clock menu more-arrow no yes

Filed under:

Austin Jackson's BABIP doesn't tell you what you might think

New, 32 comments

This much we know for sure: When Austin Jackson hits the baseball into fair territory, there's good odds he's going to make it safely to base.

How good? Well this season, at about .413, no one in baseball has a higher batting average on balls in play. MVP-candidate Josh Hamilton is second at .396, Colorado Rocky Carlos Gomez is third at .389 and Justin Upton is fourth at .354. The next-highest BABIP by a Tiger is Miguel Cabrera's .342, which ranks 14th in the MLB.

Lynn Henning wrote in the News recently that media relations director for the Tigers Brian Britten dug up the numbers for Jim Leyland to use in a pregame talk with the media. I just grabbed them off Fangraphs myself.

So, if you just look at the number in a vacuum, you assume Jackson must be doing something very awesome to be so far ahead of the rest of the pack in that stat, especially as a rookie. Unfortunately, that's not exactly the case.

To many, BABIP is interpreted to be a luck-driven stat. To others, bringing the word "luck" in makes them skeptical of the topic entirely. So I'm going to tell you it's not luck, it's a product of random variation. The stat is deeper than that, of course. How a batter hits the ball, where the batter hits the ball and how fast the batter is all go into the actual outcome of BABIP. In the end, however, the stat is not exactly repeatable.

In other words, Jackson's BABIP really doesn't tell us a lot about him as a player, and success this season is no guarantee of success next. Actually, having such a high BABIP is a pretty good indicator he's probably going to take a step backward in batting average (and on-base percentage) next year.

So I'm not going to celebrate it.

I do hope to explain it a bit more.

What does a player's batting average one year tell us about his batting average the previous year, or the next year? Not as much as you'd think. Lee Panas wrote in his book Beyond Batting Average the correlation of batting average from one year to the next to be 0.43. In this case, 0 means there is no correlation at all, and 1 means there is perfect correlation. So there is certainly some predictive ability. A batter like Brandon Inge probably isn't going to go from .243 on season to .342 the next. But higher correlated stats are strikeout percentage (.83), walk percentage (.80), isolated power (.76) and home run percentage (.76). So you could expect Inge to continue striking out and have similar power year to year.

So why does batting average vary year to year? Some years, a batter has a higher BABIP than others. Let's look at Josh Hamilton. His average is .361 this season, his BABIP .396. So what about last? His average was .268 and BABIP .319. Do you suppose he changed that much as a batter this offseason? His career totals, for comparison, are a .312 average and .345 BABIP. So I believe a lot of what we've seen in his past two seasons can be explained by random variation from year to year. One year he's less than his career norms, another he's above. In the end, he's the same batter.

What creates a BABIP? A line drive hitter is going to have more of his balls land for hits. A ground ball hitter is going to have less of his balls make it through the infield, but the unpredictability of a ground ball means he's going to have BABIP success. A fly-ball hitter isn't going to find a lot of success, because fly balls give fielders a lot more time to react and get under them to make the out. How fast a batter runs helps, of course, because the same slow grounder hit by Will Rhymes and Alex Avila probably won't have the same outcome. There's more to it than that -- a study by Chris Dutton and Peter Bendix at Hardball Times explained some of the other factors -- but those are all good predictors. So there's a reason you see a batter like Ichiro find a lot of success yearly, but that success isn't necessarily open to everyone to achieve.

In the end, take all the factors that seem to predict whether a batter will have BABIP success, and you still only explain about 35-40 percent in the variation of the stat from batter to batter. The rest? Well, that's kind of random. When a ball is pitched at 90+ mph and it's slicing through the air, and a batter has a split second to swing his bat at an even higher rate of speed than the ball is moving and the two make contact, where exactly the ball ends up is still largely a product of chance. Sure, where the pitch is located matters a bit. And the batter's ability to control what field the ball ends up in matters a bit. But give him a target and ask him to hit it repeatedly, he probably can't. In other words, putting the ball where "where they ain't" isn't exactly a repeatable skill, not for balls that remain in the field of play, anyway. BABIP is still largely a product of random variation. Thus, a batter's batting average is also a product of random variation.

Which brings us back to Jackson and his .412 BABIP. Every year, a batter finishes the season in the .390s. (.400+? that's incredible). It's yet to be carried over to the next season. Derek Jeter in 2006. BJ Upton and Chone Figgins in 2007. Milton Bradley in 2008. David Wright in 2009. They all dropped off the next year, by 30-50 points, or regressed to the mean as the sabers like to say.

Jackson uses all fields well. He's a fast -- not super fast, however -- runner. There's a lot he does right. But when someone tells you how high his BABIP is, realize what it really tells you is that everything is breaking right for him this season. He's almost certainly going to take a step backwards next year in BABIP, possibly a sizable one. Now, none of this says Jackson's BABIP is going to fall off a cliff next year. It's not going to drop to .300 without bad luck showing its face. He's always had above-average BABIP numbers in the minors because of his skill set.

Jackson's having an incredible season, but be warned: When you couple the step back you can expect to see with his BABIP average with the likelihood he'll continue striking out, you can make a good guess his batting average is going to take quite a dropoff, and he won't be nearly as successful as a leadoff batter as he was this season.

That's what Jackson's BABIP really tells us.