The starter/reliever prospect post

DMZ · December 3, 2004 at 3:05 am · Filed Under Mariners 

My distaste for having good minor league pitchers relieve in the minors has come up before, so by request here’s my view in a little more detail.

If a minor league pitcher has good stuff, even if it’s only two good pitches, they should start until their progression stalls. You want to get these kids as much experience as you can, and the regular, pitch-count protection of starting allows a team to get them a lot of playing time (starting pitcher: 200+ innings, reliever: 50-80) throwing to live batters in game situations. Between starts, you get quality side sessions with the coaching staff. If they’ve got two pitches, that’s a lot of time to try and figure out how to spot a new pitch.

Even then, a two-pitch starter can be effective if those two pitches are nasty enough, and the way to refine craft is to get them experience and instruction, and that’s starting.

There’s a case that pitchers who are unsucessful starting can succeed in relief, because they can put more on each pitch, not worrying about endurance, plus repetoire limitations become less important.

I agree — but I don’t want to concede that before you absolutely must.

Another argument is made relieving is different, that coming into a game in progress is so crazy that pitchers have to be prepared for it — that you can’t have a starter convert after a couple years of minor league experience. If you think someone’s a reliever, you should train them to be a reliever. I’ve never seen any evidence to support this. Relievers are, by and large, converted relievers. Few elite relievers were minor league relievers.

I think the opposite is true, though. If a team drafts a college closer, has them close all through the minors, then take over the closer’s job for the big team, you’ll never know if that guy could ever have been a quality starter. While there may be many starter candidates for conversion to the bullpen, someone who builds their whole career off a couple pitches thrown as hard as they can until they fall over is rarely going to be an easy conversion project.

Let your most talented pitchers start until the force a move to relief on their own. It’s that simple.


21 Responses to “The starter/reliever prospect post”

  1. Bela Txadux on December 3rd, 2004 5:45 am

    I agree with your major point here, Derek, but there’s a hitch in it, too.

    In principle, having a guy start is optimal, if for no other reasons beyond the ones you mention than to build up the durability of a guy’s arm and to have the most opportunity to groove a guy’s mechanics. Historically, it’s a certainty that elite relievers were virtually all minor league starters. The counterexample that actually bolsters your point here is Billy Koch. He never started—and has never learned how to _pitch_, either, or even to consistently locate his former triple digit stuff. The ‘surprisingly mixed’ results shown to this point by Ryan Wagner, and to a lesser extent Chad Cordero, two other power arm, never-started guys, certainly doesn’t contradict the position either. By this evidence, reliever candidates should definitely start in the minors as you suggest, no matter how they are projected by their organization. Presumably this means they should start as high in the minors as they yield passable numbers.

    At what point does the conversion happen, though? Again, historically most elite relievers were converted at the _major league_ level, usually after failing as a starter in the Bigs. Arthur Rhodes is just one guy who comes to mind, but Keith Foulke, Tom Gordon, Eric Gagne, and even Mariano the Sauce fit this pattern. Where most guys hit the wall, though, is in AA, where two pitchs aren’t enough, usually because they are really one-and-a-half pitches. With a year of AA under a guy’s belt, it’s usually a fair bet whether or not his skillset gives a reasonable projection for the majors, and it’s time to talk about what will _eventually_ be a conversion to relief for a prospect who nonetheless has at least one + pitch. Nowadays, these guys are being converted immediately, in AAA, or even in a subsequent AA season whereas again historically they would have been pushed as mediorce or sub-mediocre starters up into AAA and even a major league audition to my recollection.

    The secondary question here, then, is how high do you push a guy to start who clearly is not going to do so at the major league level? Wasting even one year of a guy’s development time on a role he can’t and won’t fill while also kicking around his confidence doesn’t seem like a good idea at all, but the decision to convert a guy is quite often not cut and dried. JJ Putz is just one example. To my recollection, he never put up any kind of decent numbers as a starter above A ball, but the Ms futzed around for several years above that level before committing to him in relief. He was still behind the curve (so to speak) in development AS A RELIEVER at the point early in ’04 when the Ms brought him up from need, and he was, frankly shaky through the first half of the season. The jury is still out on whether he and Price together made him an effective reliever in the back half of the season, but by then how old is he? How many years has it been since he stalled in AA, what four years?? And then there was the whole idiot’s ploy of sending Matt Thornton to AAA at the beginning of ’04 TO _START_, when he has never at any time in the minors been effective as a starter, when his only hope of ever helping the organization at the big league level was as a relief thrower. A team can try for _too long_ to push a guy as a starter, and more often than not this is what they do, it seems to me.

    I would suggest an ‘A level’ test, then, to purloin a coined phrase. After a pitcher’s full season in high A, I think an organizational development staff needs to make a solid, either/or/neither call on a pitcher’s future. Guys with two or more major league pitches and solid numbers get pushed as starters, period. Clint Nageotte and Felix Hernandez fall into this class, though at opposite ends obviously; Eric Gagne and Rafael Soriano, too. Guys with one or two major league pitches, but mediocre results or major flaws get converted, or get special help and an overt challenge on promotion to AA to step up as a starter or they will be converted almost immediately. Matt Thornton should have been in this group year’s ago, but I suspect Dennis Tankersley also, and several of San Fran’s so-called high-end pitching prospects that were. The third group has no + pitches, but some pitching skills, and doesn’t really project at all, so they might as well be kept as starters and ‘prove it’ promoted until they meet their match. This includes the great bulk of minor league pitching prospects, obviously, but guys like Ryan Franklin and Craig Anderson come immediately to mind. If their skills become sufficiently refined, like Franklin and probably Kirk Saarloos, they can give you something out of the major league pen subsequently and even backdoor their way into the back end of a major league rotation again exactly like Ryan did.

    Where teams usually make the wrong call is with high draft picks or power pitchers who fall into the ‘or’ category much like Thornton. The temptation to keep pushing a guy to make it as the starter you have been dreaming of in them when the results just aren’t there tends to make for poor decisions as I have seen it. On the other hand, there are guys like Matt Clement, Clint Nageotte—or for that matter Nolan Ryan, and Kevin Brown—who struggle for years as a starter with mediocre results, only to finally refine their mechanics and approach to emerge as dominant starters. If they’d been converted promptly, there’s two HoF and one solid starter careers folded up to Dave Righetti-like matchbook proportions. For the ‘either’ and ‘neither’ categories, the call on convert/don’t bother is usually pretty straightforward.

    No overall approach is going to work for every guy’s situation. But I think a mid-minors firm assessment as a starter on a pass-fail basis will optimize an organization’s yield, is what I’m saying, not a muddle through while singin’ a song approach.

  2. Bernard Aboba on December 3rd, 2004 8:21 am

    Hasn’t some analysis shown that pitchers with great longevity frequently pitched only limited innings early in their career? For example, here are the innings pitched for Nolan Ryan:

    Age Team IP ERA
    21 1968 Mets 145 3.09
    22 1969 Mets 89.1 3.53
    23 1970 Mets 131.2 3.42
    24 1971 Mets 152.0 3.97
    25 1972 Angels 284.0 2.24

    So Ryan didn’t pitch a large number of innings until he was 25, but pitched until he was 46, retiring after pitching 66 innings for the 1993 Rangers with a 4.88 ERA.

    Here is the data for Warren Spahn:

    Age Team IP ERA
    25 1946 Braves 125.2 2.94
    26 1947 Braves 289.2 2.33

    Spahn didn’t pitch a large number of innings until he was 26, but pitched until he was 44, retiring after his 1965 season with the Giants (pitching 71.2 innings with a 3.39 ERA!)

    Similarly, Hoyt Wilhelm did not pitch more than 200 innings until his stint with the 1959 Orioles, when he was 27. He pitched until he was 50; his last season he pitched 25.1 innings with the 1972 Dodgers, with a 4.62 ERA.

    On the other hand, Tommy John first pitched 200 innings when he was 23, with the 1966 White Sox.

  3. J on December 3rd, 2004 8:52 am

    Well… at the same time they’d have to have some pretty solid coaching in the minors to make sure that they can make the decision of starter vs. reliever with the most confidence they can, and, while I don’t know the specifics very well, it would seem that a large part of the coaching talent is at double-A and triple-A, and rightfully so.

    We use Nageotte as an example frequently when it comes to guys like this. My understanding of the way he pitches is that he only really started utilizing his fastball once he got to San Antonio, and prior to that it was even more sliders. It’s an interesting idea you’re presenting here, Bela, but like you said, it’s not like it works for every guy and I can’t come up with a better alternative just yet.

    I can, however, say decisively that I get pretty sick of it when 99% of the new pitching call-ups get “future closer” attached to their names.

  4. Brent Overman on December 3rd, 2004 10:18 am

    A agree mostly to what you say regarding pitchers, but do have other issues that must be considered. I would apply your rules in a general sense.

    That said, the pitcher’s mechanics must be looked at as well. Was Troy Percival a starter in the minors? I don’t recall this offhand, but when you look at his mechanics, there’s no feasible way he could be a 7-inning pitcher due to the torque he generates, which also makes him effective. You can say this about a number of guys, like Bryan Harvey, Ben Weber, John Rocker, et al. While many of these guys started at one point or another in the minors, it wasn’t until they were a reliever that they succeeded. Some, like Percival, wouldn’t start in the minors, or else they’d never make it to the majors.

    In addition, along the lines of mechanics again, one has to look at guys who have piss-poor mechanics, like Kyle Peterson (former Brewers prospect). Peterson had a ridiculous cross-fire, a la Randy Tomlin (former Pirates pitcher). Peterson blew his arm out, but I’d venture to guess that he could have made it further as a reliever than as a starter, due to the stress he put on his elbow from his cross-fire (which also tanked Tomlin’s elbow as well).

    I equate this similarly to the splitfingered fastball/forkball. Most starters who rely on this pitch WILL have Tommy John surgery. Not if, just a matter of time. However, if you’re a reliever, you have the tendency to last a lot longer, since your arm isn’t continuing to throw that stressful pitch beyond the point of muscle/tendon fatigue, like one would in a starting role.

    There’s no easy way to disect it…

  5. Brian Harper on December 3rd, 2004 10:38 am

    Sounds like a fine plan, to have every pitching prospect start until they demonstrate that there’s no way they’ll make it as a starter. But there’s a simple problem of logistics to overcome. You’ve only got a limited number of slots for starters on your single-A and rookie league teams. Conceivably just about every pitcher on those teams would be too early in their development to give up on being a starter, so how do you get starts for all of them? And who relieves on those teams? It seems like there’s no choice but to start projecting players into starting or relieving early in their development. What kind of system would allow for more than 10 starters between our two single-A clubs?

  6. Dave on December 3rd, 2004 10:46 am

    The A’s and Rangers, under the leadership of Grady Fuson, both established tandem-starter programs in the lower levels of the minors. The program paired pitchers together who would work on strict pitch counts, usually limiting them to about four innings apiece, and insuring that eight pitchers per team got regular work and could prepare in advance. The system also encouraged pitch efficiency, as the only way to qualify for a win by going five innings was to get your 15 outs relatively quickly.

    Fuson, by the way, is still unemployed after John Hart forced him out in the power struggle in Texas. While it has no chance of happening, I still endorse the Mariners hiring of Fuson to any position in the organization. Having him around simply cannot be a bad thing.

    Percival was drafted as a catcher, by the way. He was converted to relief work and started only three games in the minors. Brent is correct that some pitchers simply do not have the necessary mechanics to maintain the endurance necessary to start, but those pitchers are rare enough to be the exception.

  7. Matt Williams on December 3rd, 2004 11:09 am

    Ok, I know this is off-topic…but the Bonds steroid thread is locked.

    There’s speculation that the Yankees might get out of Giambi’s nightmare contract due to his admitting steroid use. Seriously, if that happens it would be enough to make even the most level-headed person a conspiracy nut.

  8. Lefebvre Believer on December 3rd, 2004 11:16 am

    Great Post and comments. Where does Soriano fit right now? I was in Cali at the beginning of the season and really didn’t get to see him pitch before injury. Is there any indication the organization will look at him as a starter? Or is he the designated “closer of the future”

  9. Matt Williams on December 3rd, 2004 11:26 am

    Lefebvre the organization pegged him as a reliever/closer to the press, and then a week later he said he really wanted to be like Mariano Rivera. So there’s good indication he won’t be given a shot as a starter.

  10. Dave on December 3rd, 2004 11:33 am

    If Soriano had a chance at moving back into the rotation, it went out the window when he blew out his arm.

  11. paul mocker on December 3rd, 2004 11:39 am

    Let your most talented pitchers start until they force a move to relief on their own. It’s that simple.

    Well put.

    And let your best pitchers pitch to more batters than your worst pitchers.

    As for a GM’s talent acquisition strategy, the goal should be to acquire any pitcher who is better than your worst pitcher, regardless of spot in the rotation.

  12. paul mocker on December 3rd, 2004 11:40 am

    Speaking of Soriano, when would you feel comfortable in his arm health to make him a starter? 2006? 2007? Later?

  13. Pat on December 3rd, 2004 12:10 pm

    “And then there was the whole idiot’s ploy of sending Matt Thornton to AAA at the beginning of ‘04 TO _START_, when he has never at any time in the minors been effective as a starter, when his only hope of ever helping the organization at the big league level was as a relief thrower.”


    Your opinions seem to be very well considered, which is why it surprises me that you have missed the boat on this point of fact supporting your argument.

    In 2001, Matt Thornton made 27 starts and was the California League Pitcher-of-the-Year, striking out 192 in 157 innings. Not only was he effective as a starter, he was the best in his league. As a former first-rounder he had a lot going for himself. Since then, as you know, he’s had elbow tendonitis, TJ surgery and neck problems. When Thornton got healthy it took him a little while to regain his velocity, but he seems to have made progress, although it’s certainly debatable. If he were to develop ANY kind of command, who knows?

    At 28 he’s most likely running out of time, but maybe one of the questions that the Mariners are asking themselves, besides whether or not to keep him on the 40-Man Roster is: could Matt Thornton–as was alluded to in the 2003 BP–trace out a John Halama type career path as a harder throwing swing man? Regardless, there was nothing wrong with seeing what Thornton could do as a starter in Tacoma. He didn’t pitch particularly well there, but I think that your “Idiot’s ploy” comment is a little harsh.

  14. DMZ on December 3rd, 2004 12:21 pm

    Can we please keep the Giambi/Sexson/etc discussion to the Delgado/Sexson thread, available for your enjoyment just down the hall?

  15. DMZ on December 3rd, 2004 12:28 pm

    Ahhh, the 2003 Mariners chapter in BP. Boy, that was a good piece of work. I wonder who wrote it? If only they were bylined.

  16. mike on December 3rd, 2004 1:59 pm

    I like to point this out every now and then, since it is so opposite of almost every major league starter (in fact, he may be the only one…): Russ Ortiz was a CLOSER in the minor leagues, before being converted to starter.

  17. Coach on December 3rd, 2004 2:07 pm

    Interesting tidbits about Grady Fuson. In Mariner Land, who would have the responsibility to stay abreast of available FO talent throughout MLB. If that person(s) were aware of an interesting candidate, who would make the call to pursue the candidate?

  18. Brent Overman on December 3rd, 2004 4:11 pm

    Along the Russ Ortiz landscape, when the Indians originally drafted Paul Shuey, he had been a closer throughout his career. They tried to rid him of his high leg kick and make him a starter, with disastrous results. Ortiz is an interesting anomaly, though.

  19. Bela Txadux on December 3rd, 2004 10:44 pm

    Yo Pat,

    I tend to compose a lot by the seat of the pants, and I don’t always go back to check all the numbers on some of my allegations; it’s like living (a little) dangerously, but I’m pressed for time, so I’ll accept the occasional rimshot. —But that’s not the case here with Thornton. I’m aware of his numbers in the Cali League; I’m also aware that he was 25 years of age at the time, extremely old for the league. He got there late because he’d floundered for years in the low minors after being drafted out of high school. I don’t have full the numbers for Thornton’s ’01 in front of me [see above], but my recollection is that it was not a great statistical package, his excellend K9 notwithstanding. Are you recalling that in ’02 the Mariners converted Thornton from a starter to a reliever—and then Matt blew his elbow?? That’s what I recall. In other words, Thornton had ALREADY been shifted to relief at age 26, only to have the New Crew in the Ms FO shift him back two years later. What was the probability of Thornton finally becoming an effective starter at, what was it, 27, 28? After pitching short innings around a major arm injury the previous two years? I’d rather draw to an inside straight.

    This is the matrix of my ire, even if I could be fairer, sure. If Thornton had been, say 25 years old in ’04 there is a boderline reasonableness to the attempt to re-start him. At his age with recent history, there seems to me no case. It was a total failure, and wasted most of another year of Thornton’s development time such as that is and maybe.

    Now, I’m willing to be entirely fair: if Jose Lopez takes off on an accelerated development trajectory after being forced into the majors at age 20, I’ll be the first to note this as a success for Bavasi and his development team. Dave thinks _this_ one was the fool’s move, and he may very well prove to be right. By the same approach, though, I’m inclined to give the back of my hand to the same lot for a gamble with Thornton—it could not possibly have been more than a gamble at bleak odds—that didn’t play.

    Regarding Rafael Soriano: I’m terribly disappointed by the idea that he’s going to end up a reliever. I suspect that you are right, Dave, because the Ms team had been nudging him—“slender righthander, late to pitching”—toward Mariano-dom even before The Elbow Follies. Of course, many starters blow the TJ and come back right as rain, throwing BBs better than Ezra. . . . I remember Pedro Martinez being shoved into relieving and then dumped by the Dodgers, too. I have always seen Soriano as a starter, and to me he’s one who should be given that role for at least two seasons or unless he’s just shelled before anyone even _thinks_ of making him a reliever. Soriano’s value as a ‘closer’ is, like, a third of what it would be as a starter, and the idea that the Ms want to take a high end arm and employ it at a massive markdown says loads to me about why they seem unable to finish the development cycle on their best guys. But this also explains to me why the Ms brain(t)rust are so determined to sign a frontend starter this offseason: If Soriano is projected out of the rotation, the organization definitely needs one more high end arm in the mix. What a total, total bummer. : (

  20. stan on December 4th, 2004 2:13 pm

    bela, from what i saw of soriano as a starter when he was in tacoma, he had some difficulty the second time through the line-up… i remember pedro pitching at cheney stadium when he was in the dodger system and he looked to me like he held his velocity deeper in a game than did soriano… if soriano is going to make it as a starter he will have to either develop more stamina so he can get his fastball by hitters the second and third time through the line-up or be able to change speeds like pedro… too bad pedro is not on the radar to sign with the mariners; soriano would then be in a position to learn from a master….

  21. Bela Txadux on December 5th, 2004 12:40 am


    I’m with you on the fact that Rafe S. was and is a very incomplete project, and without several improvements in his overall game he was and is not going to make it as a starter. I hesitate to even use the Pedro analogy, because even prior to his injury Soriano was far from having the secondary pitches that have made Pedro so great, even before we get to consistency with this hard stuff, which as you point out could be a major problem by itself. Maybe the Mariners brass are in a sufficient position from scout and pitching coach evals to determine that there is too much wanting in velocity and lack of development on his secondary pitches to warrant a starting projection; that’s beyond what I know. Now that Soriano has blown his elbow, I suspect the Ms braintrust aren’t goint to want him to throw his slider too much, and without that he has no shot at starting.

    What I saw is that: a) he had only _just_ learned to throw an effective slider, and it was a good, even plus pitch if he executed it, and b) he still hadn’t mastered a change-up, but was only just beginning to work with the pitch. To me, it would be to soon to write him off as a starter given that he didn’t really have development time enough to give him a reasonable time to work his slider and incipient change into his pitching patterns. Basically, he was throwing mostly fastballs because, mostly, that was what he _could_ throw; in effect, he was throwing too many fastballs, and this would impact his ability to go deep in a game.

    I’m sure you know all this, Stan, I’m just going through my thinking on it. I would really like Soriano to get a solid year of starting while throwing his slider and change, however good they are, at reasonable ratios and in appropriate counts and game situations before slotting him for relief. In ’03, he was brought up twice from need, and really didn’t get a full year of working on these secondary pitches; up in the Bigs, he was throwing mostly fastballs, both as a starter and once he came back to set-up, and even then didn’t even have his slider working at a minimally reliable level until late in the season. In ’04 . . . well, the word was ‘he showed up in camp hurt,’ not that I’m sure I believe that, and he wasn’t able to work on anything, really.

    I think a better mention than Pedro would be Johann Santana. There are various reasons why the Twins left him in a relief role as long as they did, but mostly it was a lack of confidence in his all-round game, to me. I see the same thing with the Ms and Soriano, although Rafael is obviously much farther back along the development cycle than Santana. The Ms just aren’t showing any _confidence_ in his ability to start, so in my view they are finding reasons to under-project him in that role—as opposed to simply letting him start with proper instruction and see what happens. That is what bothers me, really, because it is just so pervasive in the Ms organization in my view: they have too little confidence in the guys in their own system, and are pretty darn quick to reduce their roles and projections, and this is still very much the case with Bavasi, Player Development Specialist, in the chair as opposed to Gillick, Veteran Acquisition Specialist. The Ms would just rather go out and sign a veteran starter than really work to get the best out of their own guys, and to me that’s bassackwards at best, hence my ire.