Pennant, 1 World Series Title
Pythagorean: -2
Per Season: -0.13

12. Jimy Williams

Record: 866-746 (.537)
Standing: 2.6
Postseason: N/A
Pythagorean: -22
Per Season: -2.44

As everyone can probably tell, I am an Astros fan, so I found the results very interesting when I got to the bottom of the list. When you see his Pythagorean rating with the rest of them you have to wonder what the Astros were thinking when they hired him. Well, that is pretty obvious. Williams has the third best winning percentage among the twelve and his standing rating was tied for fourth among the twelve. Obviously, that’s pretty good right?

Well, when you see that he has never won a division championship your interest has to be piqued some. In some seasons, those two and a half games would have been enough to win the division. A quick look at the teams he has managed shows why his winning percentage is so high. A trained seal probably could have done about as well.

When we look at the top of the list we see that Bobby Cox has about as large a lead as Jimy Williams has a grip on last place. His team has won the divisional crown the last ten years in a row. Joe Torre and Dusty Baker finished second and third even though Tony La Russa has had more post-season appearances than both of them. The key in their case has been an incredible consistency over the past decade.

So, how do we use this information? Organizations could start looking at Pythagorean records when deciding on managers. This might prevent teams from being blinded by overall records and hire managers that will get the most of the talent that they have. Furthermore, when we study further we find that most of the great managers have strategy ideas in common. Then, finding a successful manager might be a little bit easier.

Scott Barzilla is a teacher in League City, Texas. He is the author of Checks and Imbalances and The Status of Baseball Management. They can be found on Mcfarlandpub.com, Amazon.com, or Barnesandnobles.com.

PissedPrincess
06-28-2004, 12:04 PM
Excellent! I knew Francona sucked. :angry:

barzilla
06-28-2004, 02:12 PM
Princess,

I didn't give you any numbers for Francona yet.

PissedPrincess
06-28-2004, 02:18 PM
You did in Pm. He is -1.5 games a year right?

barzilla
06-28-2004, 02:28 PM
No, that was minus 0.25 games per season. He basically is an average manager up until this year (I haven't done this year's). I used Baseball-reference.com for the others.

Sandy
06-28-2004, 04:57 PM
Not a bad little article.

1) The data is definitely gonna need to be in a more readily digestable format. Don't know how much you know about HTML, but those stats need to be in a grid, (sorted by either the per-season +/- or the total differential in games.

2) The obvious (to me) sidebar issue to help place your article points in context would be a look at two individuals actual year-by-year runs - (Jimy and Dusty would be fun, simply because they're in the same division at this point). It's one thing to see the totals, it's another to see the consistency in pattern.

3) Not that it needs to go into THIS article - but in the general discussion of 'manager grading', I'd love to see a team RC vs. actual runs scored method developed. That's probably more applicable as a SABR suggestion, though.

barzilla
06-28-2004, 05:49 PM
Sandy,

1. I'm hoping MLBcenter will format it for me. I'm not much for HTML but I know exactly what you are saying.

2. What I noticed about most of the managers is that they had several seasons in both categories but either had bigger positive seasons or more positive seasons. Jimy has had only one positive season (+2 in 1999 when he won AL Manager of the Year) in nine complete seasons. Plus, he has had two incomplete seasons, so I would have to guess he was also negative in those seasons, but I have no way of knowing for sure. Your comparison idea is a good one though. If you simply go by the averages you see that Jimy Williams managed teams gave up about 3.5 games to the Yankees under Joe Torre, 4 games to Dusty Baker, and 3 games to Tony La Russa. That's quite a hole.

3. That sounds pretty good too. Incidentally, does baseball-reference.com have RC against? It would be good to see both ends of that equation.

bagger
07-03-2004, 02:22 AM
brenly's only had a few years under his belt, but where does he rank if we pretend he's had five?