Relief pitchers are a volatile bunch. We’ve heard this justification countless times after a reliever follows a good season with a bad one, or vice versa. I’d wager a pretty penny that the so called volatility of relief pitchers compared to their starting brethren has less to do with a unique characteristic defining bullpen dwellers, and far more to do with the relatively small sample sizes they offer on a year to year basis to those attempting to predict future performance.

Earlier this week, the Toronto Blue Jays extended middle reliever Casey Janssen, signing him up to a two year deal that avoids arbitration in his last year of eligibility, and then takes care of his first year of free agency. His $5.9 million contract is largely the result of a very successful 2011 season in which he struck out a career high 23.8% of the batters that he faced, while maintaining a very decent 6.3% walk rate. His ERA, WHIP and FIP for the season were all the best of his career.

The previous season, Janssen did not find such success, striking out less, walking more and giving up more home runs than in 2011. The difference in results one year from the next should make us stop to consider whether or not Janssen’s 2011 season is a better representation of his true talent level than his 2010 season, and to that end, if he’s deserving of the two year $5.9 million contract that the team gave to him.

Such investigation becomes even more necessary when we discover that all of his career bests this past season were matched by a career low BABIP, a career high strand rate and the fourth lowest HR/FB ratio among relievers in the American League, also a career low. Typically, variance in such numbers is attributed to luck, randomness or at least, things outside the control of the pitcher. Which brings us to the following question: Was everything going Janssen’s way in 2011, or was he doing something different to the batters he faced to find an increased measure of success?

Digging deeper into Janssen’s numbers last year I noticed something exceedingly rare. The right handed reliever actually faced more left handed than right handed batters (119 vs. 104). This, despite having splits over the course of his career that definitely put him at a distinct disadvantage against lefties. For most of John Farrell’s first year as a manager, the Blue Jays didn’t have a left handed specialist in the bullpen. This was most painfully obvious when Octavio Dotel was left in to face a lefty.

However, for the small sample size loving Farrell, allowing Janssen to face batters on the other side of the plate was a valid option if we assume he was only interested in results. Left handed batters had a .539 OPS against Janssen, while righties managed a .659 OPS. This is a stark contrast to the season before when lefties ransacked Janssen’s arsenal of pitches for an OPS approaching .800. The average LHB got on base more than 38% of the time against Janssen in 2010, compared to less than 27% of the time in 2011.

Earlier, I mentioned Janssen’s BABIP being a career low, but the reliever actually had a higher BABIP against RHB than the previous season(.346 vs. .342). His overall low BABIP was almost entirely the result of his performance against LHB, where he went from having a .342 BABIP in 2010 to a .273 BABIP in 2011.

Despite the massive difference, Janssen’s approach to LHB doesn’t appear to have changed all that much from one year to the other. Thanks to BrooksBaseball.net, we can see his pitch counts and results from year to year.

These are the pitches and results he had against LHB in 2010:

Pitch Counts Foul/Swing Whiff/Swing GB/BIP LD/BIP FB/BIP PU/BIP GB/FB
Fourseam (FA) 118 35.48% 19.35% 64.29% 28.57% 7.14% 900.00%
Sinker (SI) 89 45.00% 5.00% 54.55% 45.45%
Cutter (FC) 159 37.04% 24.69% 61.29% 16.13% 22.58% 271.43%
Slider (SL) 49 44.83% 20.69% 80.00% 20.00%
Curveball (CU) 60 57.14% 9.52% 28.57% 28.57% 42.86% 66.67%
Changeup (CH) 42 26.32% 21.05% 40.00% 30.00% 30.00% 133.33%

And these are the pitches and results he had against LHB in 2011:

Pitch Counts Foul/Swing Whiff/Swing GB/BIP LD/BIP FB/BIP PU/BIP GB/FB
Fourseam (FA) 148 32.69% 17.31% 34.62% 26.92% 23.08% 15.38% 150.00%
Sinker (SI) 57 30.77% 23.08% 42.86% 14.29% 28.57% 14.29% 150.00%
Cutter (FC) 197 54.72% 10.38% 51.28% 20.51% 23.08% 5.13% 222.22%
Slider (SL) 3 50.00% 100.00%
Curveball (CU) 66 42.31% 23.08% 77.78% 11.11% 11.11%
Changeup (CH) 6 33.33% 100.0%

While throwing fewer sliders and change ups, Janssen is still using his four seam fastball, sinker and cutter the most. Even when we look at how he pitches with two strikes, the only year to year difference was a reduction in the number of sliders, which curiously was one of his more effective pitches against LHB in 2010.

Yet, even though Janssen was doing practically the same thing (his four seam fastball and cutter had slightly more horizontal movement last season compared to the year before), his similar approach resulted in fewer ground balls off left handed bats from season to season. In 2010, his GB:FB ratio was 2.67. In 2011, it dropped to 1.86. Again, giving up fewer ground balls and alternatively more fly balls and line drives, likely isn’t the best strategy for a pitcher in the AL East making half of his appearances at Rogers Centre. But somehow, it worked for Janssen. Unfortunately, it worked for him in a fashion that would lead one to believe that his not giving up a single home run to left handed batters last season had far more to do with randomness than skill, and is therefore likely unsustainable.

This is what makes the extension handed out by the Blue Jays to Janssen so questionable. Heading into arbitration, the two sides were only $400,000 apart. At the very most, Toronto would’ve owed Janssen $2.2 million in 2012. Offering to pay an additional $3.7 million for Janssen’s first year of free agency means that GM Alex Anthopoulos and the rest of the front office either undervalues him for this year or overvalues him for the year after.

As this off season draws to a close, here are the right handed relievers who signed a free agent contract for $3.7 million or more:

  • Jonathan Broxton: 1 year; $4 million
  • Francisco Cordero: 1 year; $4.5 million
  • Matt Capps: 1 year; $4.75 million
  • Ryan Madson: 1 year; $8.25 million
  • Frank Francisco: 2 years; $12 million
  • Joe Nathan: 2 years; $14.75 million
  • Heath Bell: 3 years; $27 million
  • Jonathan Papelbon: 4 years; $50 million

Every single one of those pitchers has closing experience. Every single one of those pitchers has a history that includes multiple seasons of success. Every single one of those pitchers has done things that Casey Janssen has not. A quick look at the numerous righty relievers who signed for less this winter than Janssen will make in 2013 reveals several talented pitchers who can offer exactly what Janssen does if not more.

I realize that this is nitpicking to a degree, that an extra $3.7 million to a pitcher with all of his service time in one city isn’t going to break any bank. It’s just that Blue Jays fans have heard all season about the organization’s payroll parameters. If the team is being promoted to its fan base as being one that can’t afford to eschew fiscal responsibility, then it should probably practice what it preaches throughout its entire roster, and not merely use it as an excuse to not go after the more expensive free agents.

The fact that there are so many indicators that 2011′s results didn’t match Janssen’s process, makes me wonder if this deal is something of an oversight, forced more out of the club’s policy of not negotiating single year contracts after arbitration figures have been exchanged than a firm belief that one year’s performance from a middle reliever will carry through to the next season.