The measure of a manager

In this spring of discontent, every member of the Royals' brain trust is under heightened scrutiny, none more so than manager Tony Pena.

For baseball managers, it's the nature of the beast. When a team falls short of perceived expectation, the chorus will shout, "Bring me the head of the manager."

The Royals have been remarkably patient in this regard during the Glass family/Allard Baird regime.

Former skipper Tony Muser posted the lowest winning percentage in team history among the ten managers (prior to Pena) who managed at least a season's worth of games. Yet only Dick Howser managed more games in a Royals' uniform.

Through Sunday, Tony Pena has compiled a career .416 winning percentage. That's eight points worse than Muser.

Of course, this only tells us so much. What kind of players has Pena had to work with? Have an inordinate amount of players been injured? Wins and losses only tell us a part of the story when evaluating a manager.

Unfortunately, there is no generally accepted method for measuring a manager's contribution to the bottom line. But that doesn't mean that there aren't indicators that we can look at. Perhaps looking at a variety of categories can yield something close to an empirical truth.

Let's start with five abstract "responsibilities" for a baseball manager. We'll rank the categories in importance, with the most important being worth five times the least important and so on. We'll then select a metric for each area that suggests ability one way or another. Then we'll rank each manager in each category and tally up the rankings. It's like a managerial BCS.

Here are the categories, ranked from most important to least important:

* MEETING EXPECTATIONS: The general manager sets the roster. The roster has a collective performance record that can be analyzed, thus yielding reasonable expectation for victories. It's up to the manager to meet these expectations. DiamondMind Baseball is the consensus king of baseball simulations. Thankfully, the DMB staff publishes their projected standings each spring, so we'll use these to establish our win expectancy.

* MAXIMIZING TALENT: Most would agree that getting the most from the hand you are dealt is a key measure of a manager. To do this, we'll look at how teams have over or under performed in relation to the record they should have posted based on run differential (i.e., the Pythagorean record).

* MATCHING TACTICS: It's a controversial topic but Bill James created a method for evaluating managers based on performance in one-run games. We'll put that to use.

* TEACHING: This is probably the area where Pena gets the most criticism. There is no perfect measure of a team's fundamentals. James has used double plays-to-errors ratio to approach this problem. We'll do the same here.

* RISK MANAGEMENT: How does a manager oversee his baserunners? We'll look at percentage of baserunners squandered, using a Stat Guy method introduced last season.

A couple of qualifications are in order. We'll give Pena a pass for the end of the 2002 season, when he took over mid-stream for Muser. So we're measuring from the beginning of the 2003 season through games on Sunday.

We'll only look at managers who have been active during that entire period. There are 19 such managers. Also, prior to the study, you would expect a natural bias towards teams with better players. We'll see how it plays out.

Expectations

The top of the rankings are largely populated by the most successful teams of the last couple of years. But not entirely. Devil Rays' manager Lou Piniella has consistently outperformed expectations during the period of this study but the Devil Rays have not been anywhere close to .500.

Sure-fire Hall of Famer Bobby Cox is at the top of the list with 26 wins above expectation. As for Pena, the news is not good. Only Colorado manager Clint Hurdle has done worse. The Royals are nine wins below expectation under Pena through Sunday.

ACTUAL WINS VS. EXPECTED WINS

	2003"		2004"		2005"
Manager	EXP	W	EXP	W	EXP	W	NET
B.Cox	86	101	86	96	10	11	26
J.Tracy	82	85	77	93	10	13	22
L.Piniella	54	63	64	70	7	8	16
B.Showalter	73	71	72	89	9	10	16
F.Alou	90	100	87	91	9	8	13
J.Torre	86	101	101	101	11	8	12
T.LaRussa	89	85	92	105	10	12	11
R.Gardenhire	90	90	83	92	9	10	10
L.McClendon	71	75	67	72	7	6	8
D.Baker	81	88	90	89	9	9	6
K.Macha	95	96	92	91	9	9	0
A.Trammell	56	43	61	72	8	7	(3)
F.Robinson	78	83	77	67	9	10	(4)
E.Wedge	73	68	78	80	9	8	(4)
B.Bochy	72	64	83	87	9	8	(5)
D.Miley	77	69	75	76	7	9	(5)
M.Scioscia	91	77	86	92	9	11	(6)
T.Pena	70	83	78	58	7	5	(9)
C.Hurdle	86	74	73	68	7	6	(18)

KEY: EXP - expected wins for season, based on DiamondMind Baseball's preseason projections; W - actual wins; NET - total differential between expected wins and actual wins.

Getting the most out of your talent

There is a surprise winner in this category: Cincinnati's unassuming Dave Miley is 17 wins above the expectations set by the Pythagorean formula. However, most of the managers with good reputations have migrated to the top of the list.

Under Pena, the Royals have under performed by one win based on their run differential. That puts him in the middle of the pack.

MAXIMIZING PERFORMANCE

	2003"		2004"		2005"
Manager	EXP	W	EXP	W	EXP	W	NET
D.Miley	63	69	67	76	7	9	17
J.Torre	96	101	89	101	9	8	16
F.Alou	93	100	88	91	8	8	10
R.Gardenhire	85	90	87	92	10	10	10
J.Tracy	83	85	89	93	11	13	8
K.Macha	94	96	86	91	8	9	8
B.Cox	96	101	95	96	12	11	5
B.Showalter	69	71	87	89	9	10	5
F.Robinson	80	83	67	67	8	10	5
T.LaRussa	88	85	100	105	11	12	3
L.McClendon	76	75	74	72	4	6	(1)
T.Pena	78	83	64	58	5	5	(1)
L.Piniella	68	63	68	70	7	8	(2)
M.Scioscia	80	77	91	92	11	11	(2)
D.Baker	85	88	94	89	10	9	(3)
B.Bochy	66	64	87	87	10	8	(4)
E.Wedge	73	68	81	80	9	8	(7)
C.Hurdle	78	74	73	68	7	6	(10)
A.Trammell	49	43	79	72	10	7	(16)

KEY: EXP - expected wins for season, based on Bill James' Pythagoreum theory, which projects a team's wins and losses based on its run differential; W - actual wins; NET - total differential between expected wins and actual wins.

Performance in one-run games

The full explanation of James' method is here: http://www.diamond-mind.com/articles/james_onerun.htm. In a nutshell, the method calculates the expected won-loss record in one-run games based on overall runs scored and runs allowed. The difference between the expected record and the actual record is our measurement.

Miley wins another category. The Reds' performance in one-run encounters is likely the explanation of why the team has outperformed its run profile in recent years, so that is not really surprising. Some of the managers with reputations as masterful tacticians, like Tony LaRussa, don't fare as well as you'd expect. Pena again posts a negative score in relation to expectation.

Measuring managers by this criterion is questionable. Taken by itself, any conclusion here could be misleading. Much of the success in one-run games is often attributed to strong bullpens. Even more is attributed to luck.

However, when you hear the relative virtues of a manager being discussed, the record in one-run games is usually going to come up. This approach looks at the question rationally and gives us further grist for our mill.

ONE-RUN PERFORMANCE

	2003"		2004"		2005"
Manager	W	EXP	W	EXP	W	EXP	NET
D.Miley	30	23	25	21	6	4	13.8
J.Tracy	26	25	32	25	2	2	8.4
K.Macha	25	24	33	27	3	3	7.2
B.Bochy	21	19	25	21	2	2	6.5
J.Torre	22	20	24	21	3	3	5.9
R.Gardenhire	22	22	24	21	4	4	4.0
L.Piniella	23	24	17	17	4	2	1.5
B.Showalter	17	17	24	22	4	4	1.2
F.Alou	28	21	18	22	1	2	1.2
L.McClendon	24	25	20	22	2	1	-2.4
B.Cox	17	23	27	24	3	3	-2.9
E.Wedge	15	19	26	23	3	5	-2.9
T.LaRussa	14	20	29	27	4	3	-3.3
T.Pena	18	20	14	15	2	3	-3.3
M.Scioscia	16	18	19	21	5	5	-3.9
D.Baker	27	23	19	26	2	3	-4.0
C.Hurdle	17	19	16	19	1	0	-4.7
F.Robinson	22	23	16	21	3	2	-4.9
A.Trammell	19	15	12	19	1	3	-5.5

KEY: EXP - expected wins for season in one-run games, based on a Bill James' formula; W - actual one-run wins; NET - total differential between expected wins and actual wins.

Fundamentals

Errors and double plays - two obvious manifestations of fundamental baseball. But does molding the two into a ratio really say anything about a manager?

By keeping this category in the study, we are, in essence, saying that the manager is responsible for what happens on the field. If the fielders are failing, well, the manager put them there.

At the same time, it must be acknowledged that the year-to-year correlation of double plays-to-errors from team to team is not as high as we'd like. Since we're dealing with relatively rare events, we are at the mercy of sample size limitations. So take the results of this category with a grain of salt.

Our man Miley turns up dead last in this area. Pena is 17th out of 19. So far in 2005, the Royals have turned 13 double plays and committed 20 errors, for a ratio of 0.65. In 2003 and 2004, no team had a ratio less than 1.00.

Our two leaders in double plays-to-errors are a pair of venerable potential Hall of Famers - Felipe Alou and Tony LaRussa.

DOUBLE PLAYS / ERRORS RATIO

	2003	2004	2005	TOTAL
Manager	Ratio	Ratio	Ratio	Ratio
F.Alou	2.04	1.51	1.14	1.703
T.LaRussa	1.79	1.59	2.00	1.692
C.Hurdle	1.42	1.81	2.00	1.615
K.Macha	1.36	1.89	1.56	1.599
F.Robinson	1.49	1.74	1.33	1.596
J.Tracy	1.38	1.99	1.00	1.568
L.McClendon	1.29	1.83	1.50	1.538
B.Showalter	1.79	1.30	1.23	1.500
D.Baker	1.48	1.47	1.69	1.488
R.Gardenhire	1.31	1.56	1.58	1.455
B.Cox	1.37	1.47	2.00	1.448
E.Wedge	1.41	1.43	1.00	1.394
M.Scioscia	1.31	1.40	1.25	1.348
B.Bochy	1.38	1.35	1.00	1.345
J.Torre	1.11	1.49	2.09	1.326
L.Piniella	1.53	1.17	0.94	1.310
T.Pena	1.32	1.29	0.65	1.255
A.Trammell	1.41	1.11	1.15	1.251
D.Miley	1.08	1.09	2.00	1.114

Risk management

In this category, we first estimate the number of baserunners for each team by taking on-base percentage times plate appearances. Then we calculate runners squandered by subtracting runs scored and runners left on base.

What remains are the players who disappeared by means of a caught stealing, a double play or by failing on an attempt to take an extra base. Divide runners squandered by total base runners and you have the percentage of baserunners squandered.

It's debatable whether or not this should be the one-point category but, for the time being, it is. What's interesting is that among the leaders in this statistic are the members of baseball's managerial pantheon. Cox, LaRussa, Piniella, Torre and Alou all rank highly in this area.

The leaders are a pair of managers who are conservative with the running game - Ken Macha and Buck Showalter. Tony Pena is in the middle of the pack, just below the overall average. While Showalter stands head and shoulders above the pack, no one is really too far below the group at the bottom of the rankings.

It's also interesting to note that Ron Gardenhire ranks second-to-last in this statistic. He's obviously led his team to some success anyway.

PERCENTAGE OF BASERUNNERS SQUANDERED

	2003	2004	2005	TOTAL
Manager	SQR	SQR	SQR	SQR
B.Showalter	5.3	4.9	3.8	5.0
K.Macha	6.2	6.7	3.2	6.3
B.Cox	6.7	6.5	9.0	6.7
J.Torre	7.6	7.1	3.3	7.1
T.LaRussa	7.5	7.2	2.9	7.2
F.Alou	7.3	7.3	5.9	7.3
L.Piniella	7.2	7.2	9.0	7.3
D.Miley	7.4	7.3	9.6	7.5
D.Baker	8.0	7.3	5.8	7.6
C.Hurdle	7.0	8.5	7.7	7.7
J.Tracy	8.0	7.7	6.1	7.8
B.Bochy	8.8	7.2	5.6	7.9
T.Pena	7.8	8.2	7.1	8.0
L.McClendon	7.8	8.1	8.1	8.0
M.Scioscia	9.8	8.3	4.6	8.8
F.Robinson	9.1	8.5	8.7	8.8
A.Trammell	10.0	8.2	7.0	9.0
R.Gardenhire	8.8	9.4	10.5	9.2
E.Wedge	9.6	9.2	5.5	9.2

KEY: SQR - estimated percentage of baserunners lost on basepaths

The overall results of our study are indicated by the accompanying chart. A more thorough examination of each area can be found on the Stat Guy Blog, at. THIS IS FOR THE PRINT VERSION

Tally it all up and Tony Pena is tied for 16th out of the 19 managers who have been employed since the beginning of 2003. At the top of the list are have pretty solid reputations - Jim Tracy is on top, followed by Felipe Alou, Joe Torre and Bobby Cox. Ranking fifth is Buck Showalter, who lost out on the Royals' job when Pena was hired in 2002.

It's tough to say whether the list is actually biased towards the better teams, as was anticipated. Showalter comes in fifth despite guiding a team that has not made the playoffs over the span of this study. Also, Mike Scioscia, who has led some very successful squads, ranks near the bottom.

What does all this say about Tony Pena? The answer to that question really depends upon how much stock you put in the criteria used to compile these rankings.

Pena comes up short in across the board - by the numbers. If you believe in the numbers then you're already a member of our aforementioned chorus.

However, if you think these numbers are malarkey, then you must rely on what you see. Fair enough. The question for you, then, is this:

What do your eyes tell you?

Manager rankings

	Manager	Exp	Max	1-run	Fund	Base	Total
1	J.Tracy	90	60	54	28	9	241
2	F.Alou	75	68	33	38	14	228
3	J.Torre	70	72	45	10	16	213
4	B.Cox	95	52	27	18	17	209
5	B.Sh0walter	80	48	36	24	19	207
6	K.Macha	45	56	51	32	18	202
7	R.Gardenhire	60	64	42	20	2	188
8	T.LaRussa	65	40	21	36	15	177
9	L.Piniella	85	28	39	8	13	173
10	D.Miley	20	76	57	2	12	167
11	L.McClendon	55	36	30	26	6	153
12	F.Robinson	35	44	6	30	4	119
13	D.Baker	50	20	12	22	11	115
14	B.Bochy	25	16	48	12	8	109
15	E.Wedge	30	12	24	16	1	83
16	M.Scioscia	15	24	15	14	5	73
16 (tie)	T.Pena	10	32	18	6	7	73
18	C.Hurdle	5	8	9	34	10	66
19	A.Trammell	40	4	3	4	3	54

KEY: The numbers listed in each column represent the score for each manager in that category. Managers are ranked on a 19-18-17, etc. basis in each category. The score for the most important category is multiplied by five, the next most by four and so on. Categories are as follows: Exp - expected wins vs. actual wins; Max - win as predicted by team's run profile vs. actual wins; 1-run - actual wins in one-run games vs. expected wins; Fund - double plays to errors ratio; Base - percentage of baserunners squandered.

NOTE: astute reader Ray from Washington noticed an error in the overall standings. Tony Pena should have had a 6 in DP:E ratio. There was a bad cell in my spreadsheet. I re-checked everything and that appears to be the only faux paus. It's a small error but it does move Pena into a tie for 16th. My thanks go out to Ray.