Managers: Does age matter?
by Kirk Allen
July 1, 2003
The idea of analyzing how age affects managers first came to me after Felipe Alou was hired to manage the Giants
this year at the ripe old age of 68 (some think he may be older). The concept was further promoted when someone
asked Rob Neyer his opinion of the idea in a chat. I decided the best way to learn anything would be to look at the
numbers.
Using baseball-reference.com, I looked at every manager who has ever managed in the AL or NL (i.e., Federal League
not included) except for seasons in which a manager only worked a single game (e.g., Ted Turner once managed a game for the Braves).
I also only took the seasons since 1900 even for managers whose careers straddle the turn of the century.
This gives a data set of 433 managers, who range in age from 23 (Roger Peckinpaugh, 1914)
to 88 (Connie Mack, 1950). Every age except 24 is represented. For age, I subtracted the manager's birth year from the
year of the season (e.g., Alou born in 1935: 2003 - 1935 = 68)
rather than looking at the manager's "baseball age" of June or July 1. This was simply done to
ease the calculations.
I will spare myself the torture of trying to come up with many explanations
at this time and simply run some of the numbers..
Cumulative Records of Managers by Age
Age Total W Total L Over/Under 500 Total %
20's 1858 1743 +115 51.6
30's 29548 29699 -151 49.9
40's 68906 69217 -311 49.9
50's 44941 44693 +248 50.1
60's 10250 9514 +736 51.9
70's 944 1349 -405 41.2
80's 572 810 -238 41.4
First, those of you who are sticklers for details may calculate that there are 5 more losses than wins.
I believe this is due to not counting single-game managers and also could be a couple typos.
The youngsters (age 20's) are virtually all player-managers and from the early 1900's. No one in his 20's has
managed since Lou Boudreau in the 1940's. The next 3 groups (30's through 50's) are about as close to 50% as you
can expect, and they also represent the bulk of the total games managed. The 60's have the best overall winning percentage
and the most games over 500. I suspect that most bad managers have already been weeded out by the time they are 60.
The 70's are made up almost entirely of Casey Stengel with the early Mets and Connie Mack with some really
bad Philadelphia A's teams. The 80's group is only Connie Mack.
For the final stage of the study, I attempted to determine if managers get better with age. I looked only at managers
who managed at least 1000 games. I compared each person's winning percentage during his 30's through 60's (insufficient data
for the extreme ages). I only considered the gain/loss in percentage for people who managed at least 2 seasons in each
of his age decades (e.g., must have at least 2 seasons during his 30's and 40's to calculate the 40's % - 30's %).
Age Range Number Increasing Number Decreasing
30's to 40's 16 25
40's to 50's 25 39
50's to 60's 10 10
Cumulative Records of Managers by Age (minimum 1000 games)
Age Total W Total L Over/Under 500 Total % % of Age Group's Games
20's 1324 1084 +240 55.0 66.9
30's 16693 15493 +1200 51.9 54.3
40's 43886 41116 +2770 51.6 61.5
50's 32245 30372 +1873 51.5 69.9
60's 8978 7875 +1103 53.3 85.3
70's 944 1349 -405 41.2 100.0
80's 572 810 -238 41.4 100.0
A little help interpreting: Of the managers who qualify (mentioned above), 17 had a higher winning percentage in their
40's compared to 30's, while 26 had a lower winning percentage in their 40's compared to 30's.
I interpret this chart as meaning that more people decline with age, regardless of what range you're looking at. Perhaps
it's boredom or failure to adapt to changes in the game or the natural process of a manager aging along
with his core players. These managers are then shown the door, and the group aging from their 50's into 60's is evenly
split between improving or declining.
Further, when you limit the population to those with 1000 games, you see the chart shift "up" by about 1.5% (except
for the aforementioned Mack & Stengel teams).
This agrees with my earlier observation that only the best managers stick around into their 60's and beyond. I also
noted that as age increases, a higher percentage of the age group is occupied by these long-time managers. This group
consists of 99 managers, 61 of whom are above 500 (61%). This compares with only 151 of 433 total managers (35%).
(Note: Only 6 are exactly 500, and this does not change the percent.)
To sum up the obvious.. Better managers manage for longer. Big shock, I know. There is some interesting information
out there if you're willing to take the time to look. I will continue to update this work at the end of each season
and hope to be able to draw some more useful conclusions at a later date.
copyright 2003, Kirk Allen