Setting a pace at which to propel one’s minor-league assets through the talent pipeline is among the most challenging tasks facing any player-development program. Each prospect’s route to the big club is fraught with potential hazards, as evidenced by both TNSTAAPP and the name of Jason Parkscolumn; by acting injudiciously, front-office handlers can inadvertently place additional obstacles in a young player's path. Grant promotions too early or too often, and they risk jeopardizing his future by burying him on the bench or subjecting him to the mental and physical rigors of major-league life before he’s equipped to handle them. Delay advancement too long, and they threaten to sabotage his development in a different way, blunting his talents against inferior competition while more expensive players with shorter shelf lives take up space on the big-league roster.

Faced with this uncertainty, each club must construct its own approach to navigating the pitfalls of promotion. This should be a scientific process, but there’s an element of art to it; in addition to deciding whether a player’s on-field performance merits a ticket to a higher level, the arbiters of a minor leaguer’s fate must also judge whether he has the necessary maturity to take a bigger stage in stride, and whether his physical gifts are such that a stiffer challenge might unlock his potential, rather than bury it beneath a residue of self-defeat.

The result is that some teams develop a reputation for especially aggressive or conservative tendencies in promoting their players. When I asked Kevin Goldstein to name an organization known for being conservative in this respect, he immediately mentioned the Twins, confirming my own impression. Kevin placed the Marlins on the opposite end of the spectrum, and we both agreed on the Tigers, at least where young pitchers are concerned. The Rays, on the other hand, are known for holding their hurlers to a slow and steady slog through the system.

With that in mind, I thought I’d take a look at what we might uncover by examining some data on major-league call-ups. I decided to limit the call-up window to 2005-2009, excluding 2010 both because of some mapping issues that were causing a few of last season’s promotions to slip through the cracks (I vaguely remember there having been some sort of fuss about a “Strasburg” guy) and because I wanted to be able to examine two years of performance by each of the promoted players. I could have gone back further—in a five-year period, we’re talking about an average of only 34 call-ups per team—but the deeper into the past one looks, the less likely any particular call-up is to have any bearing on the current team’s tendencies: it would be difficult to draw a connection between, say, the Astros’ call-up of Adam Everett in 2001 and their promotion of Brett Wallace last season, given the turnover in their front office over that span.

From 2005-2009, the average player making his major-league debut was 24.4 years old (taking the average of their seasonal ages). The average position player had 2070 minor-league plate appearances under his belt before he first stepped in against a major-league hurler; the average pitcher (without distinguishing between starters and relievers) had thrown 391 minor-league frames before getting his first crack at The Show. Sorting by position, we see the following breakdown in debut age:

Things appear much as we would have expected at each end of the spectrum. As Nate Silver observed several years ago, catchers take a long time to get ready, and shortstops tend to derive much of their value from speed and contact hitting, abilities that peak early. As a result, the more athletic infielders make it to the majors over a year earlier than their decidedly less mobile counterparts behind the plate.

Here’s how the experience stats break down by team:

As expected, the Twins led the pack in being slow to the bigs. Twins position-player call-ups accumulated an average of 2617 minor-league PA before making it to the Metrodome, roughly 200 more than the next-highest organization (the Angels) and a full season’s worth of action above average.

Since some evidence suggests that major-league pitchers don’t improve with age or experience, it might seem that minor-league hurlers shouldn’t be allowed to waste their bullets on bush leaguers after they’ve been deemed capable of big-league service, as Steven Goldman argued elsewhere yesterday. The following graph reveals the organizational discrepancies in minor-league seasoning for pitchers:

Although David Price made it to the majors in a mere 110 innings, the Rays were indeed the team most cautious with its pitchers, as James Shields, Jason Hammel, and Wade Davis, among others, did extensive work on the farm before reaching Tampa Bay. The Mets were the most aggressive club when it came to promoting pitchers as well as hitters—and we’re not even counting Jenrry Mejia. They also broke in the fewest players over this five-year period, minting just 23 major leaguers. The most rapidly multiplying team, the Giants, doubled that total.

We’ve looked at how quickly each team calls up its prospects, but what about their performance once they’ve arrived? Here are the WARP scores mustered by each organization’s average prospects in their debut and sophomore seasons:

The Mets again earn a place at the undesirable end of this scale, suggesting that the Minaya regime wasn’t particularly successful at any aspect of developing or handling prospects. Interestingly, if we limit our pool to players who amassed a combined minimum of three WARP in their debut and sophomore seasons—essentially, those who were unquestionably “ready” when called upon—we find that the most successful call-ups neither blew through the minors nor took their sweet time climbing the ladder. In fact, they averaged 2055 plate appearances and 387 innings, just slightly lower totals than the overall average. Oakland claimed the most immediately successful players, with seven surpassing the three-WARP threshold, followed by six for Cincinnati and five apiece for Minnesota, Baltimore, and Toronto.

Until now we’ve been pretending that each team makes decisions about its prospects against an identical backdrop. Of course, that’s an overly simplistic way of looking at things. Not only does each organization confront a different set of financial and competitive demands—some expect and depend on regular playoff appearances, while others have the luxury of throwing their prospects into the deep end to see if they’ll float—but each club boasts a unique collection of prospects, some of which are more advanced and deserving of promotion than others.

There was a weak (r=-.25) correlation between combined team winning percentage and number of call-ups from 2005-2009; the higher the winning percentage, the fewer call-ups we see, which makes sense, since relying on unproven players typically isn’t a surefire route to success. However, there are only extremely weak or nonexistent (r=.13 and .07, respectively) correlations between team winning percentage and PA/IP prior to debut; if there’s any tendency for weaker teams to rush their prospects to the majors, it’s a very slight one.

Adjusting for prospect talent is difficult, but as a quick-and-dirty attempt, I averaged Baseball America’s annual organization talent ratings from 2005-2010 to arrive at a single number representing the strength of each system over that span. The correlation between that combined ranking and the average WARP produced by each system’s call-ups over their first two seasons in the league was -.32, suggesting that as the ranks got lower (or in other words, as the systems got more talented), the graduates grew more productive, as one would expect. There was no correlation between system talent and average time spent in the minors prior to debut.

There’s plenty more meat on this topic’s bones, and other factors that we haven’t yet considered—for instance, the specter of arbitration has convinced many a club to postpone a top prospect’s arrival. It’s almost impossible to say whether any team is promoting its prospects in a suboptimal manner, since—aside from the fact that there’s so much we don’t know—we can’t retroactively alter a prospect’s timeline and watch his future play out differently. Still, it’s helpful to know that teams approach the dilemma differently; as long as most prospects continue not to resemble 2001-era Albert Pujols, development timetables will remain an area in which astute organizations can potentially set themselves apart.   

Thanks to Colin Wyers for research assistance.

You need to be logged in to comment. Login or Subscribe
The Mets numbers are skewed by their incredibly poor team health combined with atrocious major league depth, leading to youngin's like Tejada, etc. to be forced into duty.
This is a really interesting analysis...thanks!
It would be interesting to provide some control for draft position/international signing bonus for these guys. If a team debuts four 1st round picks during these years, you might expect those guys to spend less time in the minors than a team that debuts four 5th round picks.
Twins prospects are going to be realy disappointed if they are called up to the Metrodome, especially this year. :P Great article though. Would be interesting to see this pool of players split by HS, college, and international players and similar comparison tests run.
I had to re-read that Metrodome reference, too, initially thinking it was a mistake, but he was talking about 2005-2009, when the team was still playing there.
Right--I figured that might cause a couple double takes.
Interestingly, a team's average WARP out of rookies seems to have essentially no correlation with how the team has performed over the last five years (measured by W-L %, which isn't a perfect metric, but it'll do for something quick). There is a positive correlation between high WARP for rookies and winning, but the correlation constant is only about 0.12, which is very small and quite likely not significantly different from zero. Furthermore, there is scatter all over the place, with very good and very bad teams at both ends of the WARP continuum and in the middle. The message seems to be that good teams will be good and bad teams will be bad, regardless of how good their rookies/early-career guys are.
Very interesting stuff, although there are certainly a lot of covariates that obviously can't be accounted for. One thing that may be worth looking at, however, is how a team's preference for signing FA relievers effects the IP of their pitchers before callup. Particularly, teams that tend to sign a lot of FA relievers will mostly be calling up starting pitchers who will necessarily have logged more innings in the minors. Teams that call up a lot of relievers (or perhaps simply draft more) will necessarily be calling up players with fewer IP.
The Rays are simply having too much pitching talents, therefore their pitchers prospects have to wait in the minor for so long. It is an embarrassment of rich.
In a macro sense I think much of the learning curve may be societal/psychology based. Just like baseball players are the last element of American society to openly spit on a regular basis, the idea of a Greg Maddux slider inducing psychological havoc may be more a planted idea that is readily adopted. Look at the NFL; a rookie left tackle is expected to learn a ginormous amount of plays/information, refine his technique and footwork, build strength through the weight programs and then apply all that on game day while a vet dlineman literally tries to kick his az. Somehow, many of them pass the test well enough in year one and almost all are expected to be up to speed and contributing by year two. I would like to see new professional contracts in baseball only allow for two years in the minors then the player is either in MLB with his drafted team or is granted free agency. My guess is the players would adapt more quickly, or at least those capable of adapting, to the hand/eye coordination requirements needed to be successful on the field. There would be a constant buzz about new talent as it would rise more quickly; and such a system would build more excitement into the MLB draft each year too as fans project the near term impact of the draftee... I can see a lot baseball people strongly debating the above which, to me, ties back to the societal/psychology aspect of the current perceptions on player development being a planted idea that has been adopted...
The MiLB IP were lower than I expected. If you consider that a MLB starter could be expected to hurl 200 IP a year, having a median MiLB IP of 400 IP is just two years of development. Obviously, this is skewed because many of the pitchers are relivers and the minor league seasons are shorter than the major league seasons, but even so, I would have expected to see more MiLB IP. Perhaps a split between starters and relivers would be more illustrative. By comparison, the median MiLB PA for hitters is around 2000, or over 3.3 years worth of development if we assume that a starting MLB hitter has 600 PA per year. Nominally speaking, the hitters have over 50% more development time compared to pitchers. What makes this more interesting is that in spite of apparently having more development time, the average debut age of every position player except catcher is younger than the average debut age of pitchers. This could possbily be explained away if teams generally draft college pitchers and high school hitters. Thus even though the hitters had more development time, they were much younger players to start with. All this leads me to conclude that the missing picture is one that shows the average number of years players are in the minor leagues before making their MLB debuts.
Pitchers in the minors never throw as many innings as you suggest. Last year, in AAA's International League - the highest MiLB level - and the one that is friendlier to pitcher, to boot - was led by Brian Mazone of Lehigh Valley, who pitched 165 innings. In the High A Carolina League (again, the most pitcher friendly league at that level, the pitcher with the most IP was Nathan Jones, with 151.2. So if these are the highs, you can expect that most decent pitchers are actually a bit more babied, and thus average in the neighbourhood of 125-140 IP per season coming up. As such, it may take 3+ full seasons to rack up 400IP in the minors for a starting pitcher.
Sorry, but I'm very disappointed in this article. It seems to miss the most important issue with regards to the decision to promote or not promote any prospect. The answer to that is the source and age. Prospects come from somewhere. The Draft, international signings, independent leagues, etc. And not all Draftees are alike, High schoolers, juco, college. If a prospect enters the system at 23 (like a Tanner Scheppers), he probably won't spend as much time in the system as a kid who enters the system at 16 (like a Richard Alvarez or Roberto Perez)