Does the world really need another series of articles centering on baseball's amateur draft?  Need is a strong word.  "Could use" is, perhaps, a more appropriate phrasing.  The forthcoming series of articles, for which this piece will serve as an introduction, was inspired by a very quick and dirty listing-cum-column that I posted in July entitled "Catching Up With the Drafted Pitchers."  That piece was little more than a thumbnail summary of the pitchers drafted by the Orioles in the first three rounds of the draft beginning in 1990 with predictably awful-looking results.  But the limited view begged the obvious question:  how does the O's record compare with other franchises?

I decided to put together a comprehensive picture of the record of every team in baseball for the first three rounds of the draft since 1990 and see where that took me.  It took several weeks of on-again/off-again work to complete a dump and parsing of the relevant data, but I finally finished that work last week and started to sift through the material in qausi-analytic fashion.

What the Series Will Cover

Questions I plan to examine in the forthcoming series include, but are not limited to:

• What can generally be expected from draft picks in the first three rounds?

• What impact, if any, does draft pick slotting have on anticipated success/failure?

• What sort of impact do sandwich picks typically have and how should this impact a club's decision making process with regard to impending free agents?

• Are pitchers more or less likely to pan out long-term compared with position players?

• Do franchises have identifiably different drafting strategies and, if so, how do they compare with one another?

• Have there been significantly different levels of drafting success among franchises over this period of time?

The Rating System, etc.

Having laid the groundwork for what's to come, I want to take a few moments to establish some benchmarks and briefly discuss the rating system that underlies the entire enterprise.

Approximately four years ago I embarked on a study of the Orioles' drafting during the tenure of former scouting director Tony DeMacio.  To evaluate DeMacio's performance it was necessary to place the O's rate of success in some kind of historical context.  There's no point in reinventing the wheel; the work that was done to underpin that series can largely be re-used here.  The second article in the series laid out the player rating system that I will use again.  If you're unfamiliar with the general outline of the system, I urge you to take a look at that piece.  But even if you are familiar with the basics of the system, there are a few additional points that I feel the need to make.

1) The individual player ratings, while directed by basic guidelines, are ultimately subjective in nature.  I considered using an objective set of criteria a few years ago, and I revisited the decision not to go that route this time around.  On both occasions I arrived at the same conclusion:  an objective system wasn't worth the time or trouble it would take to produce an improvement over a carefully implemented subjective system.  While this means that there will almost certainly be disagreements over specific ratings in some cases, it's probably better to have a disagreement over a player here and there than over a standard by which an objective system is designed.  

2) Ratings, while static for players whose careers are over, are dynamic for many players whose careers are ongoing.  General standards can be fairly objectively applied to retired players.  For instance, the basic guideline for players is five years of minimum play at a relevant level.  So, to be a "contributing level" position player, someone typically has to have had at least five years as a full-time starter at a given position at the major league level to be so considered.  But should that standard be applied, say, to someone whose major league career hasn't yet spanned five years?  By that reasoning, Nick Markakis wouldn't qualify as even a contributing level player.  That's pretty silly.  So, latitude had to be granted.  Some sense of prognostication had to be made.  While it's certainly possible that Markakis, for example, won't ultimately have five years as a starting big league player, it's highly likely that he will.  That type of consideration was made in rating players.

As an extension of the point of the fluidity of ratings of active players, some individuals who might have been perceived as star players when rated at one point in time might not get that rating when their entire career was considered as a whole.  For instance, when I rated players using this system in 2004, I'm almost certain that I rated Barry Zito as a star.  At that time Zito had finished three undeniably star-caliber seasons and seemed a virtual lock to reach the requisite (by this rating system) five seasons of star-caliber play to be regarded as a career star.  Four years later, it's debatable whether Zito has achieved those five seasons--maybe he has, maybe he hasn't--and with his career in a clear downward spiral, it's now probably unlikely to expect Zito to post any additional star seasons.  Does Zito merit a star rating today?  Maybe.  The point is, a rating is only as static as a player's career.

3) Player rating categories describe a minimum level of career achievement, not necessarily the performance of that player during the current season.  For example, Jorge Posada is having an injury-plagued, generally ineffective season in 2008 and given his age and the wear and tear on his body as a catcher he may never have another meaningful season as a big leaguer again.  Does that mean that, rating him in 2008, he shouldn't be classified as a star despite obviously being one of the top performers at his position of his time?  By this rating system, the answer is no--Posada is rated as a star even if he never plays another big league game, on the strength of at least eight--and as many as ten--star caliber seasons, far more than the guideline of five. 

(Players in their first year as major leaguers in 2008 were not, under any circumstances, given "star" ratings, regardless of their performance.  First year players who were performing like stars--Evan Longoria, for instance--were rated as "contributing level" players.  Part of one season does not a trend make.  While projecting a five-year career as a starter is one thing, a five-year career as a star is another entirely.  Young players well into a second season of star-caliber play (e.g. Ryan Braun) were given the benefit of the doubt.)

4) On occasion, guidelines were bent in the interest of "justice."  Some players who, for instance, never quite achieved the minimum guideline of five years of play as a starter were given the benefit of the doubt for career duration and a certain level of significant achievement.  For instance, Matt Stairs has qualified for only three batting titles in his career and appeared in 130+ games only four times.  But with Stairs on pace this season to eclipse 300 at bats for the ninth time, appear in 100+ games for a 12th consecutive season, homer more than 250 times in his career and post an OPS of nearly .850 in a big league career that has spanned 17 years (so far) it seemed patently unfair--not to mention inaccurate--to relegate him to the journeyman category.  Stairs was rated as a "starting player" for purposes of this analysis.  Examples of this sort are quite rare.

5) Having a star season or two isn't the same thing as being a star player.  Nor does a season or two as a regular necessarily make you a regular in a career sense.  For example, Mike Caruso was a starting big league shortstop at ages 21 and 22.  The rest of his big league career consisted of 20 additional ABs when he was 25.  Starting caliber player?  No.  Hank Blalock had a couple of star caliber seasons at ages 22 and 23; he followed that up with a pair of pedestrian seasons--as a starting player, however--and has had two injury-riddled seasons since then.  Star caliber player?  No.

6)  There are stars and there are stars.  One of the weak points of a system with as few categories as this rating system contains is that granularity is lost.  As a result, players of palpably different impact may be lumped together.  For instance, Alex Rodriguez and Paul Konerko both garnered a "star" rating under this system.  Now, Paul Konerko has been a fine ballplayer--a three-time All-Star (arguably should be at least a five-timer) who, despite an injury-riddled 2008, still has an excellent chance of eclipsing the 400-mark in career home runs, Konerko's arguably had eight star-caliber seasons in his career.  I don't think there's much question that he qualifies as a "star" under this rating system.  But it's pretty obvious--or ought to be--that he doesn't hold a candle to Rodriguez.

The same problem rears its head with regard to "contributing" level players.  Chad Qualls is now deep into his fourth consecutive season as a solid, contributing reliever, but is there any way that he has the same impact as, say, Jon Garland (now well into his seventh straight year as a solid starting pitcher)?

These shortcoming are accepted for the sake of, on the one hand, ease of administration and, on the other hand, to lessen the problems of limited sample size, particularly as it pertains to team-by-team comparisons.  The number of cases in each cell are pitifully small as things currently stand; further parsing of player rating categories would make for an almost meaningless display of data.

Why the First Three Rounds Only?

Two words:  manageable workload.  Ideally, I'd have gone through every round of the draft, but the data dump for the first three rounds of these 19 draft sequences alone amounted to nearly 2000 players.  Tack another 2350 players per year onto the task and you have...well, you have nothing, because I wouldn't have taken the time to do it at all.  And the truth is, there's a tremendous diminishing rate of analytical return associated with each additional round.  As I showed in the series four years ago, there's approximately a 3% chance of a fourth round player becoming something more than a major league journeyman.  Only about 1/5 of players drafted in the fourth round reach the big leagues at all.  Consider that for a moment:  in the average draft, only six of the players selected in the fourth round can be expected to reach the major leagues; only one can be expected to become a player of significance.  And, of course, the numbers decline with each succeeding round.  

During the period of this study (1990-2008), for instance, the Orioles have had exactly two fourth round picks reach the major leagues:  1992 selection pitcher Vaughn Eshelman (a big league journeyman, with 83 appearances, 30 starts, over three seasons with the Red Sox) and 2003 choice, pitcher Bob McCrory who had a very unsatisfactory cup of coffee with the O's this season.  Some subsequent fourth round picks will surely reach the big leagues; one or two eventually will become players of consequence; but you get the point.  (In fact, since the draft began in 1965, only six O's fourth round selections have reached the big leagues; Jesse Jefferson (1968) came the closest to having a meaningful career, but is still best regarded as a journeyman; the others were Mike Darr (1974), who managed 1 1/3 innings in the big leagues; catcher Carl Nichols (1980), who accrued roughly 200 major league plate appearances over parts of six seasons; lefty Brian DuBois (1985) whose career was derailed by a series of arm injuries after 18 big league appearances; right-hander Chuck Ricci (seven major league innings pitched), Eshelman and McCrory.)

For the sake of comparison, the Red Sox fourth round experience is a bit better.  Boston hit the jackpot with Jonathan Papelbon in 2003's fourth round, but only two other fourth rounders have reached the big leagues--both of them "cup of coffee" types (outfielder John Barnes in 1996 and shortstop Angel Santos in 1997).  Only three Yankees' fourth round draftees during this time frame have reached the major leagues:  journeyman left-hander Kirt Ojala in 1990, journeyman right-hander Mike Buddie in 1992 and cup of coffee right-hander Matt Smith in 2000.  The Blue Jays have four fourth round selections who have reached the majors:  cup of coffee position players Howard Battle (1990) and Tom Evans (1992); cup of coffee right-handed pitcher Adam Peterson (2002) and journeyman Casey Janssen (2004).  Janssen appeared to be on his way to contributor status (at least) prior to tearing his labrum in spring training this year, putting his career--and ultimate rating--in serious doubt.  The Rays have had only two fourth rounders reach the bigs:  cup of coffee lefty Todd Belitz in 1997 and right-hander David Bush in 2002.  Bush has become a reliable back-of-the-rotation starter for the Brewers and nets a "contributor" rating, thus becoming only the second player drafted by a current American League East team in the fourth round since 1990 to become a major league career player of consequence.

The truth of the matter is, a simple focus on the first round (including supplemental picks) would probably be attention enough. Expanding the analysis to the first three rounds, however, provides a larger pool of players per team to examine, providing something of a critical minimum mass, without providing an overly burdensome task.  Historically, roughly half of the star and contributing level players to come out of the draft have been selected in the first three rounds despite representing only 7-8% of the selections.

There are, of course, success stories from later rounds of the draft, relatively few though they may be.  These, however, are flukes--happy accidents if you will, and in the aggregate say nothing about a team's systemic ability to assess and/or develop talent other than its facility for luck.

Looking Ahead

In the next installment we'll jump into the numbers and take a look at how the teams stack up against one another.

Back to the Belfry

Discuss this article on the Belfry message board