24 October 2013

Making the Orioles a Champion in 2014: Left Handed Relievers

This post is part of the Making the Orioles a Champion in 2014 Series.  Below you will find links to the other articles.  We will do our best to make sure the links go live with each new update.
C | 1B | 2B (1, 2) | 3B | SS | LFCF | RF | DH | Bench | SP (1, 2) | RHRP | LHRP | Conclusion

Looking Back

Courtesy of prosportblogging.com
For the portsiders in the Baltimore Orioles bullpen, the 2013 season was one that saw the team call upon their services frequently; in response to doing so, they provided a contrast of results. The trio of Brian Matusz, T.J. McFarland, and Troy Patton dutifully fulfilled a number of roles, including a one-off start by McFarland, but in general, came into games in one inning spurts, with Matusz essentially taking over the more traditional LOOGY role more so than the other two leftys. While one of the more unheralded lefty bullpens in the AL East, Baltimore's trio had a lot of say when discussing the overall production (in the form of wins above replacement) of the Oriole relief pitching corps in spite of their overall low fWAR, as compared to the rest of the division:

Team LHRP fWAR % Total RP fWAR
TOR 2.1 38
TBR 2 44
BOS 1.6 26
BAL 1.3 45
NYY 0.1 3
*based on relief pitchers with >20IP

Let's now see how this 1.3 fWAR breaks down by individual and see in what ways each of the Baltimore Three contributed:

Overall numbers:
Brian Matusz 65 51 3.53 2.91 3.59 7.93 3.57 1.0
T.J. McFarland 37 72 4.00 3.94 3.75 -0.96 4.86 0.3
Troy Patton 56 56 3.7 4.42 4.1 1.69 5.06 0.0

In this broad view, we find that Matusz had an above average year for a lefty specialist; in particular, his Fair Run Average (FRA) and Value Added (RE24) reflect positively on Matusz' ability to get out of jams and also to the out sequencing of his innings pitched. Comparing the trio's FIP and xFIP numbers, we also find that Matusz outperformed his peripheral stats (FIP < xFIP), while both McFarland and Patton underperformed. Comparing FIP to ERA, it also appears that Patton was the beneficiary of some significant defensive plays behind him (ERA < FIP), while the converse is true for Matusz.

The nature of the role of relievers necessitates the inclusion of platoon splits into the discussion; let's see how each of the Baltimore Three fared, taking lefty/righty splits into consideration:

Lefty splits:
Brian Matusz 30 0.163 0.225 0.217
T.J. McFarland 33 0.282 0.330 0.351
Troy Patton 25 0.277 0.346 0.299

Not surprisingly, each of the lefties in the 'pen did a reasonable job against left handed hitters, but with Patton showing a propensity to give up a big hit more so than Matusz or McFarland; the mustachioed McFarland apparently fell victim to some bad luck, in the form of a slightly elevated BABIP against lefties, compared to his overall .319 BABIP. Once again, we see the talent level of former 2009 first round pick Matusz shining through, with above average numbers across the board.

Whether it was to get a tough lefty hitter out, or to bridge the gap between an abbreviated outing from a starter to closer Jim Johnson, each lefty did fulfilled their roles in differing ways, this notion being evident with the following table of batted ball results:

Brian Matusz 67.6% 18.2% 31.8% 50.0% 6.1%
T.J. McFarland 85.2% 18.9% 57.9% 23.2% 18.2%
Troy Patton 96.3% 17.7% 41.8% 40.5% 15.6%

McFarland's success with the two seam fastball is confirmed in this table, with him leading the pack in terms of ground ball rate (GB%), at nearly 60%. While Matusz is the more extreme fly ball pitcher (FB%) compared to Patton, his ability to keep those fly balls in the park (HR/FB) is in stark contrast to Patton, whose aforementioned elevated wOBA could be well explained by his propensity to give up the gopher ball, with nearly one out of five fly balls becoming homers against lefties. All that being said, Patton did display a knack for keeping runners stranded with lefties at the plate - stranding all but a couple in his outings.

Overall, how did their methods fare against left handers?

Brian Matusz 2.70 2.51 3.15
T.J. McFarland 3.00 3.08 2.41
Troy Patton 2.88 4.85 4.00

In general, we see from the three a knack for neutralizing lefty hitters, in particular, Matusz. However, we do see that Patton in terms of raw ability trails both Matusz and McFarland in getting lefties out consistently, having to take advantage of the stellar Oriole defense more so than his lefty bullpen mates.

Courtesy of baltimoresportsreport.com

Let's now see how they fared against righties:

Righty splits:
Brian Matusz 21 0.302 0.330 0.368
T.J. McFarland 41.2 0.267 0.315 0.299
Troy Patton 31 0.252 0.318 0.292

While all three relievers in question were former starters, Matusz does begin to show where his weaknesses with respect to starting lie in the above table -- getting right handed hitters out can problematic, and he does show in 2013 a bit of an unlucky streak, in the form of an inflated BABIP. All three tend to have issues with big innings when righties are faced, with each having a .300+ wOBA.

Brian Matusz 72.30% 23.2% 46.4% 30.4% 4.8%
T.J. McFarland 66.40% 16.9% 57.7% 25.4% 8.3%
Troy Patton 66.10% 17.6% 48.4% 34.1% 9.7%

With respect to batted ball results, all do a good job of keeping the ball in the park against righthanders, with each inducing a lot of ground balls. Matusz, in the form of a 23.2% line drive rate (LD%), appears to have a propensity to give up hard contact to righties more so than lefties, and more so than the other two in the trio.

Brian Matusz 4.71 3.48 4.22
T.J. McFarland 5.18 4.44 4.68
Troy Patton 4.35 4.08 4.19

In the above table, we see why each of our trio are in the bullpen; inconsistent and ineffective production against right handers, leading to bloated FIP and earned run results. Even with all three outperforming their peripherals in some modicum, it was not enough to warrant much confidence in their ability to consistently get righties out.

Moving Forward

As previously mentioned, the productivity from the left side of the bullpen last season was a mixed bag and was tied into the success of Matusz as a LOOGY; while not a severely disappointing portion of the bullpen, the lefty situation could stand to use an upgrade in 2014, either in the form of improved performances by the trio discussed, or by addition of other lefties. With regards to improvement by the three, we have projection systems such as Steamer to give us a hand with gazing into our 2014 crystal balls and giving us an idea of whether the trio will build or lose ground on their past seasons:

Steamer 2014 Projections:
Brian Matusz 45 8.03 2.96 1.18 0.292 3.54 4.04 0.2
T.J. McFarland 35 6.15 3.04 0.79 0.301 3.78 3.93 0.2
Troy Patton 55 7.18 2.83 1.00 0.295 3.58 3.96 0.3

By the looks of it, next year isn't terribly encouraging, especially when it comes to Matusz. While we see the potential for a bounce back season for Patton and for McFarland to continue to be a steady, no frills type of bullpen arm, we also see Steamer is not terribly convinced another 1.0 fWAR year is in the cards for Matusz. In particular, it expects his home run rates to inflate, which, considering how low his HR/FB ratio was in 2013 given his flyball heavy batted ball rates, isn't terribly surprising. Another interesting trend seen with the Steamer projection is McFarland's innings being cut in half, going from 72 in 2013 to a projected 35 in 2014. While some of this could be a sign of him being used more as a one inning guy and not being stretched out via multiple inning outings, it could also portend the arrival of another lefty, making our trio a quartet, leading us to discuss...

Other Options

Looking internally, the Orioles do have a promising lefty arm in Mike Belfiore, who saw very limited work as a September call up in 2013, but spent essentially all of last year in Class AAA Norfolk. His AAA numbers were respectable, tallying a nearly 3-to-1 strikeout to walk ratio, and a so-so 1.44 WHIP; as previously mentioned here at Camden Depot, he is a talented prospect whose lefty splits bode well to him taking over some of the innings allotted to our 2013 trio. Norfolk teammates Daniel Schlereth, Chris Jones, and Andy Loomis could also provide innings in 2014 in a pinch, with Schlereth previously logging major league innings with the Arizona Diamondbacks and Detroit Tigers.

Outside the organization, the pickings for free agent lefty relievers are slim and old. While some familiar names with decent pedigrees are available, 11 of the 12 available lefty relievers are over the age of 31, with 29 year old Atlanta Braves hurler Eric O'Flaherty being the youngest of the lot. O'Flaherty doesn't come without his own baggage, as his 2013 was cut short due to a torn ulnar collateral ligament in his pitching elbow, prompting Tommy John surgery. However, an incentive laden minor league deal for O'Flaherty could be conducive to the Orioles budget and bullpen prospects, assuming that he returns to his usual self, which saw him average 69 appearances and a 1.17 WHIP from 2009 and 2012. Other potential arms that could help defray some of the workload that could be smart bets include old Oriole friends Oliver Perez and in particular, Mike Gonzalez. Gonzalez, looking to bounce back from a disappointing season with the Milwaukee Brewers, has previous closing experience and is also just as devastating on righties (.229 career batting average against) as he is on lefties (.219 career batting average against), something that could play well in the Orioles bullpen, potentially providing the occasional night off for the likes of Johnson and Tommy Hunter when it comes to closing duties, but also allowing manager Buck Showalter to be less susceptible to the effects of the left-right splits of his current lefty relievers on his late inning game management.

What to Do in 2014

Looking to 2014, one of the areas in greatest need of upgrading is the left handed side of the bullpen; however, the upgrades are small in size and stature and are conducive to being taken care of with a couple of low key roster moves. While Matusz, McFarland, and Patton provide a respectable lefty trio, they come with drawbacks that potentially predicate the addition of another lefty to their corps, in particular, one that fares better against right handed hitters. Whether it is the introduction of Belfiore to a larger role with the team or the pursuit of a low risk, medium reward free agent such as O'Flaherty or Gonzalez, the Orioles can do much towards setting themselves up for a 2014 to remember by adding another lefty to the bullpen.

All data courtesy of FanGraphs.

Courtesy of wikipedia.org


Matt P said...

"Another interesting trend seen with the Steamer projection is McFarland's innings being cut in half, going from 72 in 2013 to a projected 35 in 2014."

My understanding is that Steamer doesn't try to project innings very accurately especially in its early projections. For example, if you look at the Tigers starters, it projects them to use a two man rotation of Verlander and Scherzer, each of whom will pitch 480 innings. Call me crazy but I'm skeptical. Likewise, they project 30 closers to each get 28 saves.

How many teams have ever had four lefties in their bullpen at a given time? My understanding is that three was too many.

Jon Shepherd said...

With pitchers like Matusz and Patton there was some thought that their starting repertoire would enable them to pitch both sides so them being lefties was not considered as much of a concern.

Stuart Wallace said...

Matt -

Agreed that Steamer isn't gospel and like any other projection system, has its weaknesses. That being said, I though it interesting that of the three, McFarland's IP's went down so egregiously, while Matusz and Patton stayed pat, for the most part. It also added some intrigue to my idea that a fourth lefty - one that could better handle righties - might be on the horizon. But I agree with your skepticism.

As far as the fourth lefty, both Boston and Toronto used four in 2013, with each getting over 20IP; it isn't a commonly used tactic, but if that fourth lefty didn't have such drastic splits, it wouldn't be out of the question.

Unknown said...

What's more relevant than the total number of left-handed pitchers a team used is the number of left-handed relievers a team has on the roster at a given instant. It does appear that the Red Sox DID have four lefty relief pitchers on the roster (before September roster expansion) - but only for a week. I don't know if it would work to have four left-handed relief pitchers for a full season.