April 2, 2010
Checking the Numbers
A few weeks ago I found myself engrossed in a Tommy Bennett article on the Braves and stumbled upon his usage of the term SHINO when describing Melky Cabrera. The acronym stands for Switch-Hitter-In-Name-Only, and refers to some hitters with 'S' or 'Both' under the Bats column on their player pages, and specifically the ones who might want to think about changing that status. They certainly switch, but they don’t offer much in the way of hitting. The term tickled my fancy, in part due to the fact that I’ve had an article on switch-hitters in my to-do queue for over a year now that was set to focus on those who consistently struggled from one side of the plate. Though the title of that shelved article involved Bobby Kielty and not this term; as we’ll see, maybe Kielty should have been included in the title.
Switch-hitters are fascinating, not only because they form their own platoon, but also because of their relative scarcity. There are only 227 unique hitters from 1974-2009 to amass 100 or more plate appearances from each side of the dish in at least one season. Theoretically they allow teams to gain an advantage on offense regardless of pitcher handedness, but what happens if this self-platoon doesn’t work? Who are the consistent SHINOs, frequently falling short from one side? And if so, at what point might these hitters benefit from just sticking with their dominant hand?
What Is a Switch-Hitter?
Simply put, a switch-hitter is a batter whose lead hand is determined by the handedness of the pitcher. Unless Pat Venditte is on the mound, this batter swings with his right hand out in front when southpaws toe the rubber, and leads left against northpaws. The benefit of switch-hitting springs from the general fact that hitters perform better against opposite-handed pitchers. Studies have shown the effect is undeniable in the aggregate, even if a few curve-busters surface every now and again (Raul Ibanez in 2009, for example). Over the last five years, here are the Raw TAv breakdowns for each general matchup:
The effect is more pronounced for lefties predominantly due to the fact that righties usually account for upwards of 70 percent of the batters faced in a season; from 2005-09 it is actually 72.6 percent. Regardless, it is easy to see here that the offensive improvement is in the range of 50-70 points of RTAv when facing opposite-handed hurlers. With switch-hitters, instead of a righty facing his brethren 70 percent of the time and lefties in the remaining 30 percent of plate appearances, teams get to truly optimize the matchups.
Switch-hitters carry an inherent selection bias in that, technically, only those capable of succeeding with a bat in either hand get the green light to act upon this ability. Vicente Padilla liked to joke that he was a switch-hitter when he pitched for the Phillies, but was he really? If the purpose of switch-hitting is to create added value from a self-platoon, a poor hitter is just superfluously switching sides, lacking the “-hitting” part of switch-hitting. If the choice was between 100 points below average when switching and 180 points below average when staying put, the player probably shouldn’t be hitting against that pitcher to begin with.
This is where things get tricky, however, because the opportunities to face different-handed pitchers are not evenly distributed, revealing plenty of noise in the data. A switch-hitter might struggle as a righty against lefties in one season, and lead his colleagues in production in that matchup the very next year.
Are SHINOs Real?
Do these supposed phonies exist? To find out, I computed the Raw True Average (or RTAv) for all switch-hitters from 1974-2009; I often use 1974 as the first year in any of these studies, because Retrosheet is missing a few games in each year before that point. I also computed the league-average RTAv for each hand in each year, and the delta between the individual and the league. For example, in 1995, Luis Alicea had a .797 RTAv as a righty, and a .788 RTAv as a lefty; the league averages were .788 and .789 respectively, so he was slightly above average from the right side of the dish and the definition of average from the left.
In identifying SHINOs I am looking for players consistently below the league average in a specific opposite-handed matchup for a few seasons, while succeeding from the other side of the plate. The individual must be compared to the league, and not himself, though, as self-comparisons do not usually explain the entire story. In 2004, Lance Berkman was 68 points above average as a righty and 248 points above average as a lefty; his self-delta was 195 points (1.051 to .856), which is big, but he remained well above average from both sides of the plate. Criticizing his split makes little sense because he performed very well in relation to the more meaningful comparative baseline.
Shine On, SHINOs
Since lefties don’t get to throw nearly as much as their right-handed peers, the minimum PAs for a righty hitter against a southpaw was set to 70, with the minimum for a lefty hitter facing a northpaw at 120 PAs. With that in mind, I went looking for career trends of at least four out of five or six years during which a hitter consistently fell below the same average matchup RTAv by 20 or more points while being five points below average or better from the opposite side. The other side of the ledger is a key here, as it really doesn’t matter what hand Cesar Izturis uses, since he can’t hit from either side of the plate.
I was able to identify 13 players who displayed some sort of lasting trend in their career as switch-hitters to be lacking from one side of the dish. The players and the years in question:
A motley crew of sorts; it's interesting to note is that both J.T. Snow and Jose Valentin show up on this last, given that Snow ceased switch-hitting after the 1998 season, batting exclusively as a lefty, and that Valentin hinted at and attempted to remain a lefty in his latter years. Why did both players make this decision? Qualitatively the decision hinged on how poor they had been hitting from the right side, and the belief that they could better help their teams by just always batting lefty.
Translated to math, the implicit assumption was that the difference between the inherent platoon benefit and the decreased probability of success that stemmed from facing same-handed pitchers was nowhere near significant enough to truly merit their continuing to serve as switch-hitters. In other words, they had become so poor at batting righty that it actually made next to no difference if they became permanent lefties, as the driving force of the decision to switch-hit—achieving the platoon advantage—was rendered moot.
When studying aspects of the game such as this, the generally accepted methodology is a matched-pair analysis. An example would be comparing swingmen as starters to when they relieve; we can use the same pitchers to determine the expected change in performance. That type of research method doesn’t work here, as switch-hitters not named Snow tend not to completely stop hitting from one side of the plate, and the majority of the plate appearances belonging to switch-hitters facing same-handed pitchers stem from Tim Wakefield being on the mound; they didn’t want to mess their swing up. The decision to stick to one side should be determined by comparing the RTAv of the switch side up for discussion to the matchup average for the same-handed side.
To better explain, consider the following scenario: Russell Carleton is a switch-hitter consistently below average when batting lefty, while above average as a righty. The league average for the LHB vs. RHP matchup is .794 while he generally clocks in right around .735. The matchup average for LHB vs. LHP is .722, so he is still technically benefiting from the switch, even if it doesn’t feel like it. If his RTAv as a lefty was .640, then the benefit disappears, and one could suggest that he really wouldn’t lose anything by remaining a permanent righty.
I’m not saying that this is the greatest method in the world, but it’s at least a start. Did any of the 13 players above fall into this category, by hitting so poorly from one side of the plate that they were actually 20 points below the same-handed average?
Will the Real SHINOs Please Stand Up?
Situations like this are not all that rare relative to the previously established diminutive sample, interestingly enough. In 1999 for example, Alicea had a putrid .573 RTAv from the left side of the plate, well below the LHB/RHP average that year of .816. The .573 also fell below LHB/LHP average of .758 by a significant margin. It’s easy to say with hindsight that he didn’t gain anything by switch-hitting, but this did not occur for any of his other four seasons tabled above. For fellow lefty strugglers, Jose Cruz Jr. and Bobby Kielty met the criteria in three years out of four.
As righties, Roberto Alomar shows up in 1993, ‘95 and ’97. Carl Everett in 1995, ’98, 2002-03. Jerry Mumphrey in 1979-80, ’82, 1984-85 and Ron Oester in 1981, ’83, ’85-86. J.T. Snow wisely stuck to the left side after struggling miserably from 1996-98. But none of the aforementioned players holds a SHINO candle to Jose Valentin, whose numbers merit their own table below:
The numbers indicate that for the vast majority of his career, Valentin was markedly above average as a lefty facing righties, but so poor when batting from the right side of the plate that he actually fared much worse than would an average lefty hitter against a steady supply of same-handed pitchers. This is not to suggest that Valentin would have automatically performed at an average clip against lefties as a lefty, but that there was definitely no benefit gained by his employers when he switched things up. I’m not sure why the idea of platooning a switch-hitter is so taboo, because in reality the term does nothing more than satisfy our binning cravings if the player in question cannot perform while switching.
In the case of Valentin, who was clearly a solid hitter from the left side, it may have been in the best interest of the opposition to bring in lefties no matter if he were to switch, because he wasn't going to perform regardless of whether or not he stood on the opposite side of the plate. I’ve talked about bins and categories quite a bit this offseason, but SHINOs mark another area worthy of further investigation when it comes to matching actual value with its perception. I’d love to stock my team full of switch-hitters, but that's if they were close enough in contrast to a specific-handed crusher that could be a platoon partner so that the extra benefit of saving a roster spot pushes him over the top. I would not have used Jose Valentin in this capacity, because there was no benefit other than the novelty of having a switch-hitter. Without benefits, what’s the point?
Just like switch-hitters who earn their keep, SHINOs are rare, but their performance matters in the context of their eventual platoon possibilities just as much as someone operating solely from one side of the plate.