Projecting knuckleballers like Steven Wright an almost impossible task

The career of a knuckler is maybe the one thing in baseball more difficult to predict than the flight of the pitch itself. More than any other pitcher, a knuckleballer isn't born, they're made, and usually after an attempt at a normal career path has failed. They tend to be older than your average prospect, and since there are so few of them, it also at least appears as if they have a high rate of failure.

The upside, though, is just so high. Knucklers who do make it tend to have long careers, and as National League Cy Young winner R.A. Dickey reminded the baseball universe just last year, the very best of them can go up against any pitcher on the planet. Combine that with Tim Wakefield's successful 17 years in Boston, and it's no wonder the Red Sox acquired knuckler Steven Wright during the 2012 season.

Wright came to Boston in exchange for Lars Anderson, and was protected from the Rule 5 draft this past November. Manager John Farrell is on the record as saying he likes Wright as potential starting depth as soon as this season -- this isn't surprising, as Farrell worked with Wakefield back when he was the Red Sox' pitching coach. The thing is, no one really knows how Wright is going to develop, how he'll be as a big-league pitcher, and so on. There's no real magic formula, from either scouting or the world of advanced statistics, for plotting out a knuckleballer's future. To understand why, though, you need to see a knuckler in action.

"I had a joke a few years back that Wakefield's knuckler graded a 'WTF' on the 20-80 scale," says Scouting Baseball's Kiley McDaniel. "I noticed Wakefield's would move left then right every time, but the third move could be in any direction or no move at all, which explains why players are always jumping out of the way. You're trying to respond to the third move, but sometimes you respond to the first or second."

Hitters know the pitch is coming, but unless it doesn't move, or never lands in the strike zone, they're never prepared for it. That's just with your standard knuckleball, though, the Tim Wakefield special. What about the faster Dickey model, the one Wright employs? Is there any way to know what is coming from them, more so than someone like Wakefield? McDaniel doesn't think so. "Dickey has kind of blown up the rubric, and you can see the discount the Blue Jays got him for in trade and contract. Everyone just throws their hands up and says they don't get it, and Dickey admits he doesn't know when it might disappear out of nowhere. It's just mysterious, and if there were more of them maybe we'd understand it better, but you need a self-selected group of guys that throw about 90 to even have a chance to try it...so of course they'll want to keep throwing 90."

The reigning Cy Young winner in the NL signed an extension for all of two years at $25 million, with a club option of $12 million (and a $1 million buyout) for 2016. To put that in perspective, Ryan Dempster, who is just two years younger than Dickey and has received all of four points for the Cy Young in his 15-year career, signed a two-year, $25.5 million deal with the Red Sox in the same off-season. Yes, Dickey had less leverage as he was under contract for 2013, but even with that, he just won the Cy Young, and no one was going to make him sign. Given the unpredictable nature of the knuckler, though, it's probably in his best interest to do exactly what he did, and take the money offered in the present.

Wright is one of those guys who can throw 90, but his fastball isn't great, as Mark Anderson of Baseball Prospectus details."His fastball can be a nice change of pace, particularly when he wants to change the elevation of the hitters eyes, possibly pushing his fastball up in the zone with velocity in the 88-90 mph range. He lacks life on the fastball, leaving it as an unconvential 'show-me' pitch." It's no wonder that he stuck with the knuckler rather than continued to throw 90, but it's not as if every pitcher who throws a straight heater is going to pick up on a big-league caliber knuckler, either, bringing us back to the problem posed by McDaniel.

That actually brings up another issue: how do you know if a knuckler -- meaning the pitch itself -- is good? One front office member isn't sure you can. "Scouting those guys is obviously very difficult. What's a 60 knuckleball? What's a 40? We just don't have a real good calibrated scale for that. What we can do is watch them pitch, and measure the effectiveness of it. Are guys missing it? Are guys getting weak contact? Are they uncomfortable up there? etc. Basically, is the pitch working. It's such a unique offering. I don't know how to project it."

He's not alone in his thoughts, as another front office executive agrees. "They are viewed in a different context because the stuff is so different. Judging by velocity or projection is pointless, and it's really difficult to make any kind of assumptions or generalities into these type of pitchers. If anything, scouting a knuckler at the professional level is going to be more about performance than with probably any other pitcher or type of player."

Then there is Jason Parks of Baseball Prospectus, who, for all of his expertise, just doesn't get knucklers. "Knuckleball pitchers confuse the [expletive] out of me. I'm not sure how to scout them in the present or project them for the future. They are a mystery to me." Parks is far from alone on this on the scouting side.

The thing is, even performance-based measures have their issues. The developer of the ZiPS projection system, Dan Szymborski, has had to tweak his methods in order to accommodate pitchers like Wright. "I did some experiments with knuckleballers and FIP*, and find the sample size for ERA being meaningful over FIP is about half. Essentially, at 3,400 batters faced, you only expect ERA to regress half towards FIP. The halfway point for the knucklers is about 1,800 batters faced. Also, I found that most knucklers simply age as if they were 4-5 years younger. 38 year-old knucklers, for example, have about as many innings remaining in their arms as 33 year-old norms."

*Fielding Independent Pitching accounts for peripherals and a pitcher's performance in order to separate it from the luck -- good or bad -- caused by the defense behind him. As with anything involving a knuckler, it is more complicated because of them.

Szymborski's findings put an even clearer stamp on the odd similarity to the off-season contracts of Dickey and Dempster, as Dickey, by this measure, should theoretically have more innings left in the tank than the 36-year-old Dempster. Uncertainty is the name of the game with knucklers, though, and it reflects in Dickey's pay.

Wright has yet to pitch in the major leagues, and has only been learning the pitch that defines him for a couple of years. It's not a surprise to see that, even with the built-in adjustments, ZiPS sees him putting up an ERA more than two runs worse than the league average in 2013. That kind of pessimism is unlikely to go anywhere in an objective projection system, in the same way scouts and front office types need to see performance from a knuckler before fully believing in them.

So, given that, it's hard to count on Wright for anything, other than knowing he should be around for an opportunity to prove he belongs. That's not something that will go away anytime soon, either: R.A. Dickey won a Cy Young, and had to sign a deal at a discount anyway; Tim Wakefield, for all his quality seasons, topped out at under $5 million per season in his career, and even signed a contract with a perpetual club option attached. That's just the nature of being a knuckler, but there's the other side of that, too. If Wright shows he has a big-league knuckler, one that can consistently put hitters away, then he'll continue to get chances to pitch and earn a major-league paycheck.

Until that happens (or doesn't happen), Wright and his ilk are one of the key mysteries of baseball. If more arms like his begin to appear, we might have a better understanding of knuckleballers from both scouting and statistical points of view, but for now, all we can do is watch and see what happens.