You’ll never see a sport with such useless stats as you will in baseball. Baseball statisticians will keep track of any statistics they possibly can just to have something new to tell television and radio audiences. I’ll never have to worry about Nomar Garciaparra’s on base percentage on his birthday when the temperature is below 80 degrees because some guy somewhere has already researched that for me.
Now any GM will tell you that they’re always looking for new strategies to win. Heck, even I’ve personally hours researching players, veterans and rookies alike, poring over every stat acronym imaginable just for fun! I’ve read about AVG, ERA, HR, WHIP, K/9, BB, RBI, OPS, FPCG, E, and SB, the typical stats that any true fan of the game should know. But if you don’t, they’re batting average, earned run average, home runs, walks plus hits over innings pitched, strikeouts per 9 innings, walks, runs batted in, on base plus slugging percentage, fielding percentage, errors and stolen bases. Phew.
But in addition to all of these, one stat that I have never followed before has really caught my eye this spring: BABIP, or batting average on balls in play. The reason that BABIP caught me eye is because it claims to be able to accurately determine something that is usually difficult to perceive in baseball: luck. No doubt this has raised a few eyebrows into the validity of a stat that claims to determine something that is not within the control of humanity.
BABIP seems to rely on the belief that once a ball is put into the field of play, anything goes. In other words, it assumes that there are no set “definite outs” and “definite hits” and any ball put into play that reaches the ground or a glove did so because luck was on a certain side. A high BABIP for a hitter tends to mean that he is extremely lucky. This could be for a number of reasons: he may be hitting ahead of better hitters and this gives him more hittable pitches, he could be making solid contact all season long, or lightning might strike the ball every time he hits it into play and this allows him to reach base safely. The same goes for a pitcher, except that pitchers with an extremely low BABIP are considered very lucky. They might play on a field that slows the ball down often or they might just have pinpoint control for a season and that results in the abnormality. The problem with this is that players don’t reach the big leagues on luck alone. They have to have skill. Regardless of how lucky a player may or may not be, chances are that in the end, good players will have good stats and bad players will have bad stats. Luck can only win over skill for so long.
Renowned baseball statistics website Fangraphs.com has the average BABIP of a hitter around .300 and the average BABIP of a pitcher at about .290. Using these numbers, a skilled general manager is supposed to be able to weed out fluky players from truly skilled ones. Take Jake Peavy for example, who I consider to be truly skilled and not fluky (and if you disagree, see a doctor). Many questioned his 2006 year in which he posted a 4.09 ERA, well above his 3.31 career average, and had only 11 wins, 4 below his average of 15. This caused him to drop slightly in drafts in 2007. A closer look into his BABIP in 2006 shows that it was nearly .320, 30 points above the average mark! This means that if a hitter put the ball into play off of Peavy, they had a good chance of reaching base. But just last year, he won the pitching triple crown of ERA, wins and K’s en route to a Cy Young award winning season. Lo and behold, his BABIP last year? Back down to .292, only 2 points above the average. BABIP shows how his 2006 campaign, which was still respectable, in no way determines what kind of player he really is. It can be said that he just didn’t have luck on his side that year. But naysayers will argue that it is just because he is better player and his skill won out returning him to his usual form last year.
But what about when you worry that a player isn’t really that skilled and might just be riding a tidal wave of luck for an entire season? Is BABIP a trustworthy stat in that situation? Well, take a player by the name of Adrian Beltre, current third baseman for the Seattle Mariners. Before he signed that massive contract with the Mariners, he was the third baseman for the Los Angeles Dodgers. In 6 seasons leading up until 2004, he had never hit more than .290 or slammed more than 21 home runs in a single season. But in 2004, which just happened to be a contract year, he hit .324 with 48 home runs and slugged .629. Pretty abnormal numbers for a previously normal player, but the Seattle Mariners bought it and signed him to a monstrous deal. He would go on to hit only .255 with 19 homers in his first season with the Mariners. And that slugging percentage? It dropped almost 200 points, down to .413. But to a keen general manager, this disastrous contract could have been avoided by just checking the BABIP stat. Beltre’s monster 2004 season saw his BABIP rise to .340, 40 points above average! 40! For a hitter who had never shown any signs of superstardom before, this should have been a telling sign that perhaps a lot of his hitting that year was luck. An educated GM would have ignored Beltre and searched for talent that had more consistent numbers throughout their careers. But still, his 2005 BABIP was only about .255, 45 points below average, meaning that a bit of a rebound should have been expected in 2006. He did rebound, hitting .268 with 25 homers that year. So was it luck or skill? Did BABIP really provide a good view of Beltre as a hitter? He had a monster year, a bad year and a decent year. So which was the real Beltre?
BABIP is clearly not without its flaws. Because it only takes account of balls put into play, home runs and strikeouts don’t count. A hitter could have a very low BABIP, but if they’re hitting 45 home runs every year, who cares? A pitcher might have a high BABIP but if their ERA is perennially in the low 4’s and they consistently have over 200 K’s, it makes no real difference. That’s why BABIP is more effective in predicting players that you don’t know much about. This year, many people will look to players like Josh Hamilton, B.J. Upton and Hideki Okajima to repeat their stellar 2007 seasons. Hamilton looked like a whole new person once he finally reached the majors, hitting .292 with 19 homers in only 298 at-bats. But his BABIP was slightly high, hovering around .315. Injuries aside, he is a bit of a gamble, but still worth the risk in my book if you want to consider his BABIP. Upton had a great campaign in 2007 but his BABIP was at .400 in more than double the at-bats at 498! That is an astronomical 100 points above the average! This guy must have had his bats lined with four leaf clovers. And Okajima, the great story as the “other Japanese guy” on the Boston Red Sox, posting a tiny 2.22 ERA with 63 K’s had a BABIP of about .250, 40 points below the average. The differences may be small in some cases, but these are all unproven talent. If someone like Upton cannot even come close to living up to his 2007 level, and his BABIP, if believed, suggests that, then there will be a lot of unhappy Rays fans.
The thing that BABIP exemplifies so well is obsession with statistics. All statistics tell a different key point but they all share one common flaw: they only give details about the past. Statistics can be helpful in determining what a player will become but they are all heavily flawed because none of them talk about the future. BABIP is a great example because it shows just how crazy baseball people have become in using numerical values for luck in deciding on players. No stat is going to tell you how a guy will do in the future. He might get hurt, he might just struggle mentally or he might simply lose interest in the game. There is no single great stat that is going to produce a winning team.
So put the pencils down, GM’s, and just think. Stats aside, can the guy simply play the game?
Recent Comments