Scott Feldman: The Poor Man’s Brandon McCarthy

Brandon McCarthy is a popular guy in sabermetric circles. He’s a good pitcher who reads FanGraphs and looks at PITCHF/x data, and in his spare time, interacts with fans on Twitter. McCarthy’s also a free agent this winter, and his ability to throw strikes and get hitters out is going to make him a popular guy with teams looking to upgrade their rotation as well.

Outbidding everyone for the popular guy doesn’t always lead to great results, though, and there are teams who will be looking for a rotation upgrade who won’t be able to afford McCarthy. So, to those teams, I present Scott Feldman, a free agent starter with none of the publicity who might be able to do a pretty decent impression of McCarthy next year.

While McCarthy’s had a lot more success than Feldman, the two are probably more similar than you might think. Let’s start with their pitch types, and because McCarthy overhauled his repertoire in 2011, we’ll focus on just the last two seasons.

Name	FB%	FBv	CT%	CTv	CB%	CBv
Scott Feldman	34%	92	32%	90	22%	77
Brandon McCarthy	41%	91	38%	90	18%	80

Those are their pitch classifications from just their starts, by the way – Feldman’s also worked out of the bullpen, but I’m excluding those outings because that’s a totally different animal, and we’re only concerned with Feldman as a starter going forward. As you can see, they’re both fastball/cutter/curveball pitchers, and according to the breakdowns at Brooks Baseball, both throw mostly two-seam fastballs. Feldman throws it almost exclusively, while McCarthy will use his four seam fastball when he’s behind in the count and wants to get a strike, but primarily pitches with his sinker and cutter. Feldman throws his fastball slightly harder and his curve slightly slower, but for the most part, they’re similar in velocity as well.

This sinker/cutter/curve skillset is notable because it features two pitches that are generally effective against opposite-handed hitters. The cutter and the curve are both pretty good weapons to combat against large platoon splits, while the two-seam fastball is very effective against same-handed hitters. We’d expect pitchers with this kind of pitch-mix to post fairly neutral platoon splits, and indeed, that’s exactly what we see. McCarthy’s wOBA vs LHBs was 11% higher than his mark against RHBs, but Feldman actually posted a reverse platoon split of the same size – he was 11% better against lefties than righties. Given the sample sizes, you shouldn’t expect Feldman to continue posting a reverse platoon split, but this pitch mix does show that he can be effective against hitters from both sides of the plate.

Of course, just having the same repertoire doesn’t mean that they’re the same pitcher, so let’s shift over to the results side of things, starting with plate discipline numbers. Same deal as above; 2011-2012 data, relief appearances excluded.

Name	O-Swing%	Z-Swing%	Swing%	O-Contact%	Z-Contact%	Contact%	Zone%
McCarthy	30%	61%	48%	71%	89%	84%	58%
Feldman	28%	60%	44%	67%	90%	83%	50%

McCarthy’s been in the strike zone more often, which has resulted in more swings, but other than that, pretty similar across the board. Both are contacted oriented pitchers, but since Feldman lives out of the zone a little more, we’d expect his walk rate to be a little higher. And that’s exactly what we see when we look at the core outcomes data.

Name	BB%	K%	GB%
Scott Feldman	6%	17%	45%
Brandon McCarthy	4%	17%	44%

A better walk rate with the same strikeout and same groundball rate makes McCarthy a better pitcher, but we also have to keep each pitcher’s environment in mind. McCarthy spent the last two years in pitcher-friendly Oakland, knowing that he could get away with catching too much of the plate and the park might bail him out. Feldman, on the other hand, had to deal with the heat and humidity in Texas, where the ball absolutely flies in the summer months.

Oakland also has the vast expanse of foul territory, which leads to more in-play outs on pop-ups, which can be a significant benefit to a pitcher. So, while Feldman didn’t live in the zone as often as McCarthy, their approaches might have been different had they switched home parks. While the samples are too small to draw any conclusions from the data, here’s how opposing hitters did against McCarthy and Feldman on the road the last two seasons.

Name	AVG	OBP	SLG	wOBA
Brandon McCarthy	0.262	0.300	0.421	0.311
Scott Feldman	0.253	0.315	0.363	0.299

McCarthy posted a 5.6% HR/FB rate in Oakland, while Feldman ran a 13.4% HR/FB rate in Texas. On the road, it was 8.4% for McCarthy and 8.5% for Feldman. You don’t want to assume that a player’s road performance is his true park neutral level, but it’s worth noting that park factors are probably a pretty big deal in this comparison.

Of course, if Feldman was perceived to be similar to McCarthy in all these areas, his price would be a lot higher, and I wouldn’t be able to call him a “poor man’s” anything. So, why is Feldman’s price likely to be deflated this winter? Simple – he’s been absolutely atrocious at stranding runners the last couple of years.

Since the start of the 2011 season, Feldman has made 23 starts. In those 23 starts, he’s posted a LOB% of just 59.8%. Of the 190 starters who have thrown 100 or more innings over the last two years, that LOB% ranks 189th – only Tim Wakefield has been worse at allowing runners on base to come home. As you’re probably aware, LOB% isn’t very predictive from one year to the next, with just a .18 year-to-year correlation over the last decade. LOB% is determined in part by how good a pitcher is — a bad pitcher will allow more hits with men on base than a good pitcher, because he’s more likely to allow hits in general — but Feldman is a clear outlier in terms of his context-neutral performance and his strand rate.

Feldman’s never stranded a ton of runners — his career LOB% was 68.1% through 2010 — but part of his early career struggles were related to the fact that he wasn’t a very good pitcher. His 4.73 ERA/4.70 FIP/4.71 xFIP over his first 78 starts show that his results were more bad pitching than bad luck. Over the last couple of years, however, Feldman’s peripherals have gotten a lot better while his strand rate has gotten a lot worse. If you’re betting on one of those two to continue in the future, you’re better off betting on his improvements in BB/K than his decline in LOB%.

In a vacuum, you’d definitely prefer McCarthy to Feldman. His track record of success is longer, he’s shown better command, and his improvements after overhauling his repertoire led to solid results in terms of run prevention. With Feldman, you’re betting on regression to the mean and hoping that he’s able to carry a starter’s workload without wearing down.

However, Feldman’s going to come at a much lower price than McCarthy, and most of the markers that project future performance suggest that Feldman’s not that far off from McCarthy in terms of talent. He’s the kind of pitcher you can almost certainly land on a one year deal, and in a better environment for a pitcher, Feldman may very well thrive.

He’s not likely to develop into an ace, but for a team looking for value at the back end of their rotation, Feldman might be one of the best buys on the market.

Update: More proof that Brandon McCarthy is awesome – his response to this post, in a pair of tweets, embedded below.

@dcameronfg @fangraphs just so everyone’s clear, I’ve been doing a Feldman impression. Not other way around. He was huge in my change.

— Brandon McCarthy (@BMcCarthy32) November 9, 2012

@dcameronfg @fangraphs and you’re dead on that he’s probably the best bargain out there

— Brandon McCarthy (@BMcCarthy32) November 9, 2012