We continue with the **Batted Ball Project**, which is our analysis of the relationship between various types of batted balls and those components of offense that derive from batted balls. We do this for 537 MLB qualified hitters with at least 1,500 plate appearances during the time that batted-ball stats are available (2002-14).

We start with these three stats:

- BABIP (batting average on balls-in-play);
- 2b3b / Bttd, which is doubles + triples per all batted balls; and
- HR / Bttd, which is homers per all batted balls [
*HRs are batted balls but not balls-in-play*].

And we apply those to get our “batted-ball” components of OPS:

- BACON (batting average on contact or hits per batted-ball)
- BtISO (“batted-ball ISO” — essentially ISO with batted balls as the denominator rather than at-bats).

This go-around we will look at how line-drive rate affects each.

***

First, note that the range of line-drive percentage is much narrower than ground balls. In an effort to make the charts more comprehensible (as much as possible), I’m keeping the scale constant between articles. So the scatterplots of LD% will look at lot more “bunched together” than the GB% plots, but while this is less pleasing to the eye, it will allow the “slopes” of the trendlines to be comparable, which is helpful.

***

So first the impact of line drives on BABIP:

Here we see the influence is pretty strong (the trendline slopes upward pretty strongly), but the scatterplot remains pretty “blob-like” and thus our “goodness-of-fit” (R-squared of 0.2) is not as strong as we might expect.

Nevertheless, the assumption that more line drives will result in more hits is borne out.

***

And, of course, line-drive hitters will crank out doubles and triples, right?

Oops …

There’s a positive relationship, but it is barely perceptible and the “goodness-of-fit” (R-squared of miniscule 0.01) indicates that line drive percentage doesn’t appear to explain anything at all about hitting doubles and triples.

Well then.

It seems that **the primary value of line drives is in the production of singles**. We can’t really conclude that hitting more line drives results in more doubles and triples.

***

And do line-drive hitters also punch them out of the park?

Not really.

The more line drives hit, the fewer home runs were hit, though the scatterplot obviously is not really indicative of anything meaningful and the small R-squared confirms as much.

It turns out that there’s not really any meaningful relationship between line drives and home runs.

***

But line drives have to be better than ground balls, right?

Yes, but not all that much according to what we’re finding.

Here’s the influence of line-drive rate on reaching base by batted ball:

Recall that the relationship of ground-ball rate to BACON was perfectly neutral. Here, we’ve got a positive relationship, but the “blobby-ness” of the scatterplot and the low “fit” (R-squared of 0.04) indicate that line-drive rate is not telling us much at this point.

[As we circle back around, we’ll find out that line-drive rate is most important for hitters who *don’t* consistently hit the ball hard in the air.]

***

And the influence of line-drive rate on batted-ball ISO:

Here again, although the trendline indicates a negative relationship, the shape of the scatterplot and the very low “goodness of fit” indicate that it doesn’t mean much.

***

I found the lack of clear positive influence of line-drive rate to be quite surprising. But there it is.

The positive effects of increased line-drive percentage (among the full universe of qualified hitters with an MLB track record) show up only in BABIP, and then only for singles, and even that is not a very potent relationship.

Overall:

- Line drives are
**positive**in terms of**reaching base**(but not super-positive)

- Line drives are
**neutral**in terms of**producing offense**

Here’s part 5.