Groundballs and You
During a discussion on fifth starters, I ended up doing some research on ground balls and their relative worth as compared to non-ground ball batted ball types. What began as a throwaway point has since captured my attention and so I bring some results to you.
First, I looked at batted ball types in complete isolation. I believe it is widely accepted that a ground ball is better than a fly ball from a pitcher’s point of view on average. Arguing whether it is better to be a ground-ball pitcher or a fly-ball pitcher is not the scope in question here. A huge amount of complexity resides in studying such a question, some of which I explore below, but for this first initial look, I just wanted to know nothing more than the relative weight of a ground ball versus a ball in the air on average run scoring.
I need to make a quick note on the terminology here. A ground ball is any batted ball classified as a ground ball or a bunt. A ball in the air is any batted ball classified as a fly ball, a line drive or a pop up. These are my own distinctions. I used the batted ball classifications provided by MLBAM since that is what I had available to me in easy to use database form. I do not expect that the results would be vastly different using other sources since large samples are in play.
I have two sets of information that helped me determine this figure. The first is the average number of outs recorded on each batted ball type. The second is the average run value, derived from changes in score and the run expectancy matrix after each play per batted ball type. For this first isolated comparison I used totals from the American League in 2009. I looked at a couple other years and the National League as well and the numbers change only slightly.
The results were that the average ground ball generated 0.04 runs and caused 0.80 outs while the average ball in air generated 0.23 runs and caused just 0.62 outs. On a runs-per-out basis, balls hit into the air created almost 7.5 times as much offense as balls kept on the ground did.
What constitutes a line drive is somewhat fuzzy and open to subjective bias. However, even with ignoring line drives, fly balls and pop outs by themselves generate an average of about 0.1 runs and cause 0.79 outs. That rate is still about three times more offensive than the average ball hit on the ground. The additional risk of yielding a home run matters, a lot.
Tomorrow, I am going to look at home run rates in closer detail.
Matthew Carruth is a software engineer who has been fascinated with baseball statistics since age five. When not dissecting baseball, he is watching hockey or playing soccer.
I heard it mentioned, I think at BP, that a fly ball in play and a ground ball in play are approximately equivalent. Is this true? Does the .1R/.79O include home runs or just fly balls in play?
It includes home runs.
Looking at just balls in play would make them close to even, but I don’t find that particularly meaningful.
That wasn’t supposed to be a vote. I’m a moron, I meant to reply.
Is GB vs. Non-GB any more meaningful than Pop-up vs. Non-Pop-up? You’re grouping in a lot of disparate in the air events. Do pop-ups have a greater risk of becoming HRs than GBs?
What about a player like Jose Reyes? I don’t have the stats in front of me and I am just a neophyte SABR person, but I was under the impression that when he was putting the ball on the ground at the start of his career and beating them out for hits his odds of scoring a run would increase, whereas when he was hitting the ball in the air during recent years he would fly out more? Or do the odds of hitting the home run or the double in the gap (or in his case a triple) outweigh the benefits of getting the ground ball infield hit no matter what?
On a per batter basis, the results fluctuate according to the hitter’s skills. This is showing league-wide trends.
If this is true, then why are GB hitters so inept at producing runs?
I believe the notion of your question is only reaffirmed by the article.
Interesting. Very Interesting.
For line drives, someone needs to come up with a “feet traveled per second” calculation that allows for a range to be defined as a line drive. Feet traveled before it hits the ground would be tricky, especially since some line drives presumably hit the wall or close to it, and some land in no man’s land between the infield and outfield. But I imagine the technology to estimate this is either here or just about.
I don’t know of anything (other than the Field f/x system) that measures the 4th dimension (time) right now. But if you had that, then you could use it to measure the quality of a the whole array of balls in play, not just line drives.
HITf/x data would give you a decent estimate of hang time. Some teams are using this data, unfortunately it’s not available to the public yet except for a small sample from April 2009.
Trackman measures the hang time, and the whole trajectory for that matter, very accurately. It’s also not available to the public.
@glen
I suppose a player like reyes would benefit somewhat from putting thbe ball on thbe ground. But these calcs are the average values of balls on the ground vs in the air. Reyes is far from tbhe average player, so (if his groundballs produce less outs than the avg player) he is an exception. Also, he wud benefit from hittin g more ld’s himself
Reyes has an OBP of .272 and an Iso of .03 on ground balls, compared to a league average of .235 OBP and .02 Iso.
His fly balls go the opposite way, with an OBP of .196 (.218 league average) and an Iso of .295 (.377 league average).
Ichiro is even more skewed: .322 OBP/.016 ISO on ground balls vs. .153/.195 on flyballs.
They’re both just a certain type of hitter where ground balls are more productive than fly balls.
Basically, limited power, but lots of speed.
Also a significant amount of bat control and a propensity for bunting, both of which complement their skillsets.
Matthew, a question regarding run expectancy. Do you use an average change in run expectancy over the whole 24 base-out situations, or do you calculate the change from the frequency of the different states that an average hitter would face? In other words, a hitter would certainly face more no on, no out base states than bases loaded, for example. Certain base states would come up more often than others.
I hope I worded that in a way that made sense.