5 replies to this topic

[philly sox fan]

For a good chunk of the winter I worked on updating and re-writing all of my draft research. I really wanted to post it all before sring training started, and then surely by the end of spring training, and then well... crap the draft is in a little over 2 weeks and I haven't posted much of anything (just the quick draft slot probabilities thing a couple months ago - it's in Geekage now).

I've got a lot of it 90% written and I really want to get it all out by the draft on the 6th, so... I'm going to overload you with old draft reviews and summaries and whatnot whether you like it or not. I need to knock out eight of these single year reviews and then, I think, 4-5 summaries. That's a little more frequently than a new post every other day. I think it's doable, but we'll see.

1987 Draft Review

Part 1: Shape of Draft

One thing that I like to do differently in these kinds of studies is to track draft class productivity on a year-to-year basis instead of just focusing on cumulative career production. In this way we can get an idea about the shape of that productivity. How quickly does a draft class as whole start to pay dividends at the MLB level? What are the peak production years for a draft class?

By answering these kinds of questions we can have more realistic expectations for draft classes and better determine about how long we should wait to grade a draft on the basis of on field productivity (note: subjective grades on the basis of prospect potential will always be made much sooner).

There have been eighteen full seasons since the 1987 draft. For the first table, I’ve designated the draft year as Y0 and each subsequent year as Y1, Y2… Y18. In order to see how player age might shape draft productivity I have also included a simple age progression for HS and C draft picks that assumes all HS players are 18 in Y0 and all college players are 21 in Y0. The 1987 draft is predominantly a college oriented draft so we would probably expect the productivity curve to more closely mimic the college age line.

The following table is a bit dense, but hopefully pretty simple to follow with a brief explanation of terms.

WARP – is the total yearly WARP3 production produced by the entire draft class
%pk – is the individual year WARP3 as a percentage of the single year WARP3 peak

I like “%pk” as a quick way to view the shape of draft productivity because it’s more intuitive than the straight WARP3 totals.

#Pl – is the total number of players who were active in MLB during each year
“3-6” – is the number of players who produced between 3 and 6 WARP3 during each year
“6+” - is the number of players who produced 6 or more WARP3 during each year
Tot – is the number of player who produced 3 or more WARP3 during each year

Although WARP3 production is the most important characteristic of draft shape, “#Pl” tells us the number of players in any given year that are contributing to the draft class production. As we’ll see most players contribute very little to the total production so it’s important to pare that large group down to key contributors. Players in the 3-6 group were at least useful complementary contributors. For a position player a 3 WARP3 season is a pretty low threshold of usefulness, but I wanted a relatively low floor to be as inclusive as possible. Any player with at least 6 WARP3 in a season is a significant contributor to his team.

   Y0  Y1  Y2  Y3  Y4   Y5   Y6   Y7   Y8   Y9  Y10  Y11  Y12  Y13  Y14  Y15  Y16  Y17  Y18
HS 18  19  20  21  22   23   24   25   26   27   28   29   30   31   32   33   34   35   36
 C  21  22  23  24  25   26   27   28   29   30   31   32   33   34   35   36   37   38   39

WARP 2   5  31  71 128  162  200  168  221  201  195  188  172  171  107   78   47   38   33
%pk  1   2  14  32  58   73   90   76  100   91   88   85   78   77   49   35   21   17   15

#Pl  1   9  29  61  85   89   99   93   93   93   76   78   59   59   52   43   26   20   12
3    0   0   3   7  11    5   16   14   19   14   20   12   19   21   13   10    7    7    5
6+   0   0   0   2   5   10    6    6   10   11    8   11    6    7    2    2    0    0    1
Tot  0   0   3   9  16   15   22   20   29   25   28   23   25   28   15   12    7    7    6

As you’d expect draft class production has a basic bell shape with a growth phase, a plateau phase and a decline phase. In general, there is very little productivity from Y0 to Y2, there is then a period of rapid growth between Y2 and Y6 followed by a broad plateau of peak productivity between Y6 and Y10 and then a decline phase beginning in Y11. To some extent all draft classes follow that pattern. College oriented drafts will accelerate more quickly in the growth phase and HS oriented drafts will decline from the plateau phase more slowly. In general, a relatively high Y4 total is a sign of a college draft and a relatively high Y11 total is a sign of a HS draft.

This draft deviates from that pattern in two ways. One, the large dip in Y7 is very unusual. Once a draft is over 90% of peak it stays that way until it begins the decline phase. From a quick look at the players I can’t see any real reason for this dip. More importantly, there is a very slow drop off between Y10 and Y13. Usually draft class production is down to 60% of peak by Y13 as opposed to 77% for this class.

One other general thing to note is that draft class production skews older than you might expect. This draft class was more productive in Y13 (ages HS 31, C 34; 171 WARP3) than it was at Y4 (ages HS 22, C 25; 128 WARP3). That difference is magnified because this draft class was especially good in Y13, but this is generally true. It happens because the growth phase is much steeper than the decline phase.

When I summarized these eight drafts I compared their productivity through the common year Y11. In that comparison this draft ranks a close second to the 1989 draft. However, the slow decline phase actually makes this the most productive draft from this period. As we go through the rest of the review keep in mind that this is a very productive draft.

Part 2: Types of Players

The following table quickly categorizes every player that made the majors by his career WARP3. The WARP3 production is broken down in 10 WARP intervals, ie “0” is the range 0-9.9, “10” is the range 10-19.9, etc. The column “total players” is the total number of players that are in each interval or higher. The last column is the “total players” as a percentage of all the players from the draft class that at least made the majors.

As it turns out the oldest drafts in my study also happen to be the most college oriented drafts. These drafts from the late 1980s are mostly settled in terms of career production of individual players. There aren’t going to be many changes in these rankings from this interim look until all the players are retired. As I move into the 90s and the draft classes become more HS oriented and the players are closer to the middle of their careers, then there will be more movement within the rankings.

                         WARP3     # players    total players    % of players
                           100+          2             2                 1
                            90           0             2                 1
                            80           3             5                 3
                            70           0             5                 3
                            60           3             8                 5
                            50           5            13                 8
                            40           4            17                10
                            30           6            23                14
                            20          11            34                21
                            10          23            57                35   
                             0          87           144                87
                            <0          21           165               100

This is the typical distribution that one would expect. There are a handful of players at the very top of the pyramid and the size of the group within each interval becomes a little larger as we move down. By far the largest groups are for the marginal players with less than 20 WARP3 for their careers. I use that as a general cutoff for a “useful” career. There are players with fewer than 20 WARP3 in their careers who had random good years, but for the most part these players are in the cup of coffee cameo to career backup category. In this draft 165 players made the majors and of those 34 or 21% exceeded 20 WARP3 for their careers and 131 or 79% did not. Again, that is pretty typical. Roughly, 20% of the players that make the majors will produce 80% of the draft class production. Sometimes draft studies will report the number of drafted players per team that made the majors. This distribution shows that those kinds of reports can be deceiving. Teams want to draft a couple 40+ WARP3 players, not a half dozen <20 WARP3 players.

Part 3: Top 40 Players

One of the reasons I like to look at specific draft classes (as opposed to more common multiple year aggregate studies) is to try to get an idea about the number of “good” players that we can expect from an average draft class. As we’ll see forty players takes us well past the “good” players. As a result, we can also get a very good idea of the overall depth for the draft class. These forty players produced 80% of the total WARP3 production to date with an average of 44.3 WARP3/player. The other 125 players that made the majors averaged just 3.6 WARP3/player career. I haven’t checked the subsequent drafts, but I’m sure that is quite typical. There just isn’t much value outside of the Top 40. Some years there isn’t much value outside of the Top 25.

* denotes that the player was active in 2005.

Rank    Rd     Pick     Team       Player           Pos      School       WARP3
  1       1       1       Sea       Ken Griffey*      OF        HS         130.1
  2       1      22       Hou       Craig Biggio*      C         C         115.8
  3       2      47       Cle       Joey Belle        OF         C          88.6
  4       1       9        KC       Kevin Appier     RHP        JC          85.5 
  5      13     324       Bal       Steve Finley*     OF         C          84.7
  6       3      72       Stl       Ray Lankford      OF        JC          67.7
  7       1s     30       Det       Travis Fryman     SS        HS          66.1
  8       7     180       Cin       Reggie Sanders*  SS-OF      JC          60.8
  9      48    1151       NYY       Brad Ausmus*       C        HS          59.5 
 10      58    1225        KC       Jeff Conine*     3B-OF       C          55.4

I’ll go through the round and school distinctions more closely in subsequent sections, but it’s nice to take a quick look in this section just to see how top heavy (by round) and how balanced (by school) the draft class is. This draft has an extremely deep JC class so we end up with almost perfect HS:JC:C balance in the top 10. That’s quite unusual. The four very good to great players were all very high picks. Finley is a big exception obviously, but this is a pretty top heavy draft class.

This draft class should produce two HoF players and a 3rd player with a HoF peak. That’s not uncommon for very good draft classes (remember this is the most productive of the eight in my study), but this top three is better than most. Appier through Sanders are all very good players. Ausmus and Conine are more WARP accumulators than anything else, but they’ve certainly had productive careers.

Rank    Rd     Pick     Team       Player           Pos      School       WARP3
 11       1       5       CWS       Jack McDowell    RHP         C          54.8
 12      32     830       Tex       Robb Nen         RHP        HS          50.6
 13       5     127       Tor       Mike Timlin*     RHP         C          50.2
 14       1      12       Mon       Deline DeShields  SS        HS          49.0
 15      30     781       Hou       Daryl Kile       RHP        CC          45.4
 16       2      33       Sea       Dave Burba       RHP         C          41.3
 17      20     510       Oak       Scott Brosius     3B         C          40.2
 18      18     454       Bal       David Segui       1B         C          39.4
 19       1s     27       Bal       Pete Harnisch    RHP         C          38.6
 20      13     323       Atl       Mike Stanton*    LHP        CC          36.1

In this group of ten we start to see the C ranks stand out and the round distribution is more random. Half the players were high round picks and half were drafted after the 13th rd. At this level we also start to mix in good players with relatively short careers (McDowell, Nen, DeShields) with some accumulators like Timlin and players who were just solid overall contributors (Burba, Brosius). I use 40 WARP3 as a quick, blunt cutoff for a “good” career, but I’m comfortable leaving the case by case determinations much more subjective. Did Burba really have a better career than Harnisch? Are they both examples of “good” players? How about Segui who was mostly known for being constantly injured? Stanton was “good” for a middle reliever, but is that really “good” overall? In general, it’s somewhere in this 11-20th interval that we have to start asking those types of questions. Even good drafts like this one start to run out of steam by the 20th best player.

Rank    Rd     Pick     Team       Player           Pos      School       WARP3
 21       2      39       NYM       Todd Hundley       C        HS          35.5
 22       3      71       Mil       Jaime Navarro    RHP        CC          35.0
 23       6     144        LA       Darrin Fletcher    C         C          33.2
 24       6     140       Cub       Frank Castillo*  RHP        HS          29.5
 25       2      49       Tor       Derek Bell        OF        HS          29.3
 26       5     123       Mil       Steve Sparks     RHP         C          29.3 
 27      13     330       Mil       Troy O’Leary      OF        HS          29.3
 28       7     165       Min       Mark Guthire     LHP         C          28.4
 29       1      16        SF       Mike Remlinger*  LHP         C          27.6
 30       1      13       Mil       Bill Spiers       SS         C          27.0

This range is generally heavy on pitchers. It’s usually filled with back of the rotation starters (Castillo, Sparks) and somewhat successful middle relievers (Guthrie, Remlinger). Hundley was a great player for one year and some of these other players were solid supporting players for several years, but I don’t think too many people would look back on their careers and refer to them as “good” players. One way to think about the “are they good players” question is to ask yourself if you’d be happy if your team’s #1 draft pick had this type of career. Would you be happy if Jacoby Ellsbury had a Troy O’Leary career? Well in terms of expected return on draft slot you ought to be, but you really want your shiny new #1 draft picks to be more than that. Ellsbury is supposed to Damon-lite. If he’s that, then you’ll be happy. I usually start wavering on that question in the bottom half of the previous group of ten. There usually aren’t any who pass that test in this interval.

Rank    Rd     Pick     Team       Player           Pos      School       WARP3
 31       6     146        SD       Dave Hollins      2B         C          26.7
 32       9     230        SF       Gil Heredia      RHP         C          23.9
 33       2      56       NYM       Pete Schourek    LHP        HS          21.8
 34       1      19       Tex       Brian Bohanon    LHP         C          21.7
 35      12     296       CWS       Buddy Groom*     LHP         C          19.3
 36       6     148       Mon       Greg Colbrunn     3B        HS          19.2
 37      14     366       NYY       Gerald Williams*  OF         C          18.7
 38      11     291       Bos       Phil Plantier     3B        HS          18.6
 39      14     354       Oak       Ron Coomer        3B        JC          18.4
 40      15     387       Cin       Butch Henry      LHP        HS          18.2

This is actually a pretty strong group. Hollins was a very good player for a couple of years in Philly before he got hurt. Schourek, Plantier and Henry all had one very, very good season. A few of the other players hung around for a several years and made some positive contributions to their teams.

Note that in the top 10 there are six active players in the 18th full season after this draft and most of them were still pretty productive in 2005. There are just six more active players in the next group of 30 and only Timlin was productive.

Part 4: Top Players Prior to Free Agency

The easiest way to rank players is by their career productivity. However, if we’re trying to tie our rankings into the draft process itself then it also makes sense to focus on just pre-free agency service time production. It’s in those six cost controlled years where teams really reap the advantage of good drafts. Unfortunately, that’s not a particularly easy thing to do. Service time is one of the very few important, common pieces of baseball data that is not easily accessible somewhere on the web.

This past year I found the Cot’s Contracts site (I think originally linked by someone at BTF) and that site does include service time for most active major leaguers. I was planning on eventually putting that data into a spreadsheet, but thankfully I procrastinated long enough until The Hardball Times Annual came out and included that info (with a tip of the hat to Cot’s) in a downloadable link for people who purchased the annual. So thanks to both of those sites for making this difficult task much more palatable.

There are roughly 1200 MLB players from these eight drafts, most of whom are no longer active and therefore wouldn’t be on the Cot’s/THT list. I certainly did not want to figure out yearly service time for 1200 players. Just by restricting myself to players that exceeded 20 WARP3 I knew I could eliminate about 80% of the players without losing many players that are of interest. It turned out that the players with high career WARP3 totals were pretty easy to figure out. Many were still active and most good players don’t have many of the partial seasons that make things really difficult. I’m probably 95% correct for players over 60 WARP3, a little lower for the 40 WARP3 group but still pretty good. The “useful” 20 WARP3 players were a little hit or miss though. These players will have a lot of partial seasons or will be relievers who are hard to gauge just from their playing time. I was accurate enough to include them (partially because the errors on these low WARP3 careers would be small themselves), but there was definitely no point trying to decipher service time for the hundreds of players that didn’t reach 20 WARP3.

The following table is every player that exceeded 20 WARP3 ranked by a good (I think) estimate of their pre-free agency service time production. I also included their career rank and post-FA production so we can get a quick idea of which players may be over or underrated by the career total ranking from Part 3.

I put an * next to the post-FA total of active players.

Rank    Car Rank    Player               Pre-FA      Post-FA
  1         3        Joey Belle            57.2         31.4 
  2         1        Ken Griffey           56.5         73.6*    
   
  3         4        Kevin Appier          47.1         38.4         
  4         7        Travis Fryman         44.2         21.9         
  5         2        Craig Biggio          42.8         73.0*        
  6        11        Jack McDowell         41.5         13.3         

  7         6        Ray Lankford          37.9         29.8         

  8         9        Brad Ausmus           29.7         29.8*        
  9         8        Reggie Sanders        28.7         32.1*        
 10        14        Delino DeShields      28.7         20.3         
 11        21        Todd Hundley          27.8          7.7         
 12        10        Jeff Conine           27.7         27.7*        
 13        25        Derek Bell            27.6          1.7         
 14        22        Jaime Navarro         27.4          7.6         
 15        17        Scott Brosius         27.3         12.9         
 16        12        Robb Nen              27.1         23.5         
 17        31        Dave Hollins          25.4          1.3         

 18        27        Troy O’Leary          24.9          4.4         
 19        26        Steve Sparks          24.3          5.0         
 20         5        Steve Finley          23.3         61.4*        
 21        29        Mike Remlinger        22.9          4.7*        
 22        19        Pete Harnisch         21.8         16.8         
 23        32        Gil Heredia           16.6          7.3          
 24        13        Mike Timlin           16.0         34.2*         
 25        16        Dave Burba            14.9         26.4       
 26        23        Darren Fletcher       14.6         18.6
 27        18        David Segui           13.8         25.6
 28        33        Pete Schourek         13.7          8.1
 29        24        Frank Castillo        13.7         15.8*
 30        28        Mark Guthrie          13.7         14.7         
 31        15        Daryl Kile            13.7         31.7
 32        20        Mike Stanton          13.4         22.7*
 33        30        Bill Spiers           11.8         15.2
 34        34        Brian Bohanon          4.5         17.2

Belle and Griffey were practically instant MVP candidates averaging over 9 WARP3 per season prior to free agency. That kind of production at CBA-restricted wages is by far the most valuable commodity in baseball. Belle didn’t do very well on his first HoF ballot and will apparently never get in, but if you have a preference for high peak (and high value) players Belle was a truly magnificent player.

The next group of players (Appier to Lankford) averaged 6-8 WARP3 per season and contains very good, very valuable players. Note that Fryman and McDowell were roughly as valuable as Biggio in their pre-FA years, but Biggio is (or should be) going to the HoF because he’s still cranking out productive seasons long, long after the other two faded away.

There’s a large gap between Lankford and Ausmus followed by a large group of players with 20+ pre-FA WARP3 production. I think somewhere around 24 or 25 WARP3 (an average of a little over 4 WARP3 per season) is a decent place to draw the line for good players. Hollins who ranks 17th with 25.4 was a very good payer for the early 90s Phillies, but players like O’Leary and Sparks are borderline. Again, this suggests to me that even good drafts start to run out of good players in the 17-20 range.

Any player who’s career rank is much higher than his pre-FA rank was a less valuable commodity to the team that drafted and developed him than his career total suggests. Steve Finley with just 23.3 pre-FA WARP3 compared to 61.4 post-FA WARP3 is the poster boy for that group. We often see relievers like Mike Timlin and Mike Stanton in this group as well. Relievers just don’t pitch enough innings to earn a lot of WARP3 per season so their totals in any small interval are going to be modest, but if they can stay healthy and hang around for many years they can still accumulate impressive career totals.

Conversely, there are players who burn out quickly and as a result are more valuable to the team that drafted and developed them than their career totals suggest. Jack McDowell was extremely valuable to the White Sox with 41.5 pre-FA WARP3 (rank of #6), but his modest post-FA total of 13.3 WARP3 dropped his career ranking to #11. If you skim down the post-FA column for single digit totals (eg Hundley, Bell, Navarro, Hollins, O’Leary) you’ll find a sizable group of these early burnout players. Ideally, the team that reaped the benefit of their valuable solid pre-FA production did not get stuck paying post-FA prices for their declines. Unfortunately, that wasn’t the case with the Sox and O’Leary.

Part 5: Players by Round

I’m eventually going post a more detailed slot by slot analysis, but this is just a quick look at productivity by round. I’ve only listed the rounds that produced a MLB player. I’m reporting the total WARP3 produced by players drafted in each round and the WARP/pick. Note that that is not the WARP per signed player (that information is too difficult to find for later round picks).

I’ve also included the 1987-1994 WARP/pick average for quick comparisons of this draft to an average expected productivity. The “Players” column is simply the number of players that made the majors from each round. The last five columns – “20”, “40”, etc – are the number of players that reached that career WARP3 threshold.

Note that there were only 26 teams in 1987. That means the rounds turned over more quickly than they do now so you can’t necessarily do an exact comparison between round x in 1987 and round x in 2005. This isn’t a big deal for the first few rounds, but it is an issue as you move further into the draft. For example, the slots that comprised round 10 in 1987 are probably in round 8 in 2005.

Round   WARP3    WARP/pk    WARP/pk (87-94)    Players    20   40   60   80   100
  1      595.5      22.9          17.9             20       3    2         1     2   
  1s     124.2      20.7           6.6              5       1         1
  2      252.1       9.7           4.1             19       3    1         1
  3      162.4       6.2           4.6              9       1         1
  4       14.5       0.6           2.8             10
  5      103.0       4.0           3.1              8       1    1
  6      115.1       4.4           3.0              9       3
  7       98.9       3.8           2.5              8       1         1
  8       23.0       0.9           2.3              4
  9       34.6       1.3           1.1              5       1
 10        0.1       0.0           1.1              3
 11       22.2       0.9           1.8              2
 12       18.9       0.7           0.8              3
 13      165.4       6.4           2.0              7       2              1
 14       53.6       2.1           0.7              4
 15       18.2       0.7           0.8              1
 16       20.8       0.8           0.3              4
 17        3.6       0.1           1.5              2
 18       44.2       1.7           1.1              5       1 
 19        9.7       0.4           0.4              3
 20       50.0       1.9           1.3              5            1
 21       12.4       0.5           0.4              3
 22       -0.5       0.0           0.5              1
 23        9.0       0.3           0.5              2
 25        1.5       0.1           0.6              1
 26        1.7       0.1           0.2              1
 27        5.3       0.2           0.1              1
 28        1.6       0.1           0.5              1
 29        9.3       0.4           0.2              1
 30       45.7       1.8           0.7              2            1
 31       -0.7       0.0           0.0              1
 32       50.5       1.9           0.3              2            1
 33       10.6       0.4           0.1              3
 36        1.4       0.1           0.3              1
 37        0.7       0.0           0.1              1
 38       11.8       0.5           0.1              3
 43       11.6       0.4           0.4              1
 45        0.1       0.0           0.3              1
 48       59.5       2.3           0.3              1            1
 53        0.4       0.0           0.0              1 
 58       55.4       2.1           0.3              1            1
 Tot    2217.6

I’m not going to spend a whole lot of time on this table. The obvious point from looking at the WARP/pk (87-94) column is that the expected return per pick drops off very quickly after the first round. There’s a decent little plateau from the 1s to 3rd rds and then maybe another slightly lower one from rds 4-8 and after that the levels are very low and they bounce around that low baseline. This draft was very good in the supplemental round as well and pretty good in rds 2 and 3, but mostly the WARP/pk column follows the typical distribution.

The pattern in the “Players” column is also pretty typical. About twenty draft picks from the first rounds will make the majors (this 2nd rd was unusually high). That number falls to 8-10 players for a few rounds, then drops to 3-5 players from roughly the 8th to 20th rd and then settles in at 0-2 for the rest of the draft.

The one big picture point I do want to make in this section is to look at the clustering of good players at the top of the draft. Everybody knows a handful of very successful low round players (Mike Piazza from the 62nd rd of the 1988 draft is often the example) and because the overall success rate of even very high draft picks is quite low, it’s become an easy cliché to call the draft a crapshoot. That’s not really true. If the draft really was a crapshoot we’d expect to see the good players more randomly sprinkled throughout this table. And that’s not the case. Four of the five players that have exceeded 80 WARP3 were drafted in the first two rounds. Six of the eight players that have exceeded 60 WARP3 were drafted in the first three rounds. That clustering of good players at the top of the draft is the product of the amateur player evaluation methods used by MLB as an industry. I guess we’d call the evaluation practices in 1987 “traditional scouting”, but what ever you call it there is obvious added value from it.

These evaluation methods weren’t perfect. Far from it and there’s certainly plenty of room to experiment and strive for improvement. But people are much too quick to disparage the process that leads to these results without really understanding the draft as a whole. The success rate for high round draft picks is low. That’s a fact. But the primary cause of that fact isn’t that teams foolishly draft too many of X types of players, it’s that there just aren’t very many good players in the draft to begin with. Projecting the career productivity of 18-21 years olds is just really, really hard. Any worthwhile criticism of standard draft practices has to start with that as the baseline. And any deviation from those practices – whether it’s drafting college kids with good translated stats or local HS kids from Georgia or whatever – has to be evaluated with that baseline in mind.

I’ll get off the soapbox, but that is a lovely segue to the next part.

Part 6: Players by School

I assume at least a few people read that last paragraph and went – blah, blah, blah, drafting is hard, but college players are better, right? Some years they are and 1987 is one of them.

In this section we’ll look at the players that were drafted from the three different school types - HS, JC/CC and C. I’m presenting all the data in two forms. Any column with “tot” as a prefix means that every player that made the majors was included. Any column with “20+” as a prefix means that only players that have reached 20 WARP3 in their careers are included. This distinction is not important for the next set of data, but in many cases we’ll see that one subgroup will have a very large edge in sub-20 WARP3 production and that can distort our perceptions of the two groups. For the most part, we’re only interested in players that exceed 20 WARP3.

The first pair of columns lists the cumulative WARP3 production from each school type. The second pair of columns lists the number of players that made the majors from each school type. The final pair of columns lists the percentage of total WARP3 production from each school type. Since the school type debate is really a HS vs C debate I also added in parentheses the HS and C percentages excluding the JC players. That makes for a more direct HS vs C comparison.

      totWARP3     20+WARP3     totPlayers      20+Players    totPercent    20+Percent
HS       653.2       500.7           45               10         29 (36)       30 (38)
JC       393.8       330.5           20                6         18            20
 C       1170.6       826.8          100               18         53 (64)       50 (62)

Almost across the table the C group doubles the HS group. The gap narrows slightly when we just look at the “20+” columns, but this is a strong college oriented draft. The most notable thing from this table though is the strength of the JC group. This is by far the best JC group in my eight year study.

All HS players over 20 WARP3

Rank    Rd     Pick     Team       Player           Pos      School       WARP3
  1       1       1       Sea       Ken Griffey*      OF        HS         130.1
  2       1s     30       Det       Travis Fryman     SS        HS          66.1
  3      48    1151       NYY       Brad Ausmus*       C        HS          59.5 
  4      32     830       Tex       Robb Nen         RHP        HS          50.6
  5       1      12       Mon       Delino DeShields  SS        HS          49.0
  6       2      39       NYM       Todd Hundley       C        HS          35.5
  7       6     140       Cub       Frank Castillo*  RHP        HS          29.5
  8       2      49       Tor       Derek Bell        OF        HS          29.3
  9      13     330       Mil       Troy O’Leary      OF        HS          29.3
 10       2      56       NYM       Pete Schourek    LHP        HS          21.8

The HS ranks produced one future superstar who was available for all of one pick. Fryman, Nen and DeShields were All Star level players. Ausmus has had a long career as a useful accumulator. The rest were useful contributors with some occasional big years (eg Hundley’s 41 HR season in 1996, and Schourek placed 2nd in the Cy Young voting in 1995).

For the most part these players are clustered at the top of the draft. Ausmus and Nen are big exceptions, but four out of the six players to produce at least 30 WARP3 were picked in the first 39 selections.

All JC/CC players over 20 WARP3

Rank    Rd     Pick     Team       Player           Pos      School       WARP3
  1       1       9        KC       Kevin Appier     RHP        JC          85.5
  2       3      72       Stl       Ray Lankford      OF        JC          67.7
  3       7     180       Cin       Reggie Sanders*  SS-OF      JC          60.8
  4      30     781       Hou       Daryl Kile       RHP        CC          45.4
  5      13     323       Atl       Mike Stanton*    LHP        CC          36.1
  6       3      71       Mil       Jaime Navarro    RHP        CC          35.0

One of the reasons this group is so good is that it includes a top 10 overall draft pick. It’s pretty rare for a JC player to be drafted that high and of course Appier was a tremendously productive player. Lankford and Sanders were both very good players with long careers. Not every draft class has one 60+ WARP3 JC position player so it’s quite unexpected that this class has two. Kile signed as an expensive draft and follow in 1988. And again, while that process may be fairly common it certainly doesn’t produce someone as good as Kile every year. There is really a confluence of a lot of unusual things in the JC ranks this year. When we get to the early 90s the JC classes will mostly be pitchers like Stanton and Navarro.

All C players over 20 WARP3

Rank    Rd     Pick     Team       Player           Pos      School       WARP3
  1       1      22       Hou       Craig Biggio*      C         C         115.8
  2       2      47       Cle       Joey Belle        OF         C          88.6
  3      13     324       Bal       Steve Finley*     OF         C          84.7
  4      58    1225        KC       Jeff Conine*     3B-OF       C          55.4
  5       1       5       CWS       Jack McDowell    RHP         C          54.8
  6       5     127       Tor       Mike Timlin*     RHP         C          50.2
  7       2      33       Sea       Dave Burba       RHP         C          41.3
  8      20     510       Oak       Scott Brosius     3B         C          40.2
  9      18     454       Bal       David Segui       1B         C          39.4
 10       1s     27       Bal       Pete Harnisch    RHP         C          38.6
 11       6     144        LA       Darrin Fletcher    C         C          33.2
 12       5     123       Mil       Steve Sparks     RHP         C          29.3 
 13       7     165       Min       Mark Guthire     LHP         C          28.4
 14       1      16        SF       Mike Remlinger*  LHP         C          27.6
 15       1      13       Mil       Bill Spiers       SS         C          27.0
 16       6     146        SD       Dave Hollins      2B         C          26.7
 17       9     230        SF       Gil Heredia      RHP         C          23.9
 18       1      19       Tex       Brian Bohanon    LHP         C          21.7

The college ranks produced two excellent players and a lot of good depth. Depending on your opinion of Segui, Harnish and Fletcher there are 8 to 11 good, solid players and that’s about double what we saw from the HS ranks. The distribution of good players isn’t as tightly clustered although the three best picks in terms of pre-FA production are towards the top of the draft. Biggio and McDowell were both 1st rd picks and Belle was acknowledged as a 1st rd talent with question marks about his attitude. It turned out that those questions about his attitude were both valid and meaningless.

[philly sox fan]

Part 7: Players by Position

In this section we’ll look at the players by their position types. Note that I used the listed positions from the BA Draft Almanac. That doesn’t necessarily mean that the player played this position as an amateur or ended up there in the majors. It’s just the position that the team used when it drafted the player. The only changes that I made were to put a handful of successful position player to pitcher converts with the pitchers.

                P         Pos       C       INF       OF       
totWARP       926.6     1,291.0    267.0    453.0    571.0
20+WARP       619.7     1,038.3    244.0    303.8    490.5

totPlayers      85         80        11      40        29
20+Players      16         18         4       7         7

Position players were more productive than pitchers although slightly more pitchers actually made the majors. That’s typical. The distinction between the “tot” and “20+” groups is always very important in the comparison between pitchers and position players. Only 67% of the total pitching production came from 20+ WARP3 players whereas that figure is 80% for position players. That happens for two reasons. One, there are a lot of pitchers that cycle up to the majors briefly and have semi-useful 10-15 WARP3 careers. These can be middle relievers or starters who put up a couple good years and get hurt and disappear. Two, good to great position players accumulate more WARP3 than their pitching counterparts. As a result, the group of sub-20 WARP3 pitching production is large and percentage of 20+ WARP3 production by position players is quite large.

The pitcher, catcher and outfielder groups are very strong. The infielder group is about average.

All Pitchers over 20 WARP3

Rank    Rd     Pick     Team       Player           Pos      School       WARP3
  1       1       9        KC       Kevin Appier     RHP        JC          85.5 
  2       1       5       CWS       Jack McDowell    RHP         C          54.8
  3      32     830       Tex       Robb Nen         RHP        HS          50.6
  4       5     127       Tor       Mike Timlin*     RHP         C          50.2
  5      30     781       Hou       Daryl Kile       RHP        CC          45.4
  6       2      33       Sea       Dave Burba       RHP         C          41.3
  7       1s     27       Bal       Pete Harnisch    RHP         C          38.6
  8      13     323       Atl       Mike Stanton*    LHP        CC          36.1
  9       3      71       Mil       Jaime Navarro    RHP        CC          35.0
 10       6     140       Cub       Frank Castillo*  RHP        HS          29.5
 11       5     123       Mil       Steve Sparks     RHP         C          29.3
 12       7     165       Min       Mark Guthire     LHP         C          28.4
 13       1      16        SF       Mike Remlinger*  LHP         C          27.6
 14       9     230        SF       Gil Heredia      RHP         C          23.9
 15       2      56       NYM       Pete Schourek    LHP        HS          21.8
 16       1      19       Tex       Brian Bohanon    LHP         C          21.7

This is a strong draft class of pitchers. At least in the years of my study, draft classes max out at about four pretty good starters. Although underrated throughout most of his career, Appier is the second most productive pitcher drafted between 1987-1994 and a rare drafted pitcher that can be characterized as an ace. McDowell and Kile were both at least good #2/3 starters and Burba and Harnisch had some pretty solid years as well. There won’t be too many drafts with 3-5 starters as good as these. MLB as an industry did a good job of identifying these pitchers. Appier and McDowell were taken within the first nine picks and Burba and Harnish were selected close to what would now be the end of the first round. Kile went back to CC and was highly regarded following his next season.

This group includes a good closer (Nen) and a very productive long career setup man (Timlin). There’s also the usual assortment of back of the rotation starters and relievers for depth.

All Catchers over 20 WARP3

Rank    Rd     Pick     Team       Player           Pos      School       WARP3
  1       1      22       Hou       Craig Biggio*      C         C         115.8
  2      48    1151       NYY       Brad Ausmus*       C        HS          59.5 
  3       2      39       NYM       Todd Hundley       C        HS          35.5
  4       6     144        LA       Darrin Fletcher    C         C          33.2

Biggio’s career total wouldn’t be nearly that high if he had stayed at catcher, of course. Still, four regulars with one an eventual HoFer after a position switch is a very strong group of catchers for a draft class. It’s pretty amazing to think that two amateur catchers drafted 1100+ picks apart in 1987 were both starters on the 2005 NL pennant winner.

All Infielders over 20 WARP3

Rank    Rd     Pick     Team       Player           Pos      School       WARP3
  1       1s     30       Det       Travis Fryman     SS        HS          66.1
  2      58    1225        KC       Jeff Conine*     3B-OF       C          55.4
  3       1      12       Mon       Deline DeShields  SS        HS          49.0
  4      20     510       Oak       Scott Brosius     3B         C          40.2
  5      18     454       Bal       David Segui       1B         C          39.4
  6       1      13       Mil       Bill Spiers       SS         C          27.0
  7       6     146        SD       Dave Hollins      2B         C          26.7

This is roughly an average class of infielders. It’s missing a superstar, but the depth is good. Both HS players (and arguably the two best players) were drafted within the first 30 picks. The C players are more randomly distributed. There will be exceptions, but that’s generally true. Good HS infielders are clustered at the top of the draft, but there is a persistent class of college infielders with Brosius-like good, solid careers that are drafted in the 5th to 20th rd.

All Outfielders over 20 WARP3

Rank    Rd     Pick     Team       Player           Pos      School       WARP3
  1       1       1       Sea       Ken Griffey*      OF        HS         130.1
  2       2      47       Cle       Joey Belle        OF         C          88.6
  3      13     324       Bal       Steve Finley*     OF         C          84.7
  4       3      72       Stl       Ray Lankford      OF        JC          67.7
  5       7     180       Cin       Reggie Sanders*  SS-OF      JC          60.8
  6       2      49       Tor       Derek Bell        OF        HS          29.3
  7      13     330       Mil       Troy O’Leary      OF        HS          29.3

Griffey and Belle make this a very strong group at the top and the pair of unusually good JC outfielders gives it good depth.

Part 8: Players by Team

This final table ranks the teams by cumulative WARP3 production. That can be a little misleading because a team with three decent players could rank higher than a team with one very good player. I’ve listed each team’s key players to provide a fuller picture than just the cumulative WARP3 number. The most important thing to note is how rare it is to get more than one good player from a draft. A good draft is really any draft that produces one good player.

* next to the team denotes a player that exceeded 40 WARP3

Rank    Team    WARP3      Key Players (WARP3)
  1      Bal     181.2      Finley (84.7), Segui (39.4), Harnisch (38.6)
  2      Sea     178.1      Griffey (130.1), Burba (41.3)
  3      Hou     167.9      Biggio (115.8), Kile (45.4)
  4       KC     151.0      Appier (85.5), Conine (55.4)
  5      Mil     123.8      Navarro (35.0), Sparks (29.3), O’Leary (29.3), Spiers (27.0)
  6      Tex     117.3      Nen (50.6), Bohanon (21.7)
  7      Mon     107.2      Deshields (49.0)
  8      Cle     106.5      Belle (88.6)
  9      Cin     102.7      Sanders (60.8)
 10      Tor      97.0      Timlin (50.2), Bell (29.3)
 11      NYM      91.4      Hundley (35.5), Schourek (21.8)
 12      Stl      87.7      Lankford (67.7)
 13      NYY      81.1      Ausmus (59.5) 
 14      Det      80.6      Fryman (66.1)
 15       SF      77.9      Remlinger (27.6), Heredia (23.9)
 16      CWS      77.4      McDowell (54.8)
 17      Oak      71.3      Brosius (40.2)
 18      Atl      63.6      Stanton (36.1)
 19       LA      61.9      Fletcher (33.2)
 20      Cub      61.6      Castillo (29.5)
 21      Min      49.6      Guthrie (28.4)
 22       SD      29.0      Hollins (26.7)
 23      Bos      24.3      Plantier (18.6), Zupcic (3.7)
 24      Phl      11.2      Borland (5.3), Trlicek (2.3)
 25      Pit       9.5      Chamberlain (7.3)
 26      Cal       6.8      Orton (3.6), Amaro (3.6)

The Orioles had a very nice, deep draft, but their three solid players aren’t more valuable than the Seattle and Houston HoF position player plus solid starter combo. That’s the problem with a cumulative ranking.

Cleveland ranks just 8th despite drafting the single most productive player in terms of pre-FA WARP3. Their ranking is hurt because they got so little out of the rest of the draft, but that’s obviously an excellent draft. It was also an extremely risky one. As I mentioned, Belle was considered a first round talent, but he had already had a number of off-field incidents and most teams shied away from him. (Topical parenthetical expression – he was kind of like Delmon Young without the thrown bat at an umpire.) The Indians did not have a 1st rd pick and when he was still on the board at #47 in the 2nd rd, they jumped at the opportunity to get a 1st rd talent. Given what we know about the success rates of draft picks after the 2nd rd, the Indians really rolled the dice and basically put their whole draft on Belle. It worked out spectacularly, but even if it didn’t I think that was a good choice. In my opinion, too many teams these days try to minimize their risk in the draft when they should be looking to maximize the reward. That doesn’t mean teams should draft every athletic kid they can find and hope they can teach him to play baseball. Every team has to find a good risk/reward balance for their particular situation, but every draft philosophy is going to be a failure in an absolute sense because there just aren’t that many good players. Given that, teams should always be on the lookout for ways to maximize their draft reward.

On my cumulative scale the Brewers rank 5th. It is extremely rare to see a team with four players that exceeded 20 WARP3. I guess in some sense they were by far the most successful at minimizing draft risk. But none of their successful players made much of a difference. The Blue Jays under JP Riccardi have had drafts that may end up producing this kind of “success” – a bunch of useful players with maybe a couple good ones thrown in – but I don’t think that’s the right approach to the draft. This shouldn’t be the kind of draft success that teams try to achieve.

If I were to forget about the cumulative numerical ranking and redo the table it would look something like the following.

Great drafts (4)

Rank    Team    WARP3      Key Players (WARP3)
  2      Sea     178.1      Griffey (130.1), Burba (41.3)
  3      Hou     167.9      Biggio (115.8), Kile (45.4)
  8      Cle     106.5      Belle (88.6)
  4       KC     151.0      Appier (85.5), Conine (55.4)

Very good drafts (7)

Rank    Team    WARP3      Key Players (WARP3)
  1      Bal     181.2      Finley (84.7), Segui (39.4), Harnisch (38.6)
 16      CWS      77.4      McDowell (54.8)
 14      Det      80.6      Fryman (66.1)
 12      Stl      87.7      Lankford (67.7)
  9      Cin     102.7      Sanders (60.8)
  6      Tex     117.3      Nen (50.6), Bohanon (21.7)
  7      Mon     107.2      Deshields (49.0)

Good, above average drafts (4)

Rank    Team    WARP3      Key Players (WARP3)
 17      Oak      71.3      Brosius (40.2)
 10      Tor      97.0      Timlin (50.2), Bell (29.3)
 13      NYY      81.1      Ausmus (59.5)
  5      Mil     123.8      Navarro (35.0), Sparks (29.3), O’Leary (29.3), Spiers (27.0)
 11      NYM      91.4      Hundley (35.5), Schourek (21.8)

Mediocre to awful drafts (10)

Rank    Team    WARP3      Key Players (WARP3)
 15       SF      77.9      Remlinger (27.6), Heredia (23.9)
 18      Atl      63.6      Stanton (36.1)
 19       LA      61.9      Fletcher (33.2)
 20      Cub      61.6      Castillo (29.5)
 21      Min      49.6      Guthrie (28.4)
 22       SD      29.0      Hollins (26.7)
 23      Bos      24.3      Plantier (18.6), Zupcic (3.7)
 24      Phl      11.2      Borland (5.3), Trlicek (2.3)
 25      Pit       9.5      Chamberlain (7.3)
 26      Cal       6.8      Orton (3.6), Amaro (3.6)

At most there will be 4 or 5 teams that had great drafts and those drafts consist of a single great player or two very good ones. Some years there may be only 1 or 2 great drafts.

Then there may be a half dozen or so teams that had very good drafts and those are teams that drafted a single very good player or a couple pretty good ones.

Then there may be a handful of decent, above average drafts. These teams end up with one low 40 WARP3 player or maybe a long career accumulator type or a few useful players.

In thin drafts those two groups will be smaller and less distinct.

It’s usually true that about half the teams will fall into one of those categories and the other half of the league will be in the mediocre to awful range. That may sound harsh, but if there is only 15-20 “good” players in any given draft class, then there just aren’t enough to go around. And any team that manages to draft more than one of them just increases the number of teams that will be shut out.

[Worst Trade Evah]

Stunning work, philly. Really, really interesting. If I owned a team, I'd be hiring you. The service time distinction is a crucial bit, but one a lot of other people overlook.

This is probably in there and I just need to digest it further, but how can teams draft better? Just better scouting (which is already adding value)? More aggessive risk-taking on high upside players? Willingness to pay big bonuses to top prospects?

Edited by Worst Trade Evah, 19 May 2006 - 08:19 AM.

[philly sox fan]

This is probably in there and I just need to digest it further, but how can teams draft better? Just better scouting (which is already adding value)? More aggessive risk-taking on high upside players? Willingness to pay big bonuses to top prospects?

<{POST_SNAPBACK}>

I'm going to try to get into that in the summary posts, but first you've got to wade thru another six single year reviews. I think taking bigger risks and spreading around extra money is an important part of it, but I'm not sure you get better at picking between similar highly ranked players. It's difficult to see just how retrospective studies can help a team pick Conor Jackson over David Murphy. Aside from better scouting and luck, I'm not sure what else there is.

I'll spend some time picking out mid-round sleeper types, but whatever lessons there are to be learned from the 1987-1994 drafts may not be all that applicable to the 2006 draft.

[saintnick912]

Re: players that have more value to teams post-FA than in their first 6 years

We often see relievers like Mike Timlin and Mike Stanton in this group as well. Relievers just don’t pitch enough innings to earn a lot of WARP3 per season so their totals in any small interval are going to be modest, but if they can stay healthy and hang around for many years they can still accumulate impressive career totals.

This seems like a solid argument for why the Sox may be trying to convert Hansen to a starter in Pawtucket. Due to the structure of his contract, they'll only control him for 4 more years (does '05 count, if not 5 more years) before he hits FA. In order to get the most value out of him before that time, it is in their interest to convert him to a starter and hope for a McDowell like career path rather than a Timlin.

edit: Thanks for all your work on this Philly, I really enjoy reading these studies as the MLB draft is such a mystery compared to the NFL or NBA (to me at least).

Edited by saintnick912, 20 May 2006 - 01:32 PM.

[Bowlerman9]

Re: players that have more value to teams post-FA than in their first 6 years
This seems like a solid argument for why the Sox may be trying to convert Hansen to a starter in Pawtucket.  Due to the structure of his contract, they'll only control him for 4 more years (does '05 count, if not 5 more years) before he hits FA.  In order to get the most value out of him before that time, it is in their interest to convert him to a starter and hope for a McDowell like career path rather than a Timlin.

edit: Thanks for all your work on this Philly, I really enjoy reading these studies as the MLB draft is such a mystery compared to the NFL or NBA (to me at least).

<{POST_SNAPBACK}>

All players are under "control" for their first 6 MLB seasons. He is under contract until the end of 2008. At that point, the sox will own his rights until he has been in the league a total of 6 years.