Zambrano an Ace?
Next year has arrived in Chicago, and with Rich Harden gone, even more pressure will be on starter Carlos Zambrano to carry a heavy load for the Cubs rotation. His season did not get off to a good start yesterday, as he was lit up for 8 runs in only 1.1 innings, including home runs by Jason Heyward and Brian McCann. For the Cubs to challenge the Cardinals for the Central division title, Zambrano will have to pitch like an ace, as without him, the Cubs are likely a sub-.500 team.
Unfortunately, Zambrano certainly hasn’t been an ace pitcher over the last four years. Since 2006, his xFIP hasn’t been outside the 4.20-4.65 range, and only a stupidly low 5.6% HR/FB rate managed to get his 2009 FIP to 3.61. It’s really quite simple – Zambrano walks too many to be an ace at this point. 2008 was the only time in the same four year stretch in which Zambrano managed a walk rate below 4.00 per nine innings.
Somehow, though, Zambrano has managed to post ERA’s under 4.00 for all of the aforementioned years – in large part relying on BABIPs lower than .280. Zambrano has pitched for three above average defensive teams (from 2006 to 2008). Last season, the Cubs fell to a below average team UZR, and Zambrano’s BABIP skyrocketed to .308. Although that is by no means conclusive, it is possible that solid fielding behind him is in part responsible for Zambrano constantly outperforming his FIP. If that has disappeared this year, Zambrano could be in trouble.
There may be some heavy regression in line for Zambrano – both in terms of career BABIP and in terms of home runs allowed – and Monday may only have been the beginning. Zambrano should still be an above average pitcher, but he no longer has the ability to constantly win games by himself that the true aces do. If Zambrano doesn’t have the position players playing great baseball behind him like the Cubs had in 2006 and 2007, the Cubs will be on the outside looking in come September and October.
Jack Moore's work can be seen at VICE Sports and anywhere else you're willing to pay him to write. Buy his e-book.
So, in other words, guys like Alfonso Soriano shouldn’t be in the field.
And Ryan Theriot shouldn’t be at SS double-clutching his way into Cubs’ fans hearts.
Yet his career UZR at SS is 11.2 (3461 innings), I don’t see anything wrong with theriot’s shortstop.
IS there a worse OF than Soriano?? I mean honestly…there cant be. At least he is playing for the league minimum.
Jose Guillen would like a word with you.
Vernon Wells feels it should be at least a 3yr minimum to qualify.
His average FB speed has declined every year, and as he’s become more reliant on nibbling the corners then wiping guys out with his previously good slider FB combo. The only problem with that is his control isn’t good enough with any of his secondary pitches to be that guy. Whether you see him in person or are just looking at the stats it’s pretty clear he’s become a 2nd or even 3rd starter. The problem with that is he’s the 3rd highest paid pitcher in the NL. Yeesh.
In over 1500 major league innings, Zambrano has a .277 BABIP against(and a .267 from 04-08) compared to a ,298 major league average, along with team averages right around that point as well. When it gets to that point in terms of shear number of innings, don’t you kind of have to stop assuming regression to the mean and start assuming there’s some sort of inherent skill there?
That’s certainly a possibility – I think we’ll see for sure this year, after a .308 BABIP last season.
I think this year will definitely show a lot more. Cubs generally have a very good defense but had some playing poorly last year (Soriano and the mess that was put at 3B), some playing out of position (Fukudome) and some who shouldnt be on the field at all (Bradley).
Moving Fukudome back to right, having a healthy Ramirez, and if Soriano can be even average this is again a well above average defense.
The BABIP numbers should come back down once again.
Regardless Z is no ace.
That’s not how it works. Over a large enough period of time… you still regress to the mean. The amount of regression is determined by the size of your sample, of course – the fewer observations you have, the more you regress.
BABIP gets such heavy regression that you would still expect him to move back a lot toward average.
Also, remember that many of those 1500 IP are very old and not very relevant to the pitchers current true talent. I wouldn’t consider his 2005 .258 BABIP important to a current projection any more than I would consider Pedro’s 8.6 K/9 in 2005.
You do realize he has been regressing towards the mean for the last 4 years straight, right?
Here is his career as a whole
.338 (01)
.286 (02)
.291 (03)
.284 (04)
.258 (05)
.265 (06)
.275 (07)
.277 (08)
.308 (09)
If there was an inherent skill there, its obvious he has been losing it steadily since 2005.
Other cubs starters have posted similarly low BABIP’s over the last few years, and not just the ground ballers, ted lilly has as well (with soriano, and bradley in the outfield!).
That probably has more to do with Lilly’s extreme fly ball tendency’s(particularly infield flys) than anything. The team’s BABIP agains from 04-09 are .295, .290, .291, .286, .284, and .291. So yes, while those are low, Zambrano well outperformed them, and has done so while being an above average groundball pitcher. So while yeah there’s probably some regression that is/has happened, it definitely appears that Zambrano has some sort of skill that allows him artificially low BABIPs, and that can’t just be glossed over.
Plus they had Harden and Rich Hill as basically the only other FB pitchers during this stretch, who also drew a lot of IFFBs.
Thought this topic had already been pretty exhaustively covered both on Fangraphs and elsewhere over the past several years. Did I accidentally hop in the Hot Tub Time Machine? Oh, yeah, there’s Mark Prior doing towel drills- I must have gotten my relaxation and time traveling hot tubs mixed up again.
It is pretty infuriating to listen to the cubs talk about Zambrano as their ace. Dempster, and Lilly have been far better in recent years, and an argument could even be made for randy wells being better then Zambrano. Zambrano is the cubs ace in salary only.
Could the author – or anyone – please give us beleagured readers a detailed, formal definition of an “ace”? It seems like this is where the article above – and others like it – should begin.
A guy who’s consistently your opening day starter, your highest paid starter on your team, the starter that the manager always refers to as “our ace”, the first pitcher in regards to the starting five rotation.
Thanks for your reply, but I’m looking for a stats-based definition. (This is Fangraphs, right?) All of those sound like pretty dubious measures to objectively judge a pitcher’s performance.
Another great pitching project brought to you by the kingmaker himself – Larry Rothschild. Seriously what does this guy have to do to lose his job? Name 1 pitcher (other than 2008 1st half J. Marquis) that Rothschild has made a better pitcher. Z got by on talent (not brains) alone but he has nothing else to fall back on because Rothschild does not know how to teach. What I wouldn’t give for Don Cooper or Dave Duncan even for a month.
Randy Wells? Or how about Ted Lilly? Hasn’t he been a better pitcher since joining the Cubs? I guess my question for you is how can we, as fans, determine whether a pitching coach has a positive or negative impact on a player? Or any impact at all?
Correlation?
Ryan Dempster would like a word with you.
Yes, hzl, I suppose that’s the word I was looking for.
I consider a ace being like a great quarterback, every team does not have one and there is probably only 5 to 6 aces in baseball according to being able constantly win games on their own. So every team has an ace/best of the five on their team. When Z got his extension he was considered the best on the Cubs staff and that is where the leverage came in to get a big boy salary and it also helped that Hendry was/is the GM.
I’d define league aces as pitchers within a certain percentile in ERA or whatever metric you wish to use among all other pitchers, over a reasonable period of time; similarly, I’d consider a team ace to be within the high percentile of team ERA over a period of time
Lilly’s (very moderate) improvement should be chalked up to the switch to the NL from the AL East, in my opinion
Thanks – something like this is what I’d like to see someone do. Maybe ERA isn’t the best metric, because it’s defense-dependent, but it would be a start.
I would define an Ace as anyone who projects to be one of the top 15 pitchers in baseball. The guys that would be the best pitcher on an above average team.
Obviously year to year consistency plays a role, but I think the projections do a good job of figuring out who the best guys are. That being said, my “FANS aces” would be
Tim Lincecum
Zack Greinke
Roy Halladay
Felix Hernandez
Dan Haren
Cliff Lee
Justin Verlander
Jon Lester
CC Sabathia
Adam Wainwright
Josh Johnson
Ubaldo Jimenez
Javier Vazquez
Josh Beckett
Chris Carpenter
Johan Santana
I realize that’s 17, but Carpenter andSantana were tied with Beckett. Also, I thought Santana-Hanson was a good place to draw the line.
The “CHONE aces” would be the same, only with Chad Billingsley, Roy Oswalt, and Colby Lewis replacing Carpenter, Santana, Johnson, Hanson, and Jiminez.
Obviously this isn’t a perfect method, since, calling Colby Lewis an “ace” is more than a little ridiculous. But I think it’s a good start.
With the exception of Vazquez (only one dominant year) and Beckett (uneven performance over the last couple years), I’d be hard pressed to argue against that list. All of those guys are able to legitimately shut out any team when their stuff is working right, even the two I take some slight issue with. I don’t think it should be based upon the top N pitchers though. I think a better metric would be to figure out their contribution to win probability when they start a game, and set a threshold. Most of these guys are likely to put in 6 or 7 innings, giving up just a couple of runs.
Using win probability as a threshold to define an “ace” sounds like a good idea, too. And I like Steven Ellingson’s list as well – I’d not put up much of an argument that any of those guys could be called an ace, though the caveats you note for Vazquez and Beckett are valid.
I also think that calling Carpenter not an ace is ridiculous. When determining an ace, it should be who is going to go out there and make the opposition look weak and feeble every time he’s on the mound. While I do believe that durability is a very important trait for a starting pitcher and his value over replacement pitcher should reflect that, the “Ace” moniker is independent of durability. “Workhorse” is the term that is more about durability.
To me, “ace” means a pitcher whose name on the lineup card spells instant doom for his opponents. As archaic as it may sound, I believe that the statistical measure of an ace can be derived from ([team W/L when pitcher x is the starter]/[overall team W/L]). WAR is nice, but it really doesn’t tell you how hard it is to beat these pitchers, which is the essence of being a staff ace.
Let’s take the Cardinals as an example. Wainwright is the most valuable starter on the team. He’s very, very good, and very durable. But on a per-start basis, Carpenter is the better option. He’s a control pitcher who throws 95, has excellent off-speed stuff and is completely unflappable on the mound. Wainwright’s projected WAR for 2010 is higher than Carpenter’s because of the difference in durability, but if you ask the Cardinals’ players, management, fans, and opponents who the ace is in the Cards’ starting rotation, more often than not the answer will be Carpenter. This is supported by the team’s winning percentage when either one of them is on the mound.
But let’s look at WPA instead (since this is a sabermetric site): Since WPA takes into account the win probability change of everything a player does, this is a great measure of a player’s overall performance in the context of winning baseball games. In the 2009 season, Chris Carpenter had a WPA of 5.41. Wainwright had a WPA of 3.60. Both are fan-friggin-tastic numbers, but Carpenter has the edge. Let’s scale it to a per start basis and add the other Cy Young contenders from both leagues from last year.
Carpenter: 5.41 WPA/28 starts = .193 WPA/start
Wainwright: 3.60 WPA/34 starts = .106 WPA/start
Lincecum: 4.26 WPA/32 starts = .133 WPA/start
Greinke: 6.07 WPA/33 starts = .184 WPA/start
Hernandez: 3.26 WPA/34 starts = .096 WPA/start
So… holy crap. Carpenter added on average almost two tenths of a win every time he took the mound last year. That’s what I call an ace.
IP needs to be a component for ‘ace’ status as well. I’d say your ace needs to be averaging 200 IP or better year in and year out to be considered an ace. And I agree that there are not 30 aces in baseball. Most teams don’t have one.
This is another good point. Big Z pitched 200+ innings for FIVE straight seasons between 2003 and 2007. Many Cubs fans conveniently ignore this fact when they bitch and moan about the size of his contract. I believe IP was a huge factor in his deal. I’m not saying Carlos IS an ace – I don’t think he is (under my own definition) – but he’s been a remarkably productive pitcher (career-wise) for the Cubs.
I respectfully disagree. IP is very important to valuing a pitcher, but the “ace” moniker is not about durability. It’s about domination on a per-start basis. Being injury prone makes you less valuable over the course of a season, but that has no effect on your ability to make a lineup of professional hitters look stupid every time you take the mound.
Be a little careful about writing articles like these after a pitcher gets bombed – seriously.
It is way too easy to predict that a pitcher will see some serious regression after giving up 8 runs in 1.1 IP. His ERA (and other stats of course) already took a .4 run hit for the season (say, 8 runs in 180 IP). Anyone can already predict regression for Zambrano!
While I am sure you would have written the same thing before the season, it is NOT the same thing. It is similar to publishing bias (where a researcher only published research when it comes out a certain way such that we end up with bias in all published results). Had Z. thrown a non-no, you would likely not have written this article.
A valid point. And can I also say how annoying it is when an author writes a post like this and then doesn’t stick around to answer questions and participate in the discussion (beyond the single 18-word comment above)? There, I said it.
Zambrano had ace like numbers from 2004-2006. But an ace is a guy who stops losing streaks, gives you at least 7 innings every outing while keeping his team in the game, and won’t blow up if something doesn’t go his way.
Also an ace can throw a complete game when it’s needed regardless of the chance of a win or not. Big Z has 3 complete games in his last 4 seasons. But I’m sure that he will hit 4 or 5 homeruns, and probably the Cubs third best hitter.
“And can I also say how annoying it is when an author writes a post like this and then doesn’t stick around to answer questions and participate in the discussion (beyond the single 18-word comment above)?”
What? Stick around? What do you mean? Is this a live chat session? There is no more discussion! There is only one more comment after mine (other than yours). And do you know how many blogs and posts and comments I make per day? If I “stuck around” for all of them, I would not be able to sleep or eat! Seriously, man, what are you talking about?
I believe dat cubfan fever meant that Jack hasn’t done much responding to this post.
The problem with this whole thing is the arbitrary definition of what is an Ace.
Another way to look at an ace is any person who could potentially serve as an ace for a team (30 teams, 1 potentially for every team.) So the 30 best pitchers in baseball. Or you could say it is the best pitcher on every staff (so every team actually has an ace). Or you could say only pitchers who have FIPs lower than 3.00 or 3.25 or 3.452321202020.
There is no definitive brightline.
Has Zambrano’s performance declined? No doubt.
Is he an Ace? Depends on the definition of Ace.
Does someone being an Ace useful to discuss? Probably not.
While defining an Ace might be difficult, it is quite easy to work in reverse and determine who easily wouldnt qualify.
No matter the persons interpretation, one things stays the same – An Ace would have to be eating innings and consistently keeping the opponents success/run-output low. So, taking even two quick stats
Quality Start %
2009 – 61%
2008 – 57%
2007 – 53%
2006 – 67%
2005 – 70%
Ok, the ML average here is generally about 50% while the league leaders post mid-70s to 80+ marks. Specifically, the 61% Zambrano posted last season ranked 39th, and its even the best mark he has posted the last three years.
IP/GS
6.0 – 2009
6.3 – 2008
6.4 – 2007
6.5 – 2006
6.8 – 2005
Average here is about 5.8 while leaders are in the 6.7 to 7+ range. Zambrano has seen a steadily decline, with lasts year mark good enough for only 61st of 117 Qualifying Starters.
Instantly we can see that Zambrano isnt going deep into games or consistently keeping his team in them. And because of that, how can we really even begin to put him into a conversation about Aces, no matter what a persons other specific qualifications are?