We Need New Terminology For Starting Pitcher Prospects

I love Kevin Goldstein’s Twitter feed. He’s conversational, blunt, and isn’t afraid to say what he feels. This caught my eye last Thursday:


I really, really hate the term “#1 starter.” I think its a useless thing to say, because it doesn’t mean a thing. At any given time, there are at least 150 starting pitchers in the major leagues. Technically, 30 are #1 starters. Obviously, not all of them are top quality. I’m willing to say that the Rockies do not have a #1 starter right now.

If we consider a #1 starter, say, the top 20% of all MLB starting pitchers, we’re probably looking at something like this list. Those are all MLB pitchers who threw at least 350 innings between 2009 and 2012 sorted by ERA-. The #30 pitcher (20% cut off) is Dan Haren. I think that’s a pretty fair cut off, which includes quite a few prospects with that kind of ceiling, but Kevin Goldstein doesn’t:

So, he’s actually talking about consistent Cy Young candidates. To call a prospect a potential #1 starter is to call him the next Roy Halladay, C.C. Sabathia, Justin Verlander, or Felix Hernandez. Instead of top 20%, he’s probably referring to something closer to two standard deviations, or top 2.5% – one of the top 4 starters in the league. Over the 2009-2012 time frame, those starters (by WAR) are Zach Greinke, Justin Verlander, Cliff Lee, and Roy Halladay (Sabathia is #5). If this is what Kevin Goldstein is referring to, I think it is very fair to say that only prospects like Gerrit Cole and Dylan Bundy are #1 starter prospects.

But that begs the question: How effective are these labels? Why do we use them in the first place? I tried to design a prospect rating system to get around this problem, but defining pitchers by their peak WAR:

Something like this is how we should talk about prospects. Kevin Goldstein would probably say that Manuel Banuelos is a potential #3 starter. By definition, that’s a league average starting pitcher. In 2012, that means an ERA of 4.37. Banuelos is a much better prospect than that. “#3 starter on the best team in the league” might be a better way to say it, but at that point the label loses most of its meaning.

This isn’t a post to pick on Kevin Goldstein. He’s a great baseball commentator because he’s both very smart and very sure of what he wants to say. His examples clearly illustrate his intended meaning of the terms he uses. But the term still leads to ambiguity, needless argument, and confusion. I would give Manuel Banuelos the #1 starting pitcher prospect title, because I think he can be one of those top-30 in the league type guys. But that’s my definition, not everyone’s. I also think that Banuelos is the type of pitcher who could put up a 3.50-ish ERA and fairly consistently give a team a solid 5 WAR in his prime, if he hits his ceiling.

E.J. Fagan been blogging about Yankee baseball since 2006. He lives and works in Washington, DC.

5 thoughts on “We Need New Terminology For Starting Pitcher Prospects

  1. I think the distinction you may be looking for is ace vs. #1 starter, which could have different meanings. I tend to think of ace as something like a top 10 pitcher in the league, whereas a #1 could still be top 30.

    A lot of the distinction in my mind has to do with consistency. You expect aces to be among the best pitchers in the league every year, whereas somebody who’s a #1 but not an ace (for instance, Matt Garza, since Mike’s post put him in my head) may have more fluctuation in his performance, pitching like a #1 guy some years and like a #2-#3 other years.

    When it comes to prospects, these labels tend to be fairly meaningless. They usually refer to ceiling rather than the highest-probability outcome, and this is not always made clear by people who use the labels, causing confusion.

    Of Bundy, Walker, and Cole (Goldstein’s trio), I would be surprised if more than 1 of them became a bona fide ace (by my definition) starter in the bigs. Not because they are not great prospects, but because becoming a true ace is so hard to do. Injuries obviously play a huge role, as do adjustments to major league hitters, changes in repertoire, and many other factors.

    • bg90027

      I agree with you that few reach their ceilings and would add that there are also a bunch of pitchers (especially amongst ones who aren’t 6’4″ or taller and don’t throw gas) who blow right past their projected ceilings.

    • Reggie C.

      Is Wandy Rodriguez a decent comp for Banuelos?

      • T.O. Chris

        I’ve always gone with Gio Gonzalez because they share a similar velocity/repetoire and they posted similar walk rates as they rose up the ranks of the minor leagues. If I had to put numbers on it I would say in his prime Manny’s probably looking at being a 3.5-4 WAR guy, A solid number 2 starter but not an ace. However since he’s been so drastically overhyped (as with most Yankee prospects) I think that would be considered a dissapointment by some in the fan base.

  2. bg90027

    I agree that the term lacks precision but I think frankly you want less precision in projecting prospects rather than more. I generally prefer terms such as “front of the rotation”, “mid rotation”, “back end” and BP/spot starter. There are only a few “can’t miss”/easy to project types so why should we use terms that imply otherwise.

    Also disagree that Banuelos is a potential #1 but what do I know? And I’d love to be proven wrong about that.

Comments are closed.