Seth Smith's Platoon Splits

The Oakland A's have spent most of the offseason selling off young, cost-controlled pitchers like Trevor Cahill, Gio Gonzalez and Andrew Bailey for a cadre of top prospects whom they hope can make the club competitive in a brand new ball park in San Jose a few years down the road. But the A's sent a pair of arms made obsolete by the addition of Jarrod Parker, Brad Peacock, Tom Milone and others to Colorado yesterday, picking up a present-day upgrade for the outfield in Seth Smith for Josh Outman and Guillermo Moscoso.

Well, Smith is an upgrade when the 73 percent of the pitching population that tosses from the right side is on the mound, anyway. The lefty batter has been platooned judiciously during his big league career, with slightly over 83 percent of his career plate appearances coming against righty pitching. There's good reason for that. While he smacks around right-handers (his batting line against them is about 25 percent better than the league average, even accounting for Coors), he has been helpless versus lefties:

	AVG	OBP	SLUG	K%	BB%	PA
Vs Left	.202	.267	.317	20.5%	7.8%	244
Vs Right	.288	.362	.515	16.7%	10.3%	1203

While some of Smith's extreme platoon split is likely small sample size noise, righty and lefty pitchers certainly do approach him differently. Right-handers pitch him cautiously, locating just 42 percent of their pitches within the strike zone against Smith (the average for righty pitchers against lefty hitters is about 46 percent). Most of what Smith gets from righties is tossed off the outer third of the plate:

Right-handed pitchers' location to Smith, 2008-2011

Lefties don't tiptoe around the zone as much, throwing 49 percent of their pitches in the zone versus Smith (right around the average for LHP against LHBs):

Left-handed pitchers' location to Smith, 2008-2011

Smith's power numbers might suggest that he hits many more fly balls against right-handers than lefties, but that's not the case. He actually has a higher fly ball rate versus southpaws (47 percent) than against righties (41 percent). It's just that his flies don't have the same oomph versus lefties: Smith's average fly ball distance when a righty is on the bump is 333 feet, compared to 315 feet against left-handers.

Smith figures to platoon with some combination of righty-hitting Colin Cowgill (picked up from the Diamondbacks in the Cahill swap) and Michael Taylor. Smith's pickup seems to signal a particular lack of confidence in Taylor. The former Phillie prospect, traded to the Blue Jays in the Roy Halladay deal and quickly sent to Oakland for Brett Wallace, once ranked as one of Baseball America's top 30 prospects but has failed to hit for power despite standing 6-foot-5, 255 pounds. The 26-year-old has slugged just .428 in over 1,000 career Triple-A plate appearances, while the average in the hitter-friendly Pacific Coast League has been between around .440 in recent seasons.

Smith can do the heavy lifting against righties, but the A's need either Cowgill or Taylor to produce against lefties. Asking Smith to play every day doesn't look like it would be in his or the team's best interests for 2012 and beyond.

Guillermo Moscoso: Ticking Time Bomb

Colorado added two more names to its cornucopia of starting pitching options on Monday, trading outfielder Seth Smith to the Oakland A's for lefty Josh Outman and right-hander Guillermo Moscoso. At first blush, getting an arm like Moscoso might seem like a coup for the Rockies. After all, the 28-year-old notched a nifty 3.38 ERA in 128 innings pitched as a rookie in 2011, easily besting the 4.21 average for American League starters. Maybe the former Tiger and Ranger was just a late bloomer, and now he's poised to be a cheap, quality rotation option for the foreseeable future.

Or not. Pitchers can do three main things to succeed over the long haul: miss lumber, limit walks and induce ground balls. Moscoso was decent when it came to issuing free passes, but he was abysmal at the other two. Check out where he ranked among starter pitchers in these three key categories that do a better job of predicting future ERA than past ERA:

Player	K%	BB%	GB%
Moscoso	13.9	7.2	26.7
Percentile Rank	29	54	1

Moscoso's walk rate was better than slightly half of starters last year, and that's actually, by far, his most impressive skill. His strikeout rate was worse than 71 percent of starters, and Baltimore's Brian Matusz (he of the highest ERA ever for a pitcher making at least ten starts) was the only guy to get grounders less often. When it comes to the metrics that extract luck and defense from the equation, Moscoso was a marginal major leaguer.

As you might expect from such an extreme fly ball pitcher, Moscoso pitched up in the zone with his 90-ish mph fastball, low-80s changeup and high-70s slider. Take a look at his pitch location in 2011, and then the league average for right-handers:

Moscoso's pitch location, 2011Average pitch location for right-handers, 2011Thirty-four percent of Moscoso's offerings were located in the upper third of the zone, well above the roughly 28 percent league average. Moscoso didn't get killed climbing the ladder last year in Oakland, surrendering about a homer per nine innings pitched. But that came with a very low home run per fly ball rate (six percent; the league average was close to 10%) while pitching in the Coliseum, which decreases homers hit by lefty hitters by 11 percent and 20 percent for righties. By contrast, Moscoso's new home and fly balls go together like peanut butter and tooth paste: Coors boosts long balls by 13 percent for left-handers and 17 percent for right-handers.

Moscoso's good fortune on fly balls hit, favorable ball park and the lowest batting average on balls in play (.222) among starters save for Jeremy Hellickson helped him post a shiny-looking ERA, but his more skill-based stats paint the picture of a pitcher with an ERA around five. He's basically a right-handed Greg Smith (another former Athletic shipped to Colorado after a superficially impressive rookie campaign) with better control. Don't be surprised if Moscoso, like Smith, turns out to be a Triple-A lifer instead of a solid big league starter.

Cahill and K's

The Arizona Diamondbacks fortified their rotation over the weekend, acquiring Trevor Cahill (and reliever Craig Breslow) from the Oakland A's for a trio of prospects: RHPs Jarrod Parker and Ryan Cook and OF Collin Cowgill.

Parker is nearly big league-ready and has big upside, having returned from Tommy John surgery to post a 112/55 K/BB ratio in 130.2 innings at Double-A Mobile and then impressing enough in a late-season cameo to earn a postseason roster spot.  Diamondbacks GM Kevin Towers says the club is in "go-for-it mode," taking advantage of a window to retain top status in the NL West. However, Towers is no doubt aware that Cahill -- signed through 2015 for a little under $29 million, with team options for 2016-2017 worth a combined $26.5 million -- is a long-term buy with remaining upside.

Cahill (24 in March) has been a slightly above league-average starter to this point in his career, sporting a modest strikeout rate (5.5 K/9), average control (3.3 BB/9) and strong ground ball tendencies (54%). But the right-hander, who whiffed nearly 10 batters per nine innings as a prospect, is starting to miss more bats in the big leagues. He struck out just 11.6% of hitters faced as a rookie in 2009, but he increased that total to 15.1% in 2010 and fanned 16.3% of batters faced in 2011 (the average for starters is 17-18%).

He's getting more Ks by incorporating his breaking pitches more often. Cahill threw his his-70s curveball and his short-breaking, low-80s slider just 10 percent of the time as a rookie. As a sophomore, he threw his breaking stuff 15 percent, and he tossed a curve or slider 19 percent in 2011. Cahill is progressively doing a better job of keeping those breaking balls down, and hitters are swinging over his low curves and sliders.

In 2009, Cahill caught the middle part of the plate often with his curve and slider:

Location of Cahill's curveball and slider, 2009

Fifty percent of his breaking balls were located low in the zone, 29 percent were thrown to the middle portion, and 21 percent were thrown high. Hitters just about never missed those middle and high breakers:

Opponent contact rate by pitch location vs. Cahill's curveball and slider, 2009

Hitters swung and missed at Cahill's curve just 22 percent of the time overall, well below the 28 percent average for starting pitchers. In 2010, Cahill did a bit better in keeping the ball down:

Cahill's curveball and slider location, 2010

He threw 52 percent of his breaking pitches low in the zone, 27 percent to the middle, and 21 percent high in the zone. Those curves and sliders spotted low in the zone induced lots of whiffs:

Opponent contact rate by pitch location vs. Cahill's curveball and slider, 2010

Opponents missed 35 percent of the time they swung at a Cahill breaker. This past season, Cahill very rarely hung a curve or slider:

Cahill's curveball and slider location, 2011

Cahill threw 58 percent of his breaking balls low, 22 percent to the middle of the zone, and 20 percent to the upper third of the zone. Those breaking pitches thrown to the arm side gave hitters fits, and Cahill slightly increased his miss rate with his curve and slider to 36 percent:

Opponent contact rate by pitch location vs. Cahill's curveball and slider, 2011

At worst, the Diamondbacks picked up a young, cost-controlled starter who has gradually increased his workload to 200+ innings and will complement Dan Hudson and Ian Kennedy nicely. But, with Cahill becoming increasingly comfortable using breaking pitches in addition to his quality sinker and changeup, he could yet take a leap forward and become a number two or fringe number one starter. Some think Parker has that upside, too. Still, considering that the D-Backs have an absurd amount of pitching prospects in the pipeline -- Trevor Bauer, Archie Bradley and Tyler Skaggs to name a few -- picking up Cahill is a solid move that improves the team in 2012 and beyond.