Labels have always intrigued me, not least of all how they are used in baseball. Some, like “closer” or “run producer,” aren’t always accurate, but they are at least indicative of what the player does. Chad Qualls keeps finishing games despite necessitating the Chad Qualls Support Group (of which both Jay Jaffe and I were members, thanks to fantasy issues) in no small part because he’s a “proven closer.” Plenty of people might dispute the proven part of that designation, but it’s hard to say he’s not a closer. He finished 44 games in 2009 and 29 games in 2010, ergo he is a closer.

While few labels in the game are as malignant as closer, there is one that is inspires more debate.

A few weeks ago, I called Francisco Liriano an ace on Seth Stohs’ Podcast and caught a reasonable amount of flak for it, so I decided to look deeper into the issues surrounding the term. Since, unlike run producer and closer, it isn’t something you do so much as something you are.

As typically happens with these broad queries, the large question broke down into several more specific ones. While a longer piece elucidating the answers is coming, I thought opening up some of those questions as a starter ahead of some more concrete data might lead to an interesting discussion.


  • On staffs with multiple top pitchers—The Red Sox and Phillies come to mind—does being the second best pitcher preclude one from also being an ace?


  • Consistency is obviously very important to pitchers, but how many seasons does it take for a good pitcher to become an ace?


  • Acknowledging the ebbs and flows that all pitchers go through, should it matter if a pitcher is consistently above average or alternately great and mediocre?


  • Since award voting is a poor measure of a player’s actual skill, how often does a pitcher need to be Cy Young-caliber?


  • Cliff Lee has obviously been a force over the last few seasons, but the start of his career was much less auspicious. What role should his lesser past play in our evaluation of his current success?


  •  How much—if at all—should reputation or recognition matter?

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 think the description "Ace" is a bit like obscenity. You know it when you see it.

I seem to recall an earlier BP piece debating which pitcher was better, one with a 0.00 ERA in 75% of starts and 9.00 ERA in 25% of them or one who has a 4.00 ERA in all starts. AKA Oliver Perez. My poor memory suggests the inconsistent pitcher was more valuable than the consistent one.
I would say an ace is a pitcher who could face off against the best pitcher in the game without either team feeling like it has a huge advantage (i.e. anybody your team could match up against Roy Halladay that doesn't cause you to go "we're screwed"). Teams can have multiple aces. In regard to timeline, I think you need to look at about a season and a half on a rolling basis, this seems like a reasonable amount of time to both ignore a slump (think Lincecum last year), but acknowledge that factors can cause pitchers to lose "Ace" status (think late career Greg Maddux).
I agree with SC. If you have an "ace" you know you have an ace. Is King Felix an is? I think there can be more than one on a team, Lester, Buchholz, Beckett???, Lackey (who most would have considered and ace before this season). I think if you have an ace you feel like your chances are good that you will win the vast majority of the games he pitches, if you score on the others teams ace.
1) No, I think teams can have more than one... if the Yankees do in fact sign Cliff Lee, would anyone argue that either is not ace quality just because the other is present? It would be absurd.

2) A pitcher can be Ace-like at anytime, though it would likely take multiple seasons of preforming as such to have the label slapped on them.

3) Every pitcher throws the occasional clunker... Wouldn't the more important thing to look for be how many times a particular guy stops a losing streak or comes in and wins after a loss the day before? I don't think the average guy or mediocre pitcher even sniffs the Ace label...

4) CY Young's are a horrible way judging pitchers for this purpose, they don't grade the award on difficulty. Does anyone honestly think King Felix out pitches David Price for the CY Young this year if he was pitching in the AL East? I don't; and I'm not saying King Felix isn't an ace, like the first poster... you know it when you see it.

5) Why should Cliff Lee's seemingly mediocre past matter if he's done it over a period of back to back years... Its kinda like Hall of Fame voting in a sense, no? Shouldn't a pitcher be dominate at some point for a few years?

6)Reputations and recognition don't matter... Numbers do.

I think it would be interesting to see if there are any trends related to age, handedness or the number of years after being called up from the minors that guys achieve Ace-dom.
In response to Jquinton82's Hernanez/Price comparison:

Hitter that King Felix faced this year posed a combined .728 OPS. Hitters that Price faced OPSed a collective .739.

There is no question that Hernandez faced weaker hitters on the whole. That said, do you honestly believe that the difference is all that significant?

I know that this comment has nothing to do with the main point of the article. Looking forward to reading the piece when it comes out.
Great point, at least some of the increased degree of difficult Price faced was offset by the fact that Felix never got to face the haples Mariner bats.
I think that 'ace' = '#1 starter' = roughly the top 30 pitchers in baseball. (Hence the etymology of the word 'ace' and the distinction between an ace and a #2, #3, etc.) If every team in baseball got to draft their rotations from scratch, the first thirty taken are #1 starters. Aces.

Obviously one team may have a couple or three of the top 30 pitchers in baseball, and other teams may have none. There's also obviously going to be no sharp distinction between 30 and 31, but there are tiers and dropoffs. The set of tiers that roughly corresponds to 30 is the set of aces.

A pitcher can become an ace quickly, I think. Lincecum was easily one after his first full season; Kershaw and Price are there. Sandy Koufax lasted four years at the top - if he wasn't an ace, no one ever was.

I don't think 'ace' = '#1 starter'. Unless we're willing to stipulate that Bronson Arroyo and Mark Buehrle are aces, while Adam Wainwright and Jon Lester, neither the #1 starter on their team last year, are not. I'd also argue that perhaps the first 30 starters taken in a draft wouldn't be 'aces', but would be the players projected to provide the most value for the money I figured to pay them over the life of their contract with me, but that's another convo altogether. I would argue that the first 30 pitchers taken in a mock draft could not possibly be aces, because I don't think there are 30 bona fide aces in baseball.

I think to be an ace you have to be undeniably awesome, for lack of a better phrase, and you have to have demonstrated it over a long enough span (at least a couple years) to show that you're likely to repeat your awesomeness.
I didn't mean #1 on their own team. When I said that some teams (Yankees, Red Sox, Cards, Phillies) might have 2-3 of the top 30, I thought I made that clear; I think it's clear enough that Wainwright's a #1 starter, wherever he might happen to pitch in his own team's rotation.

Anyway, I had something like BurrRutledge's account in mind, though he's right to say that a guy should be expected to be a top-30 pitcher regularly to be counted as a an ace. Two-of-three years sounds about right, though a one-size-fits-all formula might be out of place here. A decade or more of stats is needed to appreciate Halladay. Lincecum was an ace after one full season. Johan Santana is a question mark now, no matter what 2-of-3 say - even what the last decade says. And so on.
I see "Ace" as a subjective term. An Ace is a pitcher who makes you pencil in a "W" every time his name comes up in the rotation. If your team is facing an Ace, you mentally write off the game as an excusable loss. An Ace is dominant. He strikes out a lot of guys and often seems unhittable. A contact pitcher can be an ace (Tim Hudson), but that's the exception. And yes, a team can have two aces. If you want to put statistical guidelines on the term, I'd say that An Ace has to be a minimum 5-win pitcher for at least two consecutive seasons.

This leads to a secondary discussion as to what defines a #2,3,4 or 5 starter.
I had this same discussion three years ago, when Seattle acquired 'Ace' Eric Bedard. I put a bit of thought into it at the time. I disagreed with Bedard's Ace distinction at the time, and I argued (quite convincingly if you ask me) that one season of realized promise didn't make him an Ace, even if he was the best pitcher on his team or even one of the top 10 in the league. Earning that title meant achieving and maintaining a level of consistency and durability both...

At the time, I think I put forward that any pitcher ranked in the top 30 in two of the past three seasons would make him an Ace in the upcoming one, regardless of which roster he was on.

With that criterion, by VORP, there are 18 Aces at the beginning of 2011:

2010 Top 10, and also Top 30 in 2008, 2009, or both: Halladay, Hernandez, Wainwright, Jimenez, Josh Johnson, Jered Weaver, Sabathia;

2 of 3 Top 30s: Johan Santana, Cain, Verlander, Kershaw, Lester, Lincecum, Greinke, Haren, Carpenter, Danks, and Oswalt.

Best of the rest / Just missed the cut (1 of 3 Top 30s):
'10 only: Hudson, Price, Buchholz, Hamels, Gio Gonzalez, Cahill, Latos, Myers, Wilson, Dickey, Sanchez, Carpenter, Liriano, Pavano, Duensing, Scherzer; (these pitchers have two more years to repeat in the top 30 to qualify as 'Aces' in the future).

'09 only: Jurrjens, Vazquez, Edwin Jackson, Wolf, Wandy Rodriguez, Happ, Buehrle, Washburn, Millwood, Lilly, Beckett, Garza, Feldman; (could qualify as an Ace in 2012 if they get another top 30 finish in 2011)

'08 only: Dempster, Ervin Santana, Matsuzaka, Sheets, Webb, Peavy, Billingsley, Lowe, Saunders, Nolasco, Duchscherer, Mussina, Baker, Shields, Volquez, Guthrie, Maholm (these pitchers are having their clocks reset... gotta get 2 of the next 3 seasons to qualify as Aces again...)

Given the rare occassion for a touted rookie to be deserving of the title after (or during) the rooked season, I'd be willing to offer an exception for a rookie if he makes the top 3 or top 5 in his debut season... Lincecum yes, Strasburg unfortunately, no.

So, this may not be a perfect definition or analysis for such a nebulous term, but I think it's a defensible starting point for the discussion.
whoops, Carpenter got listed as an Ace and also as "Top 30 in 2010 only." He's an Ace.

There may be a couple other mistakes in that list, too, unfortunately.
... like Hamels... he was Top10 in 2008, so he's an ace too... damn I shoulda done a more thorough double-check/edit.
Cliff Lee. Grumble.
I appreciate a lot of the posts already made, and want to say thank you to BurrRutledge for the in-depth post. I however have a very different view point on the term "Ace".

If a SP were to become an unrestricted free agent right, would there be a feeding frenzy? Would everyone covet what you have, and in turn fear it at the same time? That is an "Ace". When you can reasonably deduct that a 30 game starter, will win 60% or more of those games, without any worry...that is an Ace.

I don't think it matters where you fall in the rotation at all. Before this season Jake Peavy could have been considered an Ace, but when he moved to Chicago he became the #2, because Buehrle has pitched every opening day for what seems like more than a decade, 8 years in actuality and a franchise record.

I think going by a 2/3 standard for performance is a bit short sighted. If Brandon Webb were to win 19 games this season, after missing almost 2 full season to injuries, would he not be considered an Ace again? Durability, consistency and reliability make a good pitcher, great. But not necessarily an Ace. That is the creme de la creme. Let's look at it this way. There a 4 Aces in a 52 card deck. So let's take the 150 SP (30 teams x 5) and that should yield roughly 10-11 Aces.

My question to everyone is, who are the consensus 10 Aces in baseball today?
Thank you for the shout out. As I said, I put a lot of thought into this a few years ago.

I can also agree with your subjective assessment of what an Ace should be: "coveted." But how do you define it objectively? I don't think Buerhle or Peavy fits that definition, especially if you're only going to limit it to ten pitchers going into 2011.

Injuries are a tough nut. I put in the 2 of 3 qualification to allow for a slightly down year, or time missed due to injury. I guess there's not much wiggle room for somebody out for two years. I guess you might call him, injured-Ace, or something like that. Former-Ace for a resurgent guy like Hudson.

Keep in mind that there are plenty of other terms besides "Ace" that denote valuable skills, caliber, and potential, such as "#1" (if he is the team's #1 even though not an 'Ace'), "phenom" (for that young kid who everybody knows will be an Ace if he stays healthy, "work-horse," "star," etc. #1 Star workhorse Mark Buerhle. Still doesn't make him an Ace (at least by VORP, which I've heard arguments undervalues him consistently, so maybe it introduces a different argument about which metric to use to objectively define Ace).
If I have to limit it to 10 to fit your limitation, then it's going to be the best pitcher from the most recent season who was also a Top 30 in at least one of the prior two seasons.

By my revised definition above, these are the Aces:
Jered Weaver
Johan Santana
Matt Cain

Price and Buchholz would have made it from 2010 alone, except they don't have the "proven track record" of 2 Top 30s in the past 3 seasons. The consolation prize for each is they can still be called "a phenom." Kershaw also came real close and doesn't have the track record...

It's surprising to see that Santana had such a good year, #12 in VORP. Kinda a stealthy good year, I suppose. If he could be signed for a 1-season deal right now, would he still get 'Ace' money?

By limiting the list to 10, noteworthy non-Aces include Lincecum, Verlander, and Lester, but they were simply out-VORPed by Hamels, Santana, and Cain in 2010...
Upon further review, Cliff Lee bumps Cain out of the top 10.
One qualification that I have for an ace that hasn't been brought up here is how long a pitcher CAN last in a game (effectively) if called upon, and how long he typically DOES last.

A pitcher is, in my view, probably not an ace if he can't be reasonably expected to go 8-9 innings IF NECESSARY. Obviously everyone has ineffective days and I don't see any evidence that anyone can make himself have a better game by wanting to, but there seems to be a skill of going deep into or completing the occasional must-win game.

This skill should not be confused with the separate skill of putting going fairly deep into games often (averaging around or over 7IP/start).

I would require both of these skills to call someone an ace. Of course, that pitcher must also have the skills and the track record for me to want him to pitch must-win games and average 220IP/season--I'm not trying to bring Livan Hernandez into this discussion.) Basically, an ace needs to be someone the team can trust to handle a playoff game more or less by himself and to take a big percentage of the team's total innings.

Francisco Liriano has elite talent--ace talent--but he just doesn't pitch enough of the Twins' innings to be an ace. And when he duels CC Sabathia in the playoffs, he's outlasted by two or three innings.

To give Andrew's post a reply, here are 10 guys that I think qualify as aces. This is gonna be off the cuff, though, but I don't mean the inevitable oversights personally:

1. Roy Halladay
2. Tim Lincecum
3. Felix Hernandez
4. Cliff Lee
5. Jon Lester
6. CC Sabathia
7. Adam Wainwright
8. Josh Johnson
9. Zack Grienke
10.Cole Hamels

This list is only sort of in order and really, any of the following could be in the 10th spot (so maybe I believe in only 9 aces...): Clay Buchholtz, Chris Carpenter, Justin Verlander, Jared Weaver, maybe Ubaldo Jimenez.
I gotta revert back to my original definition, Andrew. I just can't see limiting it to 10 Aces. I like my 19 guys from the definition I compiled a few years ago better (and I admit it's a confirmation bias):

20 Aces: Halladay, Hernandez, Lee, Wainwright, Jimenez, Josh Johnson, Jered Weaver, Sabathia, Hamels, Johan Santana, Cain, Verlander, Kershaw, Lester, Lincecum, Greinke, Haren, Carpenter, Danks, and Oswalt.

Their is undeniably some variation within this list, but I think of all them would be highly coveted.
I subconsciously use fantasy sports now to help me with terms like "ace". Basically, I want top-five position players at every position in fantasy baseball. When it comes to the pitching staff, because we use a lot more SPs than 2Bmen, that number jumps to fifteen. In other words, I want as many of the top-fifteen SPs in the game. A starter who is in the conversation for being a top-15 starter is an "ace" in my book*.

It passes the smell test. Each team in baseball has their own #1 starter. I think the term "ace" is more selective than #1 starter. We can say that maybe half of those #1 starters are "aces" and that teams can have more than one ace.

* note: this means there are possibly more than 15 of them. If your personal #15 and #16 guys are Haren and Carpenter, for example, and you think they are about equal, they can both be considered "aces".

Just my three cents.
Aces make up 7.7% of a deck of cards.

Seems like if you are using that term, that would imply out of 150 starting pitchers there would be 11 or 12 of them that would be an ace.

In this agrument, the sabermetrics lose out. In my opinion, this is the sentence uttered by an opposing manager that defines an ace: "Oh crap, we have to face this guy today."

Strat-O-Matic breakdowns for simulations can filter out your aces each year. Take the list each year and see the pitchers that repeat three out of four years and you have your aces.

Using this example over some of the prior lists, guys like Greinke and Kershaw aren't aces. Greinke has been a bit inconsistent and Kershaw hasn't pitched long enough. Young pitchers like Lincecum are because they have been dominant since joining the league with only smatterings of inconsistencies over short time periods. And folks like Halladay are obviously aces because they've done it multiple years over a greater time period than five years.

Yep, it's subjective.
I largely agree with those who say it's subjective, you know it when you see it, etc., but I also think you can kind of frame the discussion with numbers in the interesting way some have done here. I also think ace is a slippery enough term that it can be used in various ways. Is R.A. Dickey an ace? Maybe it's too soon to say that, but you can say he pitched like an ace in 2010.

I like to think of ace as a position. So how many are there? Well, not 30, because if you look at any other position, it's clear there aren't 30 guys filling it who most reasonable people would consider MLB quality starting players at that position. In other words, there aren't 30 starting MLB shortstops. Some teams are filling as best they can. How many? Well, I tried to eyeball that for this year with some blunt-object numbers. How many players at each position qualified for the batting title while putting up an OPS that's at least league average minus 10 percent for their position? (Note I used 400 PA, not 501, for catchers.)

The answer is an average of 16.75, which I'll call 17. The cutoff point for each position is between: Cuddyer/Pena, Hill/Figgins, Andrus/Aybar, Kouzmanoff/Callaspo, Lee(!)/Cabrera, Gutierrez/Morgan, Guillen/Zobrist, Suzuki/Rodriguez. (Note of course that this ignores defense, among other things it ignores.)

Probably a little generous, if this is an accurate comparison. Maybe it should be league average minus 5 percent. Or maybe ace pitcher is a little different than other positions, there are more of them, because there are five starters per team. Dunno.

But 17 as a number doesn't seem that far off to me. It's close to another definition I've played with: Good enough to be a starter on a winning team, which would be roughly 15 guys.

Going by another blunt instrument, ERA among qualifying starters, the top 17 in 2010 were: Hernandez, Johnson, Buchholz, Wainwright, Halladay, Garcia, Price, Oswalt, Hudson, Dickey, Jimenez, Kershaw, Latos, Cahill, JSantana, Weaver, Hamels.

So Santana/Weaver as the cutoff point for 15, Hamels/Sanchez as the cutoff for 17. (The rest of the top 30, starting at No. 19: Myers, Cain, Sabathia, Lee, Carpenter, Gonzalez, Lester, Hanson, Wilson, Verlander, Kuroda, Lincecum.) We're talking about very god pitchers all down the list, though by the time you get to the bottom, you're getting into some guys who didn't always pitch like aces in 2010.

Top 15 or 17 ERA is a lot more strict for starting pitchers than it is for position players. So it's pretty clear that those numbers shouldn't match up. But I think it helps to think about it that way to understand why the number of aces should be considerably lower than 30.

I'm pretty comfortable with any number between about 15 and 20, defined using whatever metric you want to use. One-year ERA wouldn't be the best, of course, although it's a good starting point if you want to talk about guys who "pitched like an ace this year."