• Feb
    27

    The Draft: Examining Changes Over Time


    by Kerry Leibowitz

    The Draft:  Examining Changes Over Time

     

    I wasn’t planning on producing another general draft piece, but some of the comments in the message board threads pertaining to the earlier installments raised some interesting questions about the possibility of an improvement in draft success over the years.  I decided to take a quick look at the first round in decade-long segments beginning with the debut of the amateur draft in 1965.  The same player typology explained in the earlier articles in this series was used here. (Since my database was complete through 2002, the final segment is only eight years long rather than ten, but this can be dealt with by focusing on the percentages rather than the raw numbers in the tables below.) 

     

    Amateur Draft, 1965-2002, Entire 1st Round

     

    Span

    N

    0

    1

    2

    3

    4

    % BL

    % Reg

    % Star

    % Cont

    % Non Con

    Avg

    1965-74

    224

    85

    34

    45

    44

    16

    62.1%

    19.6%

    7.1%

    26.8%

    73.2%

    11.8

    1975-84

    262

    91

    40

    63

    55

    13

    65.3%

    21.0%

    5.0%

    26.0%

    74.0%

    13.6

    1985-94

    346

    102

    54

    81

    77

    32

    70.5%

    22.3%

    9.2%

    31.5%

    68.5%

    18.3

    1995-02

    336

    136

    47

    70

    61

    22

    59.5%

    18.2%

    6.5%

    24.7%

    75.3%

    22.1

     

    KEY:

    N = Total 1st Round Selections at Corresponding Slot

    0 = Never Reached Big Leagues

    1 = Cup of Coffee

    2 = Journeyman

    3 = Regular

    4 = Star

    %BL = Percentage Reaching Big Leagues in any Capacity

    %Reg = Percentage Achieving Regular Status

    %Star = Percentage Achieving Star Status

    %Cont = Percentage Achieving Contributor (Regular and Star) Status

    %Non Cont = Percentage Achieving Non-Contributor (not regular or star) Status

    Avg. = Average 1st Round Draft Slot for Given Segment

     

    At first glance, the above table doesn’t seem to show much other than a bit of statistical noise.  Performance appears to be no better in the most recent draft segment than the earlier years.  But the analysis is complicated by the fact that the number of “first round selections” (including supplemental picks) varies greatly from segment to segment.  That’s reflected in the “Average” column.  Note that the average first round selection in the draft’s first decade was (approximately) the 12th; but in the 1995-2002 segment, the average slot was the 22nd.  As Major League Baseball added teams (from 20 in 1965 to 30 by the time the 1997 draft was held) and as supplemental picks were instituted (beginning in 1982) and grew in magnitude, the number of picks per season expanded dramatically.  This “watering down” of opportunity has the potential to mask changes over the years by creating an “apples-and-oranges” evaluation.

     

    In order to produce a more balanced comparison, I restricted the analysis to the first 20 selections of each year’s draft.  That produced the following table:

     

    Amateur Draft, 1965-2002, Selections 1-20

     

    Span

    N

    0

    1

    2

    3

    4

    % BL

    % Reg

    % Star

    % Cont

    % Non Con

    1965-74

    200

    75

    30

    40

    41

    14

    62.5%

    20.5%

    7.0%

    27.5%

    72.5%

    1975-84

    200

    63

    32

    46

    46

    13

    68.5%

    23.0%

    6.5%

    29.5%

    70.5%

    1985-94

    200

    50

    26

    45

    53

    26

    75.0%

    26.5%

    13.0%

    39.5%

    60.5%

    1995-02

    160

    51

    19

    30

    42

    18

    68.1%

    26.3%

    11.3%

    37.5%

    62.5%

     

    We’re beginning to see signs suggesting that drafting has improved over time.  Keep in mind that the numbers for the most recent draft segment are a bit low, as many of the players from the past few seasons reach the prime of their careers.  Were a re-rate done on the final segment five years from now, the percentages in the final segment would likely equal, or slightly surpass, those of the 1985-94 segment.

     

    You’ll recall from the last segment that the cohort including the first four picks of each year’s draft differed significantly from that of the next 18-odd picks.  To see if that difference showed up when broken down by decade, I separated the first four selections of each draft from the following 16.

     

    Amateur Draft, 1965-2002, Selections 1-4

     

    Span

    N

    0

    1

    2

    3

    4

    % BL

    % Reg

    % Star

    % Cont

    % Non Con

    1965-74

    40

    6

    5

    8

    16

    5

    85.0%

    40.0%

    12.5%

    52.5%

    47.5%

    1975-84

    40

    8

    4

    9

    14

    5

    80.0%

    35.0%

    12.5%

    47.5%

    52.5%

    1985-94

    40

    4

    2

    7

    17

    10

    90.0%

    42.5%

    25.0%

    67.5%

    32.5%

    1995-02

    32

    3

    3

    7

    13

    6

    90.6%

    40.6%

    18.8%

    59.4%

    40.6%

     

    Amateur Draft, 1965-2002, Selections 5-20

     

    Span

    N

    0

    1

    2

    3

    4

    % BL

    % Reg

    % Star

    % Cont

    % Non Con

    1965-74

    160

    69

    25

    32

    25

    9

    56.9%

    15.6%

    5.6%

    21.3%

    78.8%

    1975-84

    160

    55

    28

    37

    32

    8

    65.6%

    20.0%

    5.0%

    25.0%

    75.0%

    1985-94

    160

    46

    24

    38

    36

    16

    71.3%

    22.5%

    10.0%

    32.5%

    67.5%

    1995-02

    128

    48

    16

    23

    29

    12

    62.5%

    22.7%

    9.4%

    32.0%

    68.0%

     

    The sample sizes are starting to shrink, but the impact is unmistakable:  drafting, at least in the upper and middle reaches of the first round, definitely yielded more impact players in more recent years than was the case early on in draft history.  The likelihood of generating a contributing level player in general, and a star in particular, has definitely increased; caveats about a relatively small sample apply, of course.  (Again, keep in mind that the final segment numbers are artificially low, across the board, and will almost certainly eventually equal or surpass those of the 1985-94 cohort.)

     

    Whether this trend has plateaued or will continue to increase for choices made over the last decade is impossible to know at this time.  One of the most frustrating things about draft analysis is that it takes many years to paint a complete picture of a draft class.  By the time the full story has been told, 15-odd years have gone by and player evaluation and drafting strategies may have changed completely.

     

    In the next installment, we’ll begin to focus directly on the draft history of the Orioles.


    Comments/Questions?
    Visit the Orioles Hangout Message Board