Batted Ball Data Analysis - How Consistent Are GB%, LD%, and FB% Year To Year

2 years ago by Jon Anderson

What Appears In This Article? hide

2. Year-Over-Year Analysis

3. Takeaways and Data

4. Conclusion

5. Download Our Free News & Alerts Mobile App

6. More Fantasy Baseball Advice

When we are evaluating hitters, we talk a lot about batted ball profile. When we're talking about this, things like launch angle and ground ball rate are cited quite often. I wanted to make a post to go over these things a bit more in-depth, giving our readers more complete knowledge of the subject at hand.

I also want to investigate whether GB%, LD%, and FB% are something you can count on year-over-year. We often say things like "well his homer ceiling is capped because of the high ground-ball rate," but is that a responsible thing to say?

Does the fact that a player had a high ground-ball rate last year suggest he'll do the same again this year? Stay tuned to find out!

Be sure to check all of our fantasy baseball lineup tools and weekly lineup resources:

Batted Ball Types Defined

You will see different sites give you different numbers for GB%, LD%, FB%, and PU% (pop-ups, sometimes referred to as infield fly ball rate). In fact, these numbers can be starkly different between FanGraphs and Baseball Savant. FanGraphs breaks down these categories here, which is very interesting to read. They admit that the classification may be quite arbitrary, and we should be careful with interpreting that data because of it. That is very honest of them and very true.

This is a complex thing we are trying to put in simple terms. I'm not sure exactly how FanGraphs does the classification, but I can tell pretty much how Baseball Savant does it from doing some coding on the pitch-by-pitch dataset that shows the classification. I will be using Baseball Savant's version of things from here on.

Launch angle is not the only thing that goes into classifying a ball into one of these categories, launch velocity factors in as well. Here's a plot that shows how the classifications break down by launch angle, you'll see some overlap between the categories:

If we have to bin these into angle ranges, this is what I would say:

Ground balls: Below 5 degrees
Line drives: Between 5 degrees and 25 degrees
Fly balls: Between 25 degrees and 50 degrees
Pop-ups: Above 50 degrees

We also talk about barrel rate a lot. It's worthwhile to note that in 2021, 75% of barrels were classified as fly-balls with the rest of the 25% being line drives. The minimum launch angle for a barrel last year was seven degrees (hit by Giancarlo Stanton at 120.1 miles per hour), and the maximum angle for a barrel was 48 degrees (hit by Mike Zunino at 111 miles per hour).

Here's a table of HR, AVG, and SLG for each batted ball type from 2021.

BB Type	HR	AVG	SLG
GB	0	.241	.266
LD	602	.636	.904
FB	5,342	.281	.877
Pop	0	.017	.019

So, a high LD% should lend itself to a higher batting average and a higher slugging percentage (most doubles and triples come from line drives), and a high FB% will lead to higher home run counts. Ground balls can result in decent batting averages as well if the guy hitting the grounders has a lot of speed to leg out infield singles, but they will very rarely result in extra bases (as evidenced by the AVG and SLG being very close there). You don't want slow guys hitting ground balls, and you don't really want weak guys hitting fly-balls.

To demonstrate this, let's take the case of two extremes: Giancarlo Stanton and David Fletcher.

Giancarlo Stanton

BB Type	HR	AVG	SLG
GB	0	.302	.321
LD	13	.762	1.381
FB	22	.321	1.172
Pop	0	.000	.000

David Fletcher

BB Type	HR	AVG	SLG
GB	0	.267	.289
LD	1	.490	.652
FB	2	.141	.228
Pop	0	.024	.024

You can see the big difference in fly balls there. Stanton got a high batting average and a high slugging out of his fly balls, while Fletcher hit just .141 and slugged .228 when hitting the ball in the air. This just goes to show that not all fly ball rates (and ground ball rates and line-drive rates) are created equal - so you have to keep in mind the whole picture when looking at it.

Year-Over-Year Analysis

The question I wanted to answer was this: how consistent are these GB%, LD%, and FB% statistics? Can we count on a player that has a high FB% in 2021 to do that again in 2022?

The way I went about answering this was by compiling a list of all players with at least 100 plate appearances in each of the last five seasons and then finding their numbers for all of these categories. I then lined them up in Excel in a way where I could check standard deviations and correlations. I compared each year in the sample to the rest of the years in the sample to see if there were correlations and, if so, where those correlations lie.

The relationships will be represented by a correlation coefficient, a number between zero and one. The closer the number is to one, the stronger the correlation. The closer the number is to zero, the weaker the correlation. Anything below 0.5 can be considered a weak relationship, and anything below 0.3 should be looked at as basically no relationship. Here are the results:

Ground Ball Rate

Line Drive Rate

Fly Ball Rate

We find three major things here.

#1 GB% rate is the only one with a strong relationship.

You see all the values being above 0.67, showing that one year does quite a bit of work in predicting future years. You see much lower numbers with line-drive rate and fly-ball rate, and the line-drive rate seems to be pretty much wholly explained by randomness. This makes intuitive sense given that the angle range of a line drive is much smaller than the range for ground-balls and fly-balls.

#2 Consecutive full seasons have much stronger relationships.

You can see that most of the highest numbers there come between two years that are right next to each other. This also makes perfect sense just because of the way time works. Think about yourself. You are probably more like your 2021 than your 2017 self as you sit here reading this today. We should take last year much more seriously than 2019 or 2020.

#3 The 2020 season is not to be trusted

That 2020 season is a complicating factor in all analysis like the one we're doing here. Hitters played 60 games max (I don't include playoff data in these studies), which made that season much more ruled by randomness and variance. You see that exemplified by the weak relationships between 2020 and its adjacent years.

Takeaways and Data

Focus on ground-ball rate. This statistic is pretty steady year-to-year. There are exceptions (George Springer's line the last five years: 48%, 50%, 45%, 36%, 33%), but in general, it's safe to assume that a player's 2022 GB% will be pretty close to his mark for 2021, given you are looking at a full season's worth of at-bats.

I figured I'd share some data to finish this up. What you see below is a table of all hitters that have had 100+ PAs in each of the last five seasons with their ground-ball rates shown. The last column is standard deviation, which is a measure of the spread of a list of numbers. The higher that number is, the more variance there has been in the player's ground-ball rates and vice versa. The aforementioned Springer has the highest standard deviation, while Bryce Harper has the lowest as he has been insanely consistent in this category (41%, 41%, 40%, 40%, 42%).

Conclusion

GB% is pretty steady year-over-year, so you can feel safe in checking a player's 2021 GB% to gain insights about 2022. This isn't as true with FB%, and it's not true at all with LD%.
Home runs come from fly-balls (75%) and line drives (25%). If you are fishing for a home run hitter - focus on players with low ground-ball rates. The average GB% last year was 42%, with the lowest players being in the low thirties. Anything under 38% gives you a shot at bloated home run totals.
High rates of ground-balls can be good for batting average if a player is fast. If you are trying to bolster your fantasy team's batting average, finding a cheap, speedy ground-ball hitter is a good thing to try.
Don't trust 2020 data for anything, and don't worry much about what happened with a given player 3+ years ago. Unfortunately for this year, that pretty much means focusing solely on 2021 data since 2020 was so random with the 60-game schedule, and 2019 is now too far away to put a ton of trust in.

Download Our Free News & Alerts Mobile App

Like what you see? Download our updated fantasy baseball app for iPhone and Android with 24x7 player news, injury alerts, sleepers, prospects & more. All free!