Millwood vs Burnett
Who is the best pitcher available this winter? For most of the spring and summer, everyone agreed it was A.J. Burnett. However, as the season went on, Kevin Millwood‘s ERA just kept going lower and lower. By the end of the season, he bested Johan Santana for the AL ERA title, and people began to realize how well he had pitched. Combine that with Burnett’s ERA of near 6.00 in September and his publicized ouster from the Marlins, and, well, there’s no longer a consensus on which pitcher is going to be more highly sought after this winter.
Even within the commenters here, there’s no real agreement. Millwood or Burnett, Burnett or Millwood. Back and forth we go. Well, let’s get to the bottom of this. Which one is better?
Control
Controlling the strike zone has never been Burnett’s strength. He walked 9 percent of the batters he faced this year and 8 percent in 2004. His career mark is 10.5 percent.
Millwood is, comparatively, a strike throwing machine. He walked just 6.5 percent of the batters he faced this year. He struggled with his control in 2004, matching Burnett’s 8 percent mark, but that appears to have been an anomoly. His career mark is just over 7 percent.
Burnett can be expected to walk about 20-30 more batters than Millwood given the same number of batters faced over the course of a season. That’s a very significant number. Big edge to Millwood.
Dominance
Burnett’s a classic power pitcher, so this is where he makes up for his occassional wildness. He struck out 8.7 batters per game, fifth best in the National League, totaling 23 percent of the batters he faced. In other words, he racks up a strikeout every four batters. That’s dominance.
Millwood is no slouch here either, though. His 7.0 strikeouts per game looks vastly inferior, but he didn’t get to face the opposing pitchers. His mark, 18 percent of the batters he faced, was good for 8th best in the American League. His 2004 season in Philadelphia, he struck out 20 percent of the batters he faced.
Burnett’s certainly the more dominant strikeout pitcher. He’s one of the elite strikeout arms in the game. But Millwood isn’t chopped liver here either. Edge to Burnett, but not an earthshattering one.
Contact Type
A.J. Burnett was a dominant groundball pitcher this year, posting a 2.42 G/F rate (by ESPN’s calculations). The Hardball Times has him as the fifth biggest groundballer in the National League. He was in the same class as guys like Tim Hudson and Mark Mulder, notorious groundball machines. However, this was a substantial leap for Burnett. He’d never been anything close to this before. In 2004, his G/F rate was 1.49, and his career mark is 1.42. Burnett saw a 40 percent increase over his previous career best in G/F rate. None of the other dominant groundball pitchers in baseball have ever had seasons where there G/F rates were as low as Burnett’s have historically been. If he can continue his groundball dominance going forward, he’ll be historically unique. At this point, there’s not anywhere near enough evidence to suggest that this was a legitimate improvement on Burnett’s part; the best guess on the information we have is that it was likely a career year, one that probably won’t be repeated.
Millwood also saw a significant bump in his G/F rate this season, though that one is far easier to explain. His career G/F rate is 1.01, and he’s been a flyball pitcher his whole career. His mark this year, however, is 1.39. According to a recent study done by Dave Studeman found, Jacobs Field is the most groundball prone park in baseball, inflating groundballs by around 11 percent. It’s not a big surprise that Millwood saw a big boost in his G/F rate in the one season he’s spent in the Jake.
Safeco Field, ironically, plays as one of the most flyball oriented parks in baseball, so any pitcher coming into Safeco should be expected to see his G/F rate decline, at least marginally. Projecting anything over a 1.5 G/F rate for Burnett in Safeco would be folly; Millwood probably projects right around the 1.0-1.2 mark he’d estabilshed as his career norm. Burnett’s certainly the better bet if you want groundballs, but he’s a very poor bet to retain his elite groundball status in future years.
Durability
Burnett has thrown 854 innings over 7 major league seasons. He had Tommy John surgery in 2003. He also experienced arm problems in the minors and had to miss a few starts this season with various health issues. His mechanics are mediocre, and he has a fairly violent delivery.
Millwood has thrown 1,600 innings over 9 major league seasons. He missed a significant part of the 2001 season in Atlanta with a shoulder injury, but never had surgery. He’s also had minor back and groin problems that have kept him out of starts the past several seasons, but he’s made 30+ starts in 3 of the past four seasons. His mechanics are clean and he has a smooth, easy delivery. As far as pitchers go, Millwood’s basically as little of an injury risk as you’re going to find. He’s an innings eater and has proven quite durable.
If you want to make sure you’re big ticket pitcher investment takes the hill frequently the next few years, Millwood is your guy.
Performance
A.J. Burnett posted a 3.45 ERA this year in one of the most pitcher friendly parks in the National League. His career record is 49-50. His ERA the past four years is almost a full run higher away from the friendly confines of Pro Player Park.
Kevin Millwood posted a 2.86 ERA in a neutral park in the American League. His career ERA is lower despite pitching in significantly less friendly environments. He has 107 wins against 75 losses.
No brainer, right? Millwod by a mile.
Not so fast. Burnett’s Fielding Independant ERA this season was 3.09 vs 3.77 for Millwood. Even after adjusting for league factors, Burnett’s FIP is significantly better. His mark was 27 percent better than the NL average; Millwood’s just 13 percent better than the average AL pitcher. The main difference in their amount of runs allowed this season was in their strand rate. Millwood allowed just 21 percent of the runners he put on to score, while Burnett allowed 32 percent of his runners to cross home plate. There’s little evidence to suggest this is a repeatable skill, and the gap will almost certainly close next year. If you even up the strand rates for both pitchers, Burnett’s overall numbers are actually better than Millwood’s this year.
Millwood’s 2005 performance was a lot more valuable to his team than Burnett’s was. But we don’t care about that. We’re looking forward, and Burnett’s 2005 performance projects better in the future than Millwood’s does. Millwood has the longer track record, but Burnett’s more recent track record is better. We’ll call this a push.
For the last few months, I was on the Millwood bandwagon. However, down the stretch, he got a significant amount of help from his defense, and his late season push to the ERA crown eliminated whatever chance he had of being undervalued. There’s no way Kevin Millwood is a value buy as a free agent anymore. If you want him, you’re paying top dollar for him, because the market for him is going to be very strong.
The market for Burnett is still strong, though he’s not the undisputed number one arm on the market anymore. And that will hurt him. The teams looking for pitching aren’t going all out for Burnett. They have viable backup plans in place, lessening the risk of one team blowing everyone else out of the water. That’s why the 5 years, $65 million speculation that creeped up over the weekend isn’t like to materialize.
At this point, I see no clear advantage to pursuing either pitcher in lieu of the other. Neither one will be a bargain. They’re going to be the two most expensive free agents to sign this offseason, and their price tags will be similar.
What’s the best course of action for the M’s, then? Pursue both. The M’s did a nice job covering their bases last offseason in their pursuit of two hitters, pitting Sexson and Delgado against each other for one position and using the leverage to make sure they got one of the two. It’s time for round two. Don’t pick between them; let them fight for the M’s. Make similar offers to both pitchers and the first one to sign gets it.
Kevin Millwood or A.J. Burnett? What’s the right answer?
How about “yes please”?
Comments
59 Responses to “Millwood vs Burnett”
I’m trying to recall the class of free agent pitchers last year. Seems like other than Pedro Martinez, there’s no guy I’d rather have between that year and this year than A.J. Burnett.
And some of those guys cashed in huge contracts.
Great thought-provoking write-up.
Interesting questions about G/F ratios and if parks influence them. I find it hard to believe that they do, but if Dave says that data strongly supports that, I believe that. The question is why, or is the data picking up something else that hasn’t been found yet?
I have similar questions about strand rates not being a repeatable skill. Intuition tells me it should be — in fact relief pitchers are (in part) measured that way. Shouldn’t some pitchers be very good and others not so out of the stretch? Shouldn’t some be very worried about baserunners and others not so based on the abilities of their catcher? So why does the data not show it as a repeatable skill overall? Interesting.
The trick of course is to get beyond the data/facts and on to converting those facts into information. Sometimes that is harder than we might think, but makes for good, intelligent discussion.
I also am confused by the idea that stranding runners is NOT a repeatable skill. Isn’t that one reason good relief pitchers are good? It is probably hard to find “bearing down” or “not giving in” within the stats, but it IS a skill.
Mussina is a borderline HOFer. I don’t think he makes it, but he’s got a legit case.
Neither of these gentlemen are particularly close, health-wise, pitching-results-wise, or track record.
Hitting often depends on visibility. Could that be causing Safeco to play as more of a flyball park?
That was a really nice, balanced, intelligent analysis.
However, there are two points I sorta disagreed with:
First, I think that the health factor is not as much in favor as Millwood as you suggest. Burnett has only had one serious injury. All of the rest of his issues have been very minor. He is now over a year and a half removed from TJ surgery, and has regained his prior form. I think that both of these guys are pretty good bets as far as health. I do agree that Millwood has been more of a horse over the course of his career, but he has missed significant amounts of time in two seasons. I don’t think that this is as big of an issue as you suggest.
Second, I think that your analysis should have included upside. This is where I think that Burnett has a significant edge. As you pointed out, Burnett is dominant. He has incredible stuff, and scouts drool over him. If he makes a slight improvment in his command, he becomes an elite pitcher. This is not that uncommon among players coming back from TJ surgery, and players often get their velocity back first, but take two years to regain all of their command. Burnett improving his command is more likely to happen than Millwood suddenly increasing his K rates at age 31. With Burnett, you also have to consider age. He was 28 this year, and lost the better part of two years of development time to injury. Thus, I really think that there is more potential in him to be unlocked. The age issue also has a big effect if we are looking at 4-year deals, which is what both of these guys will almost certainly get. The age difference is only two years, but, with Burnett, you are getting him from age 29-32. With Milwood, the M’s would be getting him at age 31-34. This is not a huge difference, but it is something.
Potential is difficult to judge, and pretty much impossible to quantify. However, I think that, given the specifics of his situation and his scouting reports, that Burnett is the better bet to improve in the next four years. Since both of these guys are likely to be overpaid this offseason, it makes sense to me to consider this. If Burnett improves his command a little bit, he will make a Chris Carpenter-like jump from ‘really good’ to ‘elite’. If that happens, they he will be worth more than the M’s will have to pay to get him. With Millwood, I don’t see much reason to expect him to improve a lot. His numbers might be better from playing in Safeco and with a really good defense, but I don’t think that you will see a fundamental improvement in his game. Just my opinion, though.
I hope that the M’s pursue both, like you say. It will help keep the price within reason, and having options is never a bad thing.
I’ve been puzzling – aloud and mainly on this site – as to why Baseball Prospectus is so low in AJ Burnett compared to Millwood. They have Millwood in or close to the top ten pitchers listed by VORP, with Burnett about 40 places lower listed near decent but unexciting veterans like Paul Byrd and Kenny Rogers.
According to VORP Burnett’s a little worse than Josh Trowers and a little better than our own Jamie Moyer.
The thing that frightened me most was the incredibly low score of 3 for runs prevented above what an average pitcher would have allowed (listed as RP, as distinct from PRAA). Burnett is credited with being 3 whole runs better than average over 209 innings. With his very low HR rate, high strikeout rate, good K/BB rate, and increasing tendencies as a groundballer, just how in the hell can that be?
Why the no-love for AJ?
I agree with Jerry above. Burnett’s upside is enormous – Cy Young material, while Millwood will never be more than a solid #2-type. If we’re going to spend #1 money, let’s get a guy that has the potential to fill that slot.
Burnett’s our man. M’s in ’07…
Re: # 39 – Did you forget GREG HIBBARD?