The other day, Ben Lindbergh and I were bantering about relievers who are all of a sudden awesome. Will Smith, Louis Coleman, Neal Cotts, some other guy whose name I have already forgotten. (Ed. note: Kevin Siegrist.) And Ben asked, as somebody always will, why teams still pay for relievers when there seem to be an infinite number of humans capable of pitching like Jonathan Papelbon for a year.

The A’s don’t pay for relievers. Their top seven this year are making about $8 million and cost almost nothing in talent to acquire. The A’s also have the second-best bullpen FRA of the post-expansion era, behind only this year’s Braves. They make it look so easy! But this incredible success is also the answer to Ben’s question. Teams still pay for relievers because the A’s way of doing things takes so much friggin' effort.

Here’s the A’s bullpen this year, listed by acquisition method instead of name:


  • Seven-figure free agent

Set-up men:

  • Small part in big trade
  • Converted first baseman

Middle relief:

  • 28-year-old w/ almost no big-league experience picked up off scrap heap
  • 28-year-old w/ extensive big-league experience picked up off scrap heap
  • 31-year-old w/ experience picked up off scrap heap
  • Low-minors return in trade of undesirable veteran

That’s essentially five players acquired for nothing, one more acquired as somewhat of an afterthought in a bigger trade, and one player acquired for a medium-sized investment. But building a bullpen is a war of attrition, and those seven players stand atop a mountain of corpses. Here’s what else went into putting those seven together:

Trade undesirable veteran for unheralded minor-league reliever
Outcome: Jerry Blevins, 260 innings with 120 ERA+ as an Athletic;
Could have been: Kristian Bell (High-A reliever acquired for Marco Scutaro); David Shafer (Double-A reliever acquired for Kirk Saarloos); Justin Souza (High-A reliever acquired for Jack Hannahan); Ross Wolf (Triple-A reliever acquired for Jake Fox); Bruce Billings (Triple-A reliever acquired for Mark Ellis); Jason Rice (Triple-A reliever acquired for Conor Jackson).

Blevins was (with Rob Bowen) half of the return in the 2007 mid-season trade of Jason Kendall to the Cubs. Kendall played 57 games for Chicago. At the time, Blevins had just been promoted to Double-A and was having a fantastic season as a 23-year-old reliever. But he had entered the season with three minor-league seasons and a near-5 ERA, almost all of it out of the bullpen. He was a 17th-round draft pick.

Math: Trade seven players you don’t want anyway for seven minor-league relievers. Most will be released before they ever move up a level in your organization. One will be a useful big-league arm.

Convert a first baseman to the majors.
Outcome: Sean Doolittle, 2.54 career FIP, 2.33 career FRA
Could have been: Daniel Petitti (converted catcher); Jonatan Santana (converted shortstop/utility infielder)

Doolittle was, of course, a first base prospect whose hitting career had been interrupted by injuries. He started throwing on the side, the A’s decided to give him a shot on the mound, and a year later he was awesome. Two years later he is also awesome. A scout once told me that, on average, most teams try this with a failed prospect about once a year, and the A’s have tried this with players who weren’t Sean Doolittle. Last year, Petitti, a 37th-round catcher who had hit .174/.249/.239 in two seasons, tried to make the switch. He struck out 17 batters per nine innings in rookie ball, but the A’s let him go after the season. (He appeared in an independent league this year.) Santana was an infielder who hit .236/.320/.314 in six seasons. Before cutting him loose, the A’s put him on the mound four times. (He wasn’t good.) The last time the A’s converted a position player into a major-league pitcher was Marcus McBeth, who appeared 23 times in the majors in 2007.

Math: Take the few dozen position players who wash out every year; find the best arm among them; get a major leaguer every five or 10 years.

Trade a good player for a prospect, or multiple prospects; get minor-league reliever thrown in on the side.
Outcome: Ryan Cook, 135 innings with a 186 ERA+ as an Athletic, baseball’s 11th-best FIP this year.
Could have been: Danny Farquhar (Rajai Davis trade); Fernando Rodriguez (Chris Carter/Jed Lowrie trade); Jordan Norberto (Brad Ziegler trade); Josh Outman (Joe Blanton trade); Jamie Richmond (Mark Kotsay trade).

Cook is a former 27-round pick who had pitched 30 innings above Double-A when the A’s got him as the less-famous part of the Trevor Cahill trade. In those 30 innings, he had 24 strikeouts and 18 walks. “A move to the bullpen in 2011 helped jumpstart his career, as he missed bats with a 92-95-mph fastball and a slider that is solid but will need to improve in order for him to have above-average major-league value,” we tacked onto the end of his Transaction Analysis.

Math: Trade six good players for sixish prospects; ask for a 23-year-old reliever, instead of a 19-year-old starter with control problems, as a throw-in; one will turn out to be an All-Star set-up man.

Pick up freely available mid- or late-20s reliever with very little big-league experience
Outcome: Dan Otero, 1.41 ERA with Oakland, 4 Ks per walk
Could have been: Will Harris (signed 4/3/13), Josh Stinson (3/29/13), Pedro Viola (1/25/13), Sandy Rosario (12/10/12), Danny Farquhar (6/9/12), Fabio Castro (1/4/12), Travis Schlichting (1/4/12), Evan Scribner (10/25/11).

Pick up freely available mid- or late-20s reliever with lots of big-league experience
Outcome: Jesse Chavez, 3.36 FIP/1.0 WARP for A’s this year.
Could have been: Fernando Nieve (8/3/13), Carlos Fisher (1/28/13), Justin Thomas (11/15/12), Garrett Olson (10/27/12), Rich Thompson (4/20/12), Travis Blackley (4/3/12), Edgar Gonzalez (11/4/11).

Pick up freely available veteran in his 30s
Outcome: Pat Neshek, 57 innings with 131 ERA+ in two seasons with Oakland.
Could have been: Mike Zagurski (8/18/13), Hideki Okajima (2/12/13), Brian Gordon (12/22/12), Chris Resop (11/30/12), Jeremy Accardo (8/15/12), Chris Ray (7/13/12), Markin Valdez (1/4/12), Jim Miller (11/11/11).

Over the past 24 months, that’s at least 26 veteran relievers that the A’s have signed, claimed, purchased or traded a nothing piece for. A few of these guys were cut free before they’d thrown a pitch; a few more pitched a small handful of innings with the big club. Miller and Scribner were useful in the A’s bullpen last year.

The math: Pick up a dozen or more veteran relievers a year, get about two or three average arms, and maybe one fluky career year.

And, finally, make a moderate investment in a closer type, pray it works out.
Outcome: Grant Balfour; 192 innings, 154 ERA+ in three years with A’s.
Could have been: Brian Fuentes

Easy as Balfour makes this step look, there have been plenty of half-measure attempts at getting a half-formed relief ace that haven’t worked out during Billy Beane’s tenure: Octavio Dotel, Arthur Rhodes, maybe Juan Cruz fits here, maybe Joey Devine.

The math: Sign two veterans who each cost around half what a regular closer costs, for around half as many years; end up with a closer.

So yes, you can build an amazing bullpen. You can spend about $8 million on the entire group. You can do it by trading things you don’t want, scavenging for things nobody else wants, and asking other teams nicely, and letting your low-minors position players indulge their fantasies. It just takes 30 or 40 arms to make it work. And heaven help you if, instead of six or seven useful relievers out of those 40, you end up with only four.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
I've never really understood the proven closer thing, and as a Blue Jays fan, watching BJ Ryan explode after one season with four expensive years still on his deal, I hope the team never spends big on relief talent again.

They've built a pretty solid bullpen this year in much of the same way as the A's, albeit with a little more salary expenditure.

Janssen (failed starter), Cecil (failed starter), Delabar (acquired from SEA for AAAA outfielder), Loup (mid-round pick), Santos (acquired from CHW for non-prospect, converted shortstop), Neil Wagner and Juan Perez (minor league free agents), Dustin McGowan (converted from starter because of injuries).

I think the only major league free agent in the bullpen is ol' Darren Oliver.
1) This is how a good GM earns his money. How many millions do other teams blow on less effective bullpens.
2) Proven Closer means nothing. Proven effective short relief pitcher is the real question.
$8 million? Was that Ryan Madson's deal?
There is nothing about this article that doesn't apply to every other aspect of the 2013 first place Oakland A's. Former catcher Josh Donaldson, anyone? Multiple-team reject Brandon Moss? Talented injury reclamations Coco Crisp and Jed Lowrie? Josh Reddick - way better as a surprise offensive contributor in a perhaps fluke 2012, but still as good a defensive outfielder as there is (when healthy). And the expensive-by-A's-standards parts, Yoenis Cespedes and Chris Young, neither valueless but wouldn't you know it neither nearly as efficient a way of spending money as Donaldson-Moss-Crisp-Lowrie.

Of course, there is math, and I'd love to see BP do the whole team carefully because I sure can't. (1) The failure to do it from '04-'11 when a series of injury reclamations, middle-cost parts, and Daric Barton didn't pan out. ('06 they did it for a minute with a crazy one-time comeback from Frank Thomas, which of course fits the model perfectly.) (2) The cost to ticket sales and jersey sales that comes every time someone who establishes themselves as a popular player having a pretty good year is let go.

This last point is, unfortunately, the piece none of us SABR types wants to acknowledge. I love the A's and their baseball management and hate Wolff and their business management like any good Oaklander does, for reasons that are obvious. But their baseball management lacks a certain understanding of the long term damage you do to revenue by ill-timed trades of popular players, and their business management, hell bent on San Jose, wants them to keep doing it that way. In what other ballpark in the United States is everyone wearing a logo jersey wearing the name of a player who hasn't been on the team in five years? (The only logo jersey I will buy under these circumstances says Henderson, 24.)
Nice write up.
" A scout once told me that, on average, most teams try this with a failed prospect about once a year"

This year's appears to be the Baseclogger himself, Jeremy Barfield, but it seems the A's are a bit more aggressive than most teams. It'd be interesting to see if that's true, or if every team has a couple of kids who flamed out in rookie/short-season without famous twitter accounts who are in instructs learning to pitch.

As an aside, I wonder how long teams give a position player to pull it together before they suggest pitching. How many years, and what minimum level of production would the M's have accepted from, say, Adam Jones before they moved him to the mound as so many wanted them to do in 2004?
Great article. You always hear about successful bullpens being built on the cheap with the implication being that it's super-easy to put one together. The reality is that you have be lucky/astute enough to recognize guys with potential ahead of time, and still have to sift through a lot of chaff to succeed.

I'm curious - does anyone know how well bullpen cost and performance correlate? Presumably, the reason teams pay more is to avoid the risks layed out here.
For an example of how building a bullpen from nothing requires a great deal of luck as well as skill, look at the Houston Astros.
Not a true conversion but in the not too distant past the A's took a AA-AAA starting pitcher and had him pitch more horizontal in Brad Ziegler.
Neil Huntington has built damn fine bullpens here in the Burgh by doing
the same thing. look at his current one: his best (Grilli) signed from minors,
#2 Melansen via trade (for ace reliever), Wilson & Morris converted starters,
and Mazzaro & Gomez cheap trades.

Pretty fine!
It blows my mind that teams spend exorbitant amounts of money on their bullpen. You could write a similar story about the Atlanta Braves. David Carpenter and Varvaro off of waivers. Jonny Venters and Kimbrel were just flamethrowers with no control in the minors, never even considered for prospect lists. Downs acquired for a warm body. Walden acquired for a warm body formerly known as Tommy Hanson. Avilan was a nobody in the minors. I think even O' Flaherty was a waiver grab from Seattle. All of those guys are cheap (though Kimbrel will start getting expensive). I'm just not sure why you'd ever spend so much on any one reliever.

I think a big part of it is knowing when to throw in the towel on your SP prospects, and just turn them into two-pitch bullpen machines. The pipeline for bullpen success is still strong for Atlanta, too. Graham, Hursh, Jaime, Cabrera, and Thomas that all look like they have floors of high-leverage reliever. If any current relievers start getting expensive, they'll let them walk and slot in the next guy.

But I'm sure Brandon League is really paying off...
Thanks for the article. It frustrates me that some people think you can turn any failed starter at AAA into a bullpen piece.