We can count on a few stories to pop up multiple times every spring. Some players show up in the best shapes of their lives. While that’s probably the most common spring cliche, pitchers developing new pitches over the off-season ranks pretty close. The attached assumption is that another pitch means another weapon. Many pitchers, however, never implement this new pitch. They can work on it all off-season, but until they start throwing it in games they won’t really know how it works. And since throwing it in games can cost runs, some pitchers shy away.

We won’t know until April whether A.J. Burnett will use his changeup more in 2010, but he certainly worked on it this off-season. In fact, as Carig tells us, he worked on it harder than in any previous off-season. Adding a dependable changeup to his arsenal could keep hitters even more off-balance when his curveball is working well, and could provide a backup plan on days where his curveball falls flat. But even given his hard works and the effect it could have on his success, Burnett won’t commit to mixing it in more often. “Whether I throw it or not, I don’t know,” he said.

To help him better develop the changeup, Burnett sought out 40-year-old Reds reliever Arthur Rhodes. It sounds perfectly normal for pitchers to seek advice from their elders, especially when they live nearby in the off-season. Rhodes has been a fine pitcher over his 18-year career, striking out nearly a batter an inning while maintaining a 4.15 ERA (107 ERA+). Without the 82 innings he’s pitched against the Yankees, though, that would be a 3.88 ERA. He has also allowed eight runs in 6 playoff innings against the Yankees, though seven of those runs came during two games in the 2000 ALCS.

I’d love to see Burnett work in a third pitch, but it’s not a simple process. I assume he’ll throw it a bit more often once the spring games start to see if he can get a feel for it. But even then he might not have enough confidence, no matter how much progress he’s making. Without full confidence in a pitch we can’t expect a pitcher to use it, no matter how much it could potentially improve his arsenal. With Burnett, I’ll expect another season of fastball-curveball, and take any further changeup usage as a bonus.

Photo credit: Kathy Willens/AP

While A.J. Burnett’s first season in pinstripes was in many ways a success, he left some room for improvement. He knows it, too. As he said the other day, he needs to, “Not walk as many people and go deeper into games.” In 2009 Burnett posted his highest walk rate since 2000 and the lowest innings per start of his career (both discounting his injury shortened 2003). He would clearly benefit from improvements in both, though I do think that he left out one important aspect: his ground ball rate.

Photo Credit: Kathy Willens/AP

Burnett has never been known as a control pitcher. In some years, such as 2006 and 2004, he’s posted BB/9 rates below 3.00, which is a good mark for a starter. Yet his career rate is 3.78, meaning he’s had many years above that mark. He also sat below the mark in years where he failed to reach 140 IP. Chances are, Burnett won’t miraculously stop issuing free passes in 2010. Instead, I imagine he’ll fall back somewhere in the range of his last five years, which is 3.52 walks per nine.

What Burnett could do to help his case is to start inducing more ground balls. Or, rather, to induce ground balls like he did before 2009. Last season he posted the lowest ground ball rate of his career, 42.8 percent. Over his career he’s kept nearly 50 percent of balls in play on the ground, which helps a pitcher who hands out free passes. Unsurprisingly, Burnett induced more double plays in 2005, when his ground ball rate was 58.4 percent, than at any other point in his career. In 2009 he induced one double play for, roughly, every 60 batters faced. Prior to 2008, when he saw his ground ball rate drop below 50 percent for the first time his career, he was around one every 38 batters. In 2008 he was at one in every 50 batters.

Keeping the ball on the ground can also help Burnett reach his second stated goal, to go deeper into ballgames. While walking fewer hitters will undoubtedly help, so will inducing ground balls. In his best ground ball years, Burnett kept his hits per nine rate below 8.0. With a decent infield behind him, Burnett shouldn’t have problems with too many ground balls finding holes. Perhaps if this was the 2005 Yankees defense it would be an issue, but it’s not as much in 2010.

At FanGraphs, Matthew Carruth examines ground ball rates and what they mean for pitchers. In terms of the big picture, pitchers with higher ground ball rates saw lower FIP and runs allowed rates. That’s not to say that every pitcher follows this guideline, but Burnett seems to. In the last five years his highest ground ball rates have come in 2005 and 2007, which also happen to be the years he’s posted his lowest ERAs.

We’d all like Burnett to cut down on his walks and pitch deeper into games. The first, in fact, begets the second. Part of the problem with walks, however, is that Burnett has never been a low walks guy. But he has been a ground ball guy. If he can get back to that, and bring his walk rate back to career norms, he should not only pitch deeper into ballgames, but also pitch more effectively overall.

In a general sense, A.J. Burnett performed well in his debut season for the Yankees. Sure, he tossed a few clunkers, but he also had his share of dominating performances. At the end of the season that averaged out to a 4.04 ERA, right around his 2008 ERA of 4.07. That he pitched over 200 innings for the second straight season, the first time he’s accomplished the feat in his career, was a further positive. In a number of ways, however, Burnett’s season represented a step backward.

In 2007, after an injury-shortened debut season with Toronto, Burnett accomplished something he never did in the National League: he struck out more than a batter an inning. He did it again in 2008, and that probably played a role in the Yankees’ decision to sign him. It was an odd trend, of course, as pitchers tend to rack up more strikeouts in the NL, where pitchers hit instead of the DH. During these two years Burnett also kept his walks in check, around 3.50 per nine. Yet in his 2009 campaign Burnett declined in both areas.

His strikeouts remained high, 8.48 per nine innings, right around his career National League performance, but not quite at the level of his previous AL East experience. His walks also shot up, 4.22 per nine, his highest rate since 2001 (discounting his 23 innings in 2003). But again, despite declining about 10 percent in strikeout rate and increasing about 27 percent in walk rate, Burnett produced similar results as 2008. My question is whether this is a good sign, or whether it represents a boatload of good luck.

Looking a bit deeper into Burnett’s performances makes me think that luck played a big factor in his 2009 numbers. The first set of data that stands out is his ground ball to fly ball ratio, 1.09, the lowest of his career, and by a decent margin since 2003. His ground ball rate has declined over the past three years, going from 54.8 percent in 2007 to 48.5 percent in 2008, and finally to 42.8 percent in 2009. In that time, his fly ball rate has increased from 29.8 percent to 32 percent to 39.2 percent. This led to an xFIP of 4.29 and a FIP of 4.33, both a bit above Burnett’s actual ERA. His defense, it would seem, helped him out a bit.

We know that Burnett lives on his curveball, a nasty pitch that acts somewhat like a slider, diving down and away from right-handed batters, though the down and inward motion seems to foil lefties as well. He lives on swings and misses out of the zone in that regard. Yet in 2009 hitters made more contact in pitches outside the zone, 51.1 percent, than in any other year of his career. Burnett’s contact rate as a whole jumped last season, while his number of pitches thrown inside the zone was the lowest of his career (again, discounting 2003). What’s worse, hitters swung at fewer pitches outside the zone, 22.1 percent, than they had since he moved to the AL in 2006. His overall swing rate was, again, the lowest of his career.

The curveball, however, seems to be fine. According to FanGraphs’s pitch type values, his curve was as good as ever, perhaps among the best it has been in his career. It was worth 15.4 runs above average, higher than in any of the Blue Jay years, and higher than any year of his career except 2005. What hurt him, it appears, was his fastball, which ranked -13.0, the lowest of his career and, on a per 100 pitch basis, the 22nd worst fastball in the majors among pitchers who threw more than 150 innings.

Does the fastball decline explain Burnett’s increased walks, decreased strikeouts, and decreased ground balls? I think it has to, at least in some way. Again, look at Burnett’s last three seasons, and you’ll see increasing fly balls and decreasing ground balls. You’ll also see a decrease in his fastball value, from 8.2 runs above average in 2007 to 5.9 runs below average in 2008 to 13 runs below average in 2009. While we can’t determine specific causation, there seems to be something of a correlation there.

What’s most troubling about A.J.’s trending numbers is that we should have expected an uptick in performance over Toronto. He pitched in the AL East for three years and had to face the powerhouse Yankees offense during that time. By moving to the Yankees, he moved from that to having to face the Blue Jays lineup. So it appears that his workload got a bit easier. Yet his peripherals declined. I don’t like the looks of that.

Perhaps Burnett went through a period of adjustment to the rigors of pitching for the Yankees, and will recover his previous form in 2010. We know he has the stuff to do so. We saw A.J. at his best in 2009, one-hitting both the Mets and the Red Sox. We also saw him at his worst, giving up eight and nine runs to the Red Sox, seven to the White Sox. Hence Good A.J. and Bad A.J. These numbers don’t show whether we’ll see more of Good A.J. in 2010, though they do show why it appeared Bad A.J. showed up more than he really did.

Credit: AP Photo/Elise Amendola

Tonight, in the first elimination game of the 2009 World Series, the Yankees will send A.J. Burnett to the mound on just three day’s rest. The tactic makes sense. The alternative is Chad Gaudin, who hasn’t started a game since late September and who has a well-documented deficiency when facing left-handed hitters. With such a significant drop-off between the Yankees third best starter, Andy Pettitte, and their fourth, Gaudin, the choice was not a difficult one.

Burnett has experience starting on three days’ rest, and most of it came in the 2008 season with Toronto. His performance in those games might have helped influence Joe Girardi’s decision, so let’s take a look at exactly what happened when Burnett took the mound a day earlier than normally scheduled.

July 4, 2004

Burnett underwent Tommy John surgery in early 2003 and made his return in June, 2004. He had a few blips, including a 4.1-inning, eight-run outing against Cleveland, but generally pitched well in his first month back. Unfortunately, the Marlins could not pick him up, losing each of Burnett’s six starts that month. This included back to back starts in which Burnett allowed just two runs over seven and eight innings.

On July 4, the Marlins called on Burnett to start on three days’ rest against the Tampa Bay Rays, against whom he had thrown the aforementioned eight-inning game. He didn’t pitch quite as well, allowing three runs over 7.2 innings, but it was enough to earn his first win of the season. He struck out six Devil Rays that day, including Carl Crawford to lead off the game. Atypical of Burnett, he also didn’t issue any walks — though Tampa Bay drew the fifth fewest walks of any MLB team that season.

It might seem strange for Burnett to start on three days’ rest so shortly after recovering from elbow surgery. That seems like the kind of move that could lead to a relapse. Burnett, however, had thrown just 30 pitches on June 30, leaving the game two batters into the second inning after allowing five hits and walking two. That light workload made the short-rest start make a bit more sense.

July 13, 2008

Burnett started off July 2008 with two horrible starts. In seven innings against the Angels on Independence Day, Burnett allowed eight runs, six earned, on 12 hits over seven innings. The next time out he allowed seven runs on seven hits and three walks over 5.1 innings to the Orioles. Why, then, would the Blue Jays bring back Burnet on there days’ rest to face the Yankees on July 13?

I’m not quite sure. It was the last game before the All-Star Break, so perhaps Cito Gaston didn’t want Burnett to have such a long layoff. Whatever the reason, it worked. Burnett took a shutout into the ninth inning, though Jason Giambi ruined it with a solo home run. B.J. Ryan came on after a Jorge Posada single to record the final two outs. Still, Burnett was magnificent, and it’s one of the reasons that the players lobbied the team to sign him over the off-season.

September 13, 2008

At the end of August, Burnett found him with quite the challenge. On the ledger for his final three starts of the month: the Yankees twice, with Boston in the middle. While the Blue Jays were out of the race, it was still an audition for both teams. He killed the Yankees, but faltered a bit against the Red Sox. He’d get his chance for redemption against them, though, as Gaston named him the starter on September, just three days after his seven-inning, one-run performance against the White Sox.

It wasn’t an easy six innings for Burnett, as he used 102 pitches, walking three. But at the end of the sixth he had allowed just one unearned run. The Blue Jays went to town, scoring eight runs in support of their free-agent-to-be, helping him pick up his 18th win of the season.

The phantom three days’ rest start — April 16, 2008

When looking up Burnett’s short-rest starts, I first went to his Baseball Reference splits page, where it says he has started four games on three days’ rest. Yet I found only three such games. It comes down to a nitpick: does a start count as being on three days’ rest when the previous appearance was in relief?

On April 16, 2008, then-Blue Jays manager John Gibbons called on Burnett to come into the 14th inning of a game at home against the Rangers. It was tied 5-5, and the Blue Jays needed some more innings out of a dwindling pen. He had last pitched on April 13, also against Texas, and didn’t pitch particularly well in that start, allowing four runs over 5.2 innings. It seemed curious that Gibbons would call on Burnett two days later, but he did and paid for it. Burnett allowed two runs on three hits and a walk in the 14th, leading to a 7-5 Blue Jays loss.

Then, three days later, Burnett came out to start against the Tigers. It was six days after his last start, but just three days after his last appearance. He allowed three runs over five innings, walking six in the game. It was easily his worst start on three days’ rest, yet the Blue Jays offense put him in line for the win, his second of the season (the first was against the Yankees in his first start of the season).

In a way, I don’t want to count it because the start on three days’ rest did not follow another start, but a relief appearance. Then again, Burnett did throw 24 pitches in that span, six short of the 30 he threw on June 30, 2004, which he followed with a start on three days’ rest. Is there much of a difference there? I thought so at first, but I’m not so sure after thinking it over some more.

None of this guarantees Burnett anything tonight. It proves that he’s physically capable of throwing on three days’ rest and succeeding, but that’s about it. Knowing his track record is a bit reassuring, at least.