We're in the midst of a statistical revolution in baseball, a fascinating time in which next-level measures give us a greater understanding of player skill, and in which statistical advances every year teach us new things about the game.
Player analysis has become quite different in the 21st century: These statistical innovations provide us a virtual toolbox giving us a variety of tools with which to measure player skill.
To extend that metaphor, we've reached -- at least in this columnist's opinion -- a point where BABIP is appreciated merely as one such tool in the box, rather than the only tool. It's the Phillips-head screwdriver in your toolbox, rather than a tool that transcends the box; say, a pocketknife that contains said screwdriver.
Still, any amateur handyman is going to rely heavily on the Phillips head, just as many fantasy owners continue to lean on BABIP. For instance, perhaps you've heard this line during the winter: "Andrew McCutchen has no chance at batting .327 again in 2013 because of his absurdly high .375 BABIP."
Before you get deep into your player analysis, make sure you fully appreciate such tools at your disposal as BABIP. McCutchen's critics might have a point. However, if their point is that his BABIP was "78 points too high," then they misunderstand the statistic. Flippant conclusions using this category are the most dangerous any fantasy owner could make, and they discount the complexities of BABIP and the underlying influences that result in that final number.
BABIP is not a one-size-fits-all tool, not your pocketknife. Know this before you adjust your personal projections assigning McCutchen a .280 batting average.
This is the third year for which we've published a "BABIP Primer" in our Draft Kit -- here are links to the 2010 and 2012 editions -- and we continue to learn new things about the category each year. No one is ever too seasoned for a refresher course on, or at least an open discussion about, the category.
What, precisely, is BABIP?
BABIP, or Batting Average on Balls In Play, measures both a hitter's success producing hits or a pitcher's ability to prevent them only on batted balls put into the field of play. It's a recalculation of batting average only on those batted balls in play, meaning excluding the "three true outcomes" of home runs, walks or strikeouts. Home runs might be the puzzling exclusion; yes, those are batted balls, but they depart the field of play, and therefore aren't a useful measure of defensive influence on batted balls.
The formula is Hits minus Home Runs, divided by At-Bats minus Home Runs minus Strikeouts plus Sacrifice Flies, or:
H - HR
AB - HR - K + SF
Voros McCracken is widely credited with having "invented" BABIP, but others contributed to the research behind the theory, and many since then have offered their varying opinions on its importance as well as its formula. Some sources exclude sacrifice flies (SF) from the formula, and others give it a different name, such as Ron Shandler's Baseball Forecaster, which calls it Hit Percentage (H%).
What purpose does BABIP serve?
The most significant pitfall of BABIP for fantasy owners, and the A-number-one way they give it the pocketknife treatment, is to equate the category with "luck." BABIP is not that one-size-fits-all measure of luck that many believed in the past -- and hopefully don't still believe today. If you're quick to say that Dexter Fowler was the "luckiest hitter in baseball" in 2012 because of his major league-leading .390 BABIP, Justin Smoak was the "unluckiest hitter" (league-worst .242 BABIP), Jered Weaver was the "luckiest pitcher" (league-low .241 BABIP) and Rick Porcello was the "unluckiest pitcher" (league-worst .344 BABIP), without doing any deeper examination, you are misusing the stat.
While it's true that luck contributes to BABIP performance, it is only one of several such influences. Luck can come into play when a player gets a fortunate bounce on a ground ball, squeaking between a third baseman or shortstop, or a line drive or fly ball dunks in front of an outfielder. In either of those scenarios, though, the defenders' abilities presumably also contributed to the result.
The purpose of the category, therefore, is not to identify outliers, but rather illustrate a player's success when he put the ball in play. And to extract value from it, you must understand the context -- how he got there and why he will or will not repeat it -- by which the player came to that result.
What is the major league average for BABIP?
It varies, which illustrates a critical aspect of BABIP: annual fluctuations, both by individuals or the league as a whole. For example, in 2012, the major league average for BABIP was .297, up two points from its 2011 number (.295). Since the turn of the century, the league's BABIP has varied, ranging as high as .303 in a single year, in 2007, to as low as .293, in 2002. And if you flash back before the Steroid Era, league BABIPs were even lower -- as low as .282 exactly 25 years ago (1988).
It is often assumed that a league-average BABIP is .300, and that every player should therefore regress to that mean. Not so: While it has typically ranged between .295 and .300 the past half-decade, it can often change depending upon era.
So what, then, are these other factors that influence BABIP?
There are several, and we keep getting more reliable data on new ones each year. Let's address each of these contributing factors one by one:
• Raw hitting skills: No one will attempt to convince you that Miguel Cabrera and Jamey Carroll are skills equivalents with the bat. Cabrera led the major leagues in home runs (44), slugging percentage (.606) and well-hit average (.296), the latter the measure of at-bats that resulted in hard contact, in 2012. Carroll, meanwhile, has hit one home run in his past 1,406 at-bats, and had the second-worst slugging percentage (.317) and worst well-hit average (.119) last season. And, unsurprisingly, Cabrera's BABIP (.331) was 25 points higher than Carroll's (.301).
To put it another way, among the 11 leaders in batting average the past three seasons combined (among those with 1,500-plus plate appearances), seven also ranked among the top 11 in BABIP. Meanwhile, none of the players with a .300-plus batting average during that time had a BABIP beneath .310.
There is no question that the better the hitter, the better the BABIP. It is for this reason that a player's single-season BABIP should never be analyzed without comparing it to that individual's history in the category.
• Raw pitching skills: The same lesson also applies to pitchers -- Henderson Alvarez is not as equally skilled as Clayton Kershaw -- albeit to a slightly lesser degree. Unlike hitters, pitchers exercise less control over batted-ball outcomes, a point illustrated best by the range of BABIPs posted on either side of the ball: From 2010-12 combined, among pitchers who faced 1,500 or more batters, the major league low for BABIP was .245 (by Jeremy Hellickson) and the high was .329 (by Jeff Francis), and 87 of the 112 pitchers to meet that minimum (77.7 percent) posted a BABIP between .280 and .320.
Meanwhile, among hitters to come to the plate at least 1,500 times from 2010-12, the major league low for BABIP was .251 (by Carlos Pena) and the high was .370 (by Austin Jackson), and only 60 of the 114 hitters to meet that minimum (52.6 percent) posted a BABIP between .280 and .320.
Again, make sure to compare a pitcher's history in the category against his individual-year numbers but understand that fluctuations are a bit more telling on their side because of diminished control over their results.
• Types of contact: Whether a batted ball becomes a line drive, ground ball, fly ball or bunt has a bearing upon its result. Naturally, a line drive is the most likely to result in a hit, a fly ball the least likely, so a player's percentage of batted-ball types have influence upon his BABIP.
These were the major league averages for each batted-ball type in 2012:
Ground balls -- bunts included (46.0 percent of all balls in play): .230 BABIP
Ground balls -- bunts excluded (43.8 percent): .226 BABIP
Fly balls (35.7 percent): .132 BABIP
Line drives (18.3 percent): .714 BAIBIP
Bunts (2.2 percent of all balls in play): .402 BABIP
Remember that these averages, like overall BABIP numbers, vary from season to season: The 2011 line-drive rate was 18.5 percent and BABIP was .707, for example. So, again, it isn't a hard-and-fast rule that a player's BABIP should result in a number identical to the league average, even when breaking down batted balls further.
This is another area in which a hitter exercises greater control than a pitcher, specifically pertaining to line drives: Joey Votto was the majors' leader in line-drive percentage from 2010-12 combined (24.2 percent) and Mark Reynolds (15.1 percent) had the majors' worst number; on the mound, Kyle Lohse (20.7 percent) afforded the highest line-drive rate, Kershaw the lowest (15.7 percent). And of the 114 hitters to come to the plate 1,500 times during that span, 81 managed a line-drive rate between 16 and 20 percent (71.1 percent). Of the 112 pitchers to face that many hitters, meanwhile, all but six managed line-drive rates between 16 and 20 percent.
• Quality of contact: Here's where that "well-hit average," or statistics related to hard contact, comes into play. Generally speaking, the harder the contact, the greater the potential results for the hitter -- with the one minor exception being that speedy ground-ballers probably want to generate weaker contact to use their speed to leg out infield hits. Last season, these were the league average BABIPs by the three classifications of quality of contact:
"Hard" contact (24.8 percent of all batted balls last season): .623 BABIP
"Medium" contact (18.7 percent): .370 BABIP
"Soft" contact (56.5 percent): .146 BABIP
This isn't difficult math: Players, on average, generate soft contact more than half the time, and more than twice as often as they make hard contact. When they make hard contact, however, they are more than four times more likely to get a hit.
The following chart lists the top -- and bottom -- 25 qualified hitters in terms of BABIP in 2012, and their rankings in terms of percentage of balls in play judged hard contact (among 144 batting title-eligible players):
Here is where it gets interesting, returning to the McCutchen example at the top. McCutchen was the only player in baseball to manage a top-10 ranking in both BABIP and hard-contact rate last season, the most compelling explanation for his .375 BABIP and resulting .327 batting average. Any argument against his ability to repeat the BABIP must, therefore, assumes he cannot repeat his hard-contact rate. That's a fair point to make, considering he has a .322 career hard-contact rate, but also one that ignores that he had a .352 number in the category in 2011.
Now, here are the top (and bottom) 25 qualified pitchers in terms of BABIP in 2012, and their rankings in hard contact allowed (among 88 ERA qualifiers):
Pitchingwise, Adam Wainwright's differential should draw your attention. For a pitcher who afforded hard contact 13th-least often among ERA qualifiers, it's puzzling that he had the fifth-worst BABIP in the game -- especially if you consider that, as a pitcher who strikes out a lot of hitters, he allows fewer balls in play than your typical pitcher might. Couple that with his ongoing recovery from Tommy John surgery, and there's a compelling argument for a big step forward from him in 2013.
Speed of the hitter: The quicker the runner, the more he can drive batting average (and with it BABIP) with his legs, beating out bunts and ground balls for base hits. To that end, three of the four players to steal 40 or more bases in 2012 had BABIPs of .325 of greater, and all four had a BABIP higher than the major league average of .297. And, looking at the charts above, the three players who ranked among the game's top 25 in BABIP who also ranked among the 25 worst in hard-contact rate are all generally known for being quick: Jon Jay, Derek Jeter and Alcides Escobar.
We can break out BABIPs by batted-ball type, and doing so with only ground balls provides telling data about many of the game's most prolific speedsters. Listed below are the 25 best and worst in ground-ball BABIP in 2012, along with their stolen base totals and overall BABIPs:
• Quality of the defense: This is a key aspect of pitching analysis, and it needs be stressed that a pitcher's BABIP allowed might not be the best measure of his ability. It might instead tell us more about the quality of the defense behind him. For example, a pitcher isn't necessarily "unlucky" just because neither Yuniesky Betancourt nor Delmon Young had the range to catch that blooper to shallow left field; he is only "unlucky" because this was the day that Charlie Manuel decided it was a good day to start both Betancourt and Young, both poor defenders.
To illustrate how defense can influence BABIP, consider that, of the 13 teams to post a BABIP beneath .290 last season, six ranked among the 10 best in terms of Defensive Runs Saved. Those six teams featured five of the top six ERA qualifiers in terms of BABIP: Jered Weaver, Los Angeles Angels (.241, 1st); Ervin Santana, Angels (.241, 2nd); Mike Minor, Atlanta Braves (.252, 3rd); Jason Vargas, Seattle Mariners (.254, 4th); Jeremy Hellickson, Tampa Bay Rays (.261, 6th).
The Rays are perhaps the best example of a team's defensive influence on BABIP. They have ranked among the two best squads in the category in three consecutive seasons, leading the way in each of the past two, and have finished among the 10 best teams in Defensive Runs Saved in each of those years, leading the way in 2011. Hellickson's critics, in particular, need to take this into account. His oddly low BABIPs in back-to-back seasons are driven in part by the quality of his defense.
• Direction of the batted ball: This might seem an obvious one -- hit a line drive barely fair within the outfield foul lines and you're effectively 100 percent certain to get a hit -- but it runs deeper than merely "hit it where they ain't." Hitters can exercise some degree of control over the direction of their hits, and pitchers can throw pitches to a certain area of the strike zone in an attempt to get a batter to pull the ball, just to name two examples.
But this influence is most relevant when coupling it with the previous one. When defenses shift to cover the area of the field where the opposing hitter is most likely to hit the ball, the hitter's margin for error decreases. Look at the ground-ball BABIP chart, specifically the "25 worst" side, and you'll see two left-handed hitters who were notorious for facing defensive shifts last season, Adam Dunn and Eric Hosmer. Their poor BABIPs -- and resulting poor batting averages -- were largely the result of an inability to adjust on balls batted into play, hitting ground balls and line drives more often into covered area of the field that might have been left completely unguarded for an average hitter.
Hosmer's struggles are most relevant, because last season, he pulled 119 ground balls, had a .094 BABIP and 12 total hits on those. In other words, 27 percent of the time he put a batted ball in play, it was a grounder he pulled, meaning he was playing right into opponents' defensive shifts. There's no better explanation for Hosmer's struggles than that, and he's going to need to make some serious adjustments if he's to break out in 2013.
• Park factors: Ballparks can influence BABIP, which makes sense, considering no two have identical combinations of outfield dimensions, playing surface, foul territory and weather conditions. The total area of the field of play -- foul territory included -- in particular influences batted-ball results.
For instance, the O.co Coliseum, home to the Oakland Athletics, is widely regarded as one of the more pitching-friendly venues in baseball, mostly thanks to its spacious foul territory that swallows up a healthy number of harmless popups. As those drive down BABIPs, it's understandable that the O.co Coliseum has ranked among the five worst ballparks for BABIP in each of the past three seasons, in 2012 ranking 29th-worst with a .272 number. Conversely, Colorado's Coors Field had the highest BABIP in baseball at .345, understandable if you consider the thin air coupled with the spacious outfield dimensions at the venue. There is plenty of space in those outfield gaps for hitters to send those doubles and triples.
These were the BABIPs at all 30 major league ballparks in 2012:
Is BABIP a more valuable evaluation tool for pitchers than for hitters?
As hinted above, BABIP is more valuable in your analysis of pitchers, and the reason is that, due to all the influences above, random fluctuation impacts the category more on that side of the ball. Pitchers have less control over the outcomes of individual plays than hitters -- at least those that are not walks, strikeouts or home runs -- and you can expect more year-to-year BABIP variance as a result.
So what conclusions, then, can we draw from BABIP?
The best use of BABIP in your analysis is to examine a player's history in the category, as well as investigate how he got to that season's number. What kind of hitter or pitcher is he? Was his 2012 BABIP radically different from his previously established career norm? If it changed, is there an obvious reason why, such as a shift from being a ground-ball to fly-ball hitter, a change in ballparks, a change in the defense behind him (if he's a pitcher), or perhaps even an injury that might have resulted in diminished skills?
Torii Hunter's career-best .389 BABIP last season is an excellent such case to examine. That number was 56 points higher than his previous single-season best (59 if you count only batting title-eligible seasons) and 82 points higher than his career mark in the category (.307), and you can see in the hard-contact chart above that he ranked among the game's bottom half in terms of making hard contact. He also hit a higher rate of ground balls (52.8 percent) than he had in any other season for which batted-ball data was available, and as a 37-year-old he wasn't exactly as fleet of foot as he was a half-decade ago; he had only nine steals in 2012. If you're looking for a player whose BABIP belied his true batting-average skills, Hunter qualifies.
Also consider calculating Expected BABIP statistics based upon performance by batted-ball type, and then comparing it to his actual number. This has been discussed in our past "BABIP Primers," but it can help put those statistics into perspective. Opinions on Expected BABIP formulas vary -- some say you should just add .12 to a player's line-drive rate, which seems arbitrary -- but the best formulas calculate the major league average BABIPs by batted-ball type, multiplied by the player's batted-ball percentages.
Let's take it a step further: We'll use league-average BABIPs on soft, medium and hard contact on each batted-ball type.
Here are those league averages:
Understand that the formula does not account for a player's speed; there isn't an easy way to include it in the calculation -- at least not yet.
The following chart ranks the 25 hitters who had the most significant differential between actual and expected BABIP last season (400 plate appearance minimum):
Now, the 25 pitchers who had the most significant differential between actual and expected BABIP (100 inning minimum):
In conclusion, I'll reiterate the most important rule, one made multiple times in this space: BABIP can never, and should never, be regarded as the driving force behind your player analysis. Again, it is merely another tool in the box; you wouldn't open an auto repair shop armed only with a pocketknife, would you?
BABIP is a valuable, underrated and often misunderstood tool. And with this column's help, hopefully you've now mastered it.