Back on June 11, the Minnesota Twins were defeated by the Boston Red Sox 15-4, giving up 10 runs over the final two innings to let an otherwise close game get out of reach.

It wasn’t just another loss in another losing season. It was noteworthy because it gave fans their first look at J. T. Chargois, a highly-touted, hard-throwing pitching prospect from the 2012 draft. Chargois is one of many collegiate relief pitchers the Twins have been stockpiling with high picks in the amateur draft over the past several years, and became the first of these picks to make his major league debut.

Twins manager Paul Molitor said before the game that he would look for a low-pressure situation in which to ease Chargois into his role. It made sense. Chargois was coming off of two lost seasons to elbow surgery and had pitched on back to back days only once in 2016 to that point. The late innings of a lopsided game seemed like a good spot to give him his first appearance. Any runs Chargois might surrender would have little impact on the final result. And surrender them he did—he faced eight batters, six reached base, and he was charged with 5 runs in 2/3 of an inning.

And just as quickly as he had been called up to the majors, Chargois was immediately optioned back to AAA after the game.

Introduction

Chargois was drafted in 2012 out of Rice University with the 72nd overall pick, one of five college relievers selected among the Twins’ first eight picks that year. The Twins then drafted heavily again in this way in 2014. Obtaining more college relievers with high velocity and expectations of much quicker promotions was welcome news for many Twins fans, who were enduring a string of losing seasons in which their team would finish near the bottom in many pitching categories. The 2014 season was one in which division rival Kansas City was building a pennant winner around its overpowering, upper-90s-throwing bullpen arms. Twins used the 42nd overall pick in June 2014 on Nick Burdi, a relief pitcher from Louisville who could reach 100 mph on the radar gun.

But the 2014 draft was two years ago, and the 2012 draft was four years ago already. The Twins bullpen in 2016 is still relying on softer-throwing minor league signings made during the offseason. Where have all the college relievers gone?

From reading the discussion boards at Twins Daily this season, I noticed others asking this same question. TwinsDaily writer Seth Stohs had an article back in December on the long list of college relief pitchers the Twins have drafted over the past decade. Other than that, references were scarce; I couldn’t find much (or didn't conduct the right searches) about how college relief pitchers became major league relief pitchers. Are the Twins promoting their relief prospects too slowly? That was my main question, but I had some other questions too. I decided to frame the questions as an academic type of study. I wanted answers with Chargois and Burdi in mind, two guys who were selected with second round picks and have been indisputably developed for the purpose of becoming major league relief pitchers.

The natural response to the question of whether the Twins are promoting these pitchers too slowly is: They are promoted when they are promoted. Hard to argue with that explanation, but for many people, it’s not good enough.

I wanted to compare how other organizations promote pitchers with profiles similar to Chargois and Burdi through their minor league systems. My intent was not to perform any sophisticated statistical analyses or conclusively answer any big philosophical questions, nor do I consider roster issues, or a pitcher’s velocity or pitch repertoire, or any qualitative information that might affect a player’s progress or a team’s evaluation of it. I just wanted to dig up some data, present it, and see if it was saying anything.

I settled on three hypotheses or questions I wanted to test. First, that the Twins have drafted more of these types of pitchers than most other organizations; second, that the Twins promote these pitchers more slowly than other organizations; and third, that it makes no difference to their careers when these pitchers make their first major league appearance, whether it be a critical situation or low leverage.

Also, I believed that there are more efficient ways to obtain strong bullpen arms than targeting them in the amateur draft, but I chose not to address this last question. A good rundown of the best relievers the Twins have developed over the years is here, and they are not primarily relief pitchers drafted from college.

On the initial three hypotheses, I found the answers mixed. But I did discover a couple of things that surprised me.

Methods

In order to arrive objectively at a set of pitchers to examine, I established some rules. First, I would define what a college relief pitcher is. Then I would set some conditions for which pitchers from the draft were eligible for inclusion into the sample, and finally, I would need to decide on how to measure the promotions of these pitchers once they were in a team's minor league system.

Above all, I was interested in success stories. Again, I was targeting players who were drafted as college relievers, who pitched in relief in the minors, and who eventually reached the major leagues as relievers for the team that drafted them. I wanted to know what characteristics these ballplayers had in common, and if Chargois, Burdi and other Twins relief pitchers shared those characteristics.

Defining a college relief pitcher. I didn’t want starting pitchers in the group. During his college career, J.T. Chargois appeared in 47 games as a pitcher and started only 2 of them. (Note: I would use the Baseball Cube for retrieving college stats, and then verify these stats against other sources when possible.) I was comfortable calling Chargois a college relief pitcher. Other times it was less clear. Logan Darnell appeared in 43 games—strictly in relief—in his first two seasons at Kentucky. In his third and final season, he made 11 starts. Was Darnell a college reliever? Did the Twins consider him one? I wasn’t sure.

After browsing through the records of many former college pitchers who were selected in the draft, I settled on the following rules. A college relief pitcher would be someone who:

• pitched in relief in a majority of games his final college season; or, pitched in relief in a majority of all college pitching appearances; and
  
• pitched fewer than 200 innings total as a college pitcher.
  

There is nothing special about these rules; I thought they simply made sense for what I wanted to look at. Unfortunately those rules created a couple of odd exclusions, including Chance Ruffin, who was the closer for the Texas Longhorns during the 2010 NCAA season but who made 28 starts in his two prior seasons at Texas, exceeding the innings limit. Also disqualified as a relief pitcher here was Madison Boer, who the Twins drafted at the end of the 2nd round in 2011 but who also barely exceeded the innings limit (according to Baseball Cube). It's also not impossible that a pitcher or two in this study might have been unintentionally mis-categorized, but if so, it was without bias.

Eligibility for sample. Having defined a college reliever, I then decided I would only look at picks in the first and second rounds, including compensation picks. This would include Chargois and Burdi in my set (but not Darnell, who in 2010 was a sixth round pick).

I settled on the drafts from the years 2004-2012, however. This meant I would have Chargois, but not Burdi, who was drafted in 2014. I decided that for any players drafted 2013 or later, their teams wouldn’t have the benefit of development time. Though this endpoint excludes Burdi, I will bring him into the discussion later. I had also supposed that Chargois, a mid-second round pick in 2012, might be the final pitcher in the set chronologically, but my first surprise was that there was a college relief pitcher drafted later than Chargois in the second round who has been in the major leagues for several seasons now.

The 2004 draft was my other endpoint, or beginning point, because in my initial browsing, I discovered that Huston Street was part of that draft, and he is an example of the successful type of college pitcher I imagine the Twins are hoping to develop with these picks. I also liked the 2004 draft for the irrational reasons that it was the first draft held following the publication of Michael Lewis's Moneyball, it was the draft of Glen Perkins, and also the draft of Matt Bush, the first overall pick that year, selected as a shortstop but now pitching effectively in relief with the Texas Rangers. (Perkins and Bush are not otherwise part of this study.)

As I was finishing this project, I became aware that in 2002 the Twins drafted Jesse Crain in the 2nd round with the 61st overall pick. Crain then debuted for the Twins two seasons later. Crain would have qualified for this list had I broadened the year range back to 2002. But by that time, I did not want to expand the set of pitchers based on this selected piece of information. However, like Burdi, I will bring Crain into the discussion at various points of interest.

How to measure the speed of promotions? I thought it would be reasonable and convenient to use innings pitched to measure a pitcher’s time spent at each minor league level. Innings pitched has the advantage of familiarity over other measures like batters faced, game appearances, or even calendar days spent at a level. I would track the number of innings pitched up until the pitcher's major league debut.

I also began with the intention of tracking minor league performance, such as strikeout and walk rates, but as I went on, I felt less of a need for this. Sample sizes are too small for this exercise, and my own judgment about performance would be too uninformed and subjective. Here is an example. An argument could be made that Nick Burdi, compared with his stats in college, struggled with control (a higher BB/9) at his first minor league assignment in A ball at Cedar Rapids. Indeed, I believe the Twins even stated this. However, Huston Street, with the Oakland organization, also struggled with control in this same way (with a higher BB/9) at his first minor league assignment. I began the study wanting to measure performance, but when I considered things like park bias, and sample size, I decided it was not significant. Huston Street pitched only 2 innings in AAA, after all. Whatever his walk or strikeout rate in AAA, it wouldn't be as significant to me as the number of innings he pitched. I reasoned that Oakland would not have promoted Street after a mere 2 innings in AAA if they did not feel he was ready.

In the end, I was comfortable bypassing statistics altogether and using the promotion itself as the main indicator for satisfactory performance.

I did not account for innings pitched in exhibitions, minor league postseasons, or independent leagues. Only innings pitched in the affiliated minor leagues were part of this study; in other words, the innings pitched total you would see if you checked MiLB.com.

Likewise, other than Chargois, I had very little information on injuries for these pitchers, and I simply used their presence in games as an indicator of health.

The College Reliever Search

Using the criteria above, I found a total of 50 college relief pitchers drafted in the first or second round from the years 2004 to 2012.

Have the Twins drafted more college relief pitchers than other organizations? This was my first hypothesis, and from this pool of relievers, I would conclude yes. From this set of 50, the Twins drafted five college relief pitchers within the first two rounds between 2004 and 2012. The Dodgers also drafted five. The Diamondbacks and the Cardinals each drafted four, and no other team accounts for more than three such pitchers.

However, I determined that 27 of these 50 relief pitchers were converted to starters early in their minor league careers, or were given enough starts to cast doubt on whether they were drafted with the strict intent of becoming major league relief pitchers. For example, Mason Melotakis, drafted in the 2nd round of 2012, met the conditions of a college relief pitcher, but he made 18 starts in A ball to begin the 2013 season. Carlos Gutierrez, drafted in 2008, had a similar profile as a potential starter. Starting games in the minor leagues made Melotakis, Gutierrez and the others distinct enough from Chargois and Burdi that I excluded them from my set.

Removing those 27 converted starters from the original set of 50, and removing four others who did not sign a contract (including Josh Fields, who is actually counted twice in the set of 50 as he was drafted in consecutive years), I arrived at a set of 19 college relievers who were drafted in the first two rounds and then used almost exclusively as relief pitchers in the minors.

The Twins drafted 3 of these 19 players to become relievers exclusively, while no other team drafted more than two. Setting aside Melotakis and Gutierrez, here are the three Twins pitchers who would make up part of this group of 19 (Table 1):

Table 1: College relief pitchers drafted by the Twins between 2004-2012 in the first or second round, who were developed exclusively as relief pitchers in the minor leagues

Because I was only interested in college relievers who have reached the majors with their original teams, I sorted these 19 relief pitchers into even smaller sets: three who were traded to another organization while still in the minors, including Bullock (and also Josh Fields again, who I will mention later); three who retired or were released by their original clubs before making it to the majors; and two who are still in the minors with their original clubs, including Luke Bard. That left 11 left over. The remaining 11 pitchers were the ones I was looking for—college relief pitchers who made it to the major leagues as relievers with the club that originally drafted them.

These were the guys who again, like Chargois and Burdi, were selected with high draft picks, developed through the minors almost exclusively as relief pitchers, and made their debuts with the teams that drafted them.

Table 2 introduces these 11 pitchers, listed chronologically by year drafted and draft position: Bill Bray (drafted as a Montreal Expo), Huston Street, Craig Hansen, Joey Devine, Chris Perez, Eddie Kunz, Ryan Perry, Daniel Schlereth, Drew Storen, J. T. Chargois, and Paco Rodriguez.

Table 2: College relief pitchers drafted between 2004-2012 in the first or second round, who were developed exclusively as relief pitchers in the minor leagues, and who debuted in the major leagues as relief pitchers for the teams that drafted them

By collecting data from this set of 11 relief pitchers, I could now look at innings pitched to test my second and main hypothesis, about whether the Twins are promoting college relief prospects more slowly than other organizations. I got my answer, but found that the question was a little more complicated than that.

Before collecting this data, however, I wanted to get a sense of how well these pitchers have performed in MLB. A couple of these pitchers are well known, while a few of them I had not heard of or had forgotten about. Table 3 lists these same pitchers, tracking the pitcher by age and the fWAR accumulated in the majors at that age, through the end of the 2015 season:

Table 3: Age of relief pitchers per season, and accumulated fWAR for that season

There are some problems with using fWAR and dividing by seasons played to put a value on a player, but I was only looking for approximations. Additionally, when evaluated this way, I found these pitchers could be put into three natural groupings: a very good group, an above average group, and a below average group.

The following data re-sorted and displayed in a bar graph (Figure 1) will illustrate these three groupings better. The top and bottom of each grey bar represent the highest and lowest season fWAR for each pitcher. The black line near the middle of each bar represents the pitcher’s career average fWAR. The league average fWAR for a relief pitcher is 0.16, indicated by a dashed line.

Figure 1: Relief pitcher high and low season fWAR, represented as a bar graph, relievers sorted by general effectiveness

The first three bars in order are Huston Street, Drew Storen, and Chris Perez. These three make up what I considered the very good group. The next three bars are the above average group of Paco Rodriguez, Joey Devine, and Bill Bray. The last group of bars includes Ryan Perry, Daniel Schlereth, Eddie Kunz, Craig Hansen, and J. T. Chargois.

The pitchers will be sorted in this order in all of the bar graphs to follow.

I put Chris Perez with the first group based on his career saves total, and because Perez’s fWAR was hurt by his final two seasons. bWAR treats Perez much better than fWAR. Perhaps Perez would fit better in the middle group, but I chose to include him in the top group. Perez also was named to two All Star games.

You can also see Jesse Crain’s performance in this graph. The first bar represents his seasons with the Twins, which would fit neatly with the above average group of relievers. Crain’s second, taller bar represents his full career, including his final three seasons with the White Sox.

Results and Discussion

My second hypothesis was that college relief pitchers in the Twins system were promoted more slowly, as measured by innings pitched, than the pitchers in the other organizations. Note that this innings pitched data is not intended to differentiate across seasons, only minor league levels.

For each of the 11 relief pitchers in the final set, Table 4 lists the number of innings each pitched at each minor league level prior to their major league debuts:

Table 4: Relief pitchers, innings pitched per minor league level

Below is the same information presented in bar graph format (Figure 2). Already, one or two things start to stand out.

Figure 2: Relief pitchers, innings pitched per minor league level, represented as bar graphs

The first thing that stood out to me is the tallness of the bar representing Chargois. Chargois pitched more minor league innings prior to his debut than all but Perez.

Another clear bit of data that stood out is the number of innings Chargois pitched in rookie league, which I generally thought to be for players coming out of high school. All six of the very good and above average relief pitchers skipped the rookie level. At first, I thought that the Twins not having a short season A- team might have narrowed a choice of assignment to either rookie level or A level, but each of the top three relievers pitched in organizations with a short season A- level team, yet were assigned directly to A level.

However, looking at Chargois and the top three relievers, what also stands out is a pattern of promotions those pitchers were given through the minors. Each of the top three of Street, Storen and Perez began their careers at A ball, pitched at AA, and then at AAA. Storen also made a stop at the high A+ level. Each of the top three spent their longest minor league assignment, measured by innings pitched, at AA. On its face, it looks significant to me that this progression of promotions is shared among the top three relievers, and I find it encouraging that Chargois also followed this progression. What’s more, none of the other seven relievers in the set shows this progression in just this way. On the contrary, the promotion patterns of the others seem haphazard in comparison.

The most common characteristic shared among all the pitchers is pitching at the AA level. Ten of the eleven pitchers in this set pitched in AA prior to their major league debuts. The top three were then promoted to AAA and promoted to the majors from there. However, six of the other seven were promoted to the majors from AA.

Nine of the 11 pitchers prior to their debuts, including Chargois, pitched their highest number of innings in AA. Perhaps these organizations see AA as the most important level for these pitching prospects.

But for how long did these other pitchers stay in MLB after their debuts? Here is a chart with an additional bar to represent major league innings pitched following each pitcher’s debut:

Figure 3: Relief pitchers, length of first stay in major leagues, represented as a bar graph

Again, Chargois stands apart from this group in a few respects. His initial stay in the majors was very brief and fits more with the bottom of the group or relievers than the top group.

The table below shows the number of major league game appearances and innings pitched for each pitcher following their initial call up (Table 5):

Table 5: Relief pitchers, number of game appearances and innings pitched following MLB debut

The best three pitchers had three of the four longest stints in the major leagues. Street was with the team from Opening Day, and Storen and Perez were each called up in the middle of May. Bill Bray was called up in June and then traded in July. Paco Rodriguez was a September call up, then broke camp with the Dodgers the following spring. Only Joey Devine was demoted shortly after his first call up.

Leverage of First Appearances

Does the leverage index of a debut appearance make a difference?

I decided to look at the leverage index of the first three major league appearances of each reliever (Table 6). Approximately, low leverage is a number under 0.85, medium leverage is between 0.85 and 2.00, and high leverage is anything higher than 2.00:

Table 6: Relief pitchers, leverage index (LI) of first three major league appearances following first call up

Most of the relievers did indeed debut in low leverage situations, if not quite blowout losses as Chargois did. In fact, Huston Street’s first three appearances were also fairly late in blowout games (one win, two losses).

The big difference between Street and Chargois is that Street eventually took over the closer role later in his rookie season, something Chargois seems unlikely to do.

The highest leverage debut appearance came from Joey Devine, who entered a game in the top of the 12th inning for the Braves. Devine’s appearance was rated a leverage of 2.31. Devine was one of Bobby Cox’s last options from a bullpen that appears to have been depleted from the previous couple of games. Devine pitched a scoreless top of the 12th, struck out trying to lay down a bunt with two strikes in the bottom of the 12th, then got into trouble in the top of the 13th before surrendering a grand slam and taking the loss.

Meanwhile, Bill Bray was credited with a win in his debut appearance—without facing a batter. Pitching for the Nationals and manager Frank Robinson, Bray entered a game in Milwaukee in the bottom of the 8th inning with a runner on first and two outs, his team trailing by a run. He was brought in to face the rookie Prince Fielder, but the baserunner was caught stealing after the first pitch, ending the inning. The Nationals scored two runs in the top of the 9th and another pitcher replaced Bray to record a save. Bray got the win. The baserunner who was thrown out? Corey Koskie, in his final major league season.

Jesse Crain's MLB debut in August 2004 was in a 0.51 leverage situation, trailing by five runs in the 4th inning versus the Angels. Gardy pulled no punches with Crain’s next two appearances, however, with leverage indexes of 2.85 and 3.41 each, one of those being an extra-inning appearance.

As far as leverage is concerned, I am not sure there is enough information—here or anywhere, perhaps—to determine whether a debut appearance in a high leverage situation will impact a career negatively. Introducing a relief pitcher into a low leverage situation wouldn’t seem to hurt, however.

Other Pitchers

In the following bar graph are the innings pitched of some additional pitchers, including Twins, who were not part of the original set of 11. Each of these pitchers were also relief pitchers in college and developed as relief pitchers in the minors:

Figure 4: Relief pitchers, innings pitched in minor leagues, represented as a bar graph

The first three remain Street, Storen, and Perez. The next two pitchers, separated by extra space in the middle, are Chargois and Nick Burdi. Chargois’s stats are through his debut in June, while Burdi’s stats are through the end of 2015, before his 2016 season began.

The next group of six pitchers is made up of Twins. In order, from left: Billy Bullock, Luke Bard, Zack Jones, Jake Reed, Trevor Hildenberger, and Jesse Crain. The stats for Crain are up through his major league debut in 2004. The stats for the other five Twins pitchers run through the end of the 2015 season and do not include 2016.

The final bar at the far right is Josh Fields, who was just traded from Houston to the Los Angeles Dodgers, which is now his fourth professional organization. Briefly on Josh Fields: Fields was among the 19 relievers in the sample above who were college relievers and pitched almost exclusively in relief in the minors. Seattle drafted Fields with the 20th overall pick in 2008, and notice how they assigned him directly to AA, skipping the lower levels completely. After seeing the progression of promotions from the top three relievers, it’s not hard to imagine how Fields might have struggled in that first long minor league assignment, facing a quality of competition he was not yet suited for, and then having trouble catching up and meeting expectations from that point forward.

The second tallest bar, next to Fields, is Jesse Crain. You can see how the Twins promoted Crain comparatively slowly, but also according to the progression of minor league levels, without skipping a level. Crain spent the longest time in AAA, rather than AA.

Notes on other Twins college relief pitchers here include:

• Billy Bullock, drafted in 2009, in the 2nd round with the 70th pick. Bullock was traded to Atlanta during spring training of 2011 after 107.2 minor league innings with the Twins. Bullock pitched in AAA for a couple of seasons and then was suspended in December 2012, effectively ending his career;
  
• Luke Bard (2012), 1st round with the 42nd pick. Bard has been hit with significant injuries which perhaps prolonged his stay in A ball, but is currently healthy and pitching in high A;
  
• Zack Jones (2012), 4th round, 130th overall. Looking at his stat line, Jones seemed to pitch well and even improve on his college numbers in 3 ½ seasons in the Twins system. Jones was then taken in the Rule 5 draft by Milwaukee in December 2015, but returned to the Twins in June. He is pitching in AA;
  
• Jake Reed (2014), 5th round, 140th overall. Reed would not have made my original set, for pitching too many innings in college. Reed sputtered when promoted to AA in 2015, but is performing very close to his college numbers in AA now in 2016;
  
• Trevor Hildenberger, 2014 draft, 22nd round, 650th overall. Successful late round picks are fun stories, and Hildenberger is finally getting some attention this season. Like the others in this study, he was a college reliever who was developed as a reliever in the minors. Like the other Twins, he has progressed through every level, and was promoted comparatively slowly to Street, Storen and Perez, which I attribute partly to his low draft selection and lower expectations. Hildenberger has dominated at AA this season;
  
• Mason Melotakis. I did not include Melotakis in the bar graph above, because the games he pitched as a starter in 2013 would distort his innings compared with the others. He was drafted in 2012, 2nd round, 63rd overall. He has been promoted on the same schedule as the others, with stops at every minor league level, including rookie. Melotakis is also currently pitching in AA;
  
• Pat Light. Light was drafted 37th overall in the 2012 draft. He was a starter in college and in the low minors, but has pitched exclusively as a reliever since being assigned to AA to begin 2015. At AA he has pitched 29.2 innings in 21 appearances, and at AAA he has pitched 64 innings in 51 appearances. He debuted with Boston on April 26 of this season and had two appearances before being demoted back to AAA;
  
• Alan Busenitz. Busenitz was college relief pitcher drafted in the 25th round of the 2013 draft. He has pitched 254 innings in the minor leagues, mostly in relief.
  

All of which leads us to Nick Burdi.

On the surface, it looks as though Burdi and Chargois are being promoted on the same schedule as the other Twins college relief prospects. Here is the comparison of Burdi’s minor league innings through 2015 with the innings pitched by Street, Storen, Perez, Chargois, and Crain prior to their major league debuts:

Figure 5: Relief pitchers, comparison

Again, Street and Storen were promoted fairly quickly, while Perez and the three Twins were promoted less quickly, or even slowly, in comparison. Chargois and Crain pitched 75.2 and 85.2 innings apiece through the AA level prior to their debuts. Similarly, Burdi reached 84 innings through AA through the end of 2015 (not counting Burdi’s innings during Chattanooga’s 2015 postseason run).

Notice how Burdi skipped rookie ball and was assigned directly to A ball in 2014. As we’ve seen, for the Twins this is unusual. Remember too that Burdi struggled in the first half of 2015. After 30.1 innings in AA he was demoted back to A+ for a short time. Burdi then pitched 20 more innings in A+ before his promotion back to AA.

Now suppose that Burdi, instead of struggling in those first AA appearances, had pitched very well. Suppose he pitched well enough that instead of being demoted at the end of June, the Twins instead promoted him to AAA, and he pitched in AAA until the end of the Rochester season. Here’s how the graph might look under these new innings totals (FIG 6):

Figure 6: Relief pitchers, comparison alternative

This alternative promotion schedule is similar to Storen’s, especially through the end of AA. If the 33.1 innings pitched by Burdi after his demotion had all been credited to AAA instead of the lower levels, the AA to AAA innings ratio would have resembled Crain’s prior to Crain’s debut. It is not a stretch of the imagination to think that Burdi would have been considered for a September call-up last season had his season gone differently.

Wrapping up

Looking only at this small set of 11 pitchers, one can conclude that what is most important to the major league success of these pitchers is not the speed of their promotions, but instead is the logical progression of promotions the pitcher takes through the minor leagues. The pitching prospect is drafted and assigned to a minor league team, after which point the pitcher advances level by level, without skipping more than one level at a time, until AA is reached. A college relief pitcher on a successful development plan can skip the rookie level, but will not skip the AA level. At AA, the pitcher may be asked to pitch more innings than at any previous level. AA is then followed by a stop in AAA, which is followed by the pitcher’s MLB debut. The soundness of this progression is even more evident when looking at the major league careers of the college relief pitchers in this study who did not follow the progression (Figure 2). It’s true that better data and a larger sample of pitchers might alter how this conclusion currently looks, but for now I am standing behind it.

Unfortunately for this study, the relative slowness with which the Twins promote their relief prospects means that their success will not be apparent for a while. The quick promotion of Paco Rodriguez, selected 10 picks later than Chargois in 2012, does suggest that a late second round pick would not need to be promoted slowly out of fear of hurting the pitcher’s career. The Dodgers had already gotten more than three seasons and 1.2 fWAR from Rodriguez before trading him away during his age 24 season. With the slower timelines the Twins prefer, the Twins figure to miss out on the early value a college relief pitcher might present.

Only after I finished with the above data did I hunt for pitchers with similar profiles who fell outside of my set. This is when I discovered Jesse Crain, drafted in 2002. There are pitchers such as Jake Barrett in Arizona (2012, 3rd round) and Tony Zych in Seattle (2011, 4th round) who were taken a bit later in the draft and spent a great deal more time in the minors than Chargois, but there are also pitchers like Joe Smith (2006, 3rd round), who was promoted rapidly by the Mets and debuted in MLB with only 32.2 minor league innings behind him.

A similar, better study of minor league promotions would widen the sample of pitchers, and would also improve on the definition of college reliever as used here.

For now, it's reasonable to think that the Twins still expect a few of these relievers in their system to become valuable major league contributors. I would also conclude that most of these relievers have been promoted responsibly, if a little slowly. But again, too slowly? It's hard to know for sure. Chargois could probably have been called up by now. Once Burdi becomes healthy, the Twins will probably ask him to pitch more minor league innings than he really needs. And aside from Chargois and Burdi, there is still an abundance of young pitching prospects in the Twins system waiting for their chance. Perhaps half of the AA Chattanooga bullpen is made up of former college relief pitchers who could stand to use a promotion, but not to the majors yet—to AAA.