Effect of Seam Height on Wake of MLB Baseballs: Post 35

The huge home run rate in 2019 has led to a lot of questions from players and fans. According to the report published in 2017 by the MLB Home Run Committee (which you should read), the drag of baseballs decreased some from 2015 – 2016 – 2017 and the home run rate increased by an amount commensurate with that change. After the report, the home run rate decreased somewhat in 2018 and then increased significantly in 2019.

I do not believe that there have been a series of discrete changes in the MLB ball. The hand-made ball has always varied a lot. I don’t believe any of this is on purpose and question how anyone at MLB or Rawlings could have known how to design a ball to produce more home runs. I also do not believe that the diameter of the stitches, the smoothness of the leather nor the roundness of the ball has ever had anything to do with baseball drag or home run production. These notions do not fit what we know about aerodynamics and have never been supported by any evidence. The non-centered pill theory sounded interesting to me for about 5 minutes, but also does not carry any water.

This post will attempt to address how seam height affects drag. Most lay-persons I talk to are pretty satisfied to assume larger seams mean larger drag, but as a fluid dynamicist, I was not convinced. A smooth ball has a higher drag than a baseball, so up to some point, seams reduce drag. As you will see below, most of the time during a ball’s flight, the seam height has no impact on drag.

Some have pointed out that simultaneous to any change in the average MLB ball, MLB hitters are in the midst of a fundamental change in approach, and that this change also impacts the number of home runs.

From where I am sitting, I see a lot of evidence of each of these trends. So let me state here that the point of this post is not to weigh in on which of these effects is most important, or to even discuss changes in the ball. This post will show how the behavior of two balls with 2 different seam heights is different.

As described in my Post 32, one crate of MLB baseballs (6 dozen balls, thanks to Mike Fast and the Atlanta Braves) produced a variation of seam height from 22/1000″ to 40/1000″. To me, this seems like a surprisingly large range. The average seam height was 31/1000″. From this batch, we chose the ball with the smallest seam height and tested it against a ball provided to us by Meredith Wills (@Bbl_Astrophyscs) that was from 2009 and had a somewhat larger seam height of 45/1000″.

Both balls were un-scuffed, our ball was as out of the box, while the game-used 2009 ball was rubbed with mud. We have shown previously in Post 25 that mud does not significantly change the roughness of the leather, but we have some anecdotal evidence that it may cause the seams to raise higher. And that should not matter, because we are comparing balls with known seam heights in their current condition. I make no claims that the 2009 ball is typical of that year. A single sample provides us no information on this.

[If you are new to our measurements, you may take a minute to read here about vorticity (the colors in our plots), boundary layer separation and wakes]

Results: These data were acquired by USU undergraduate student and noted Cardinal fan, John Garrett (with assistance from MS Student and unfortunate Mariners fan Andrew Smith, no relation). Below, we show the lower seam ball on the left and the higher seam ball on the right at 4 different angles, all in a 4-seam configuration and near the point where there is a seam directly on the top and on the bottom. The ball is moving to the left at 90 mph and is not rotating. The successive sets of images at different angles simulate a ball is rotating like a fastball, or more relevantly, a backspin line drive.

The datasets are described by the angular position of the top seam. If it is on top, this is 0 degrees. If it is rotated counter clockwise from that position, that is a negative angle.

In the first set of images below, the low-seam ball is on the left and the high-seam ball is on the right. The balls are oriented similarly. The flow over both balls is similar, with the separation on top happening on the seam and on the bottom somewhat past the seam. The wake is the disturbed colored air behind the ball. .

Set 1: Seams are -14 degrees from vertical.

In the second set, the separation occurs on the seams on top and bottom of the both balls resulting in a large wake and thus large drag for both cases.

Set 2: Seams are -7 degrees from vertical.

Now, for the third set, the seam on the top of the ball on the left is causing boundary layer transition, a very late separation on the back of the ball, and low drag. But, on the right, the higher seam in the same position is causing boundary layer separation. I have added a blue and red lines to emphasize how small the wake is for the lower seamed ball compared to the higher seamed ball. In both cases, the seam on the bottom causes separation. The resultant wake of the higher seam ball is therefore much larger and the drag is therefore larger too.

Set 3: Seams are -2.7 degrees from vertical.

The situation in the fourth set is similar with the lower seam having a much smaller wake.

Set 4: Seams are 2.5 degrees from vertical.

With a small additional rotation (not shown), both balls now act similarly. They will act similarly until the seam which is now at the rear of the ball reaches the bottom of the ball (while the seam currently on the front reaches the top).

So, most of the time, the flow pattern, and thus the drag, of these two balls is similar. In some common but narrow and brief configurations, the drag on the higher seam ball is much larger. It is not hard to imagine that the net result is a somewhat larger drag for the higher seam ball.

I’d also like to note that we say similar behavior with a very high-seam old-style NCAA ball. It was similar to the MLB ball except when seams where near the top or bottom of the ball.

Now, batted balls do not necessarily travel in a 4-seam configuration. And that is not important to this result. Any time the ball has a seam on the extreme edge of the ball, the larger seam can lead to a massive separated flow and a big wake. Note that all my talk of “top” and “bottom” also applies equally to the sides that we are not seeing in our data sets.

My hypothesis is this: The bigger the seam, the wider range of orientations that lead to this large wake and higher drag. Even for big seams, I believe that most of the time the wake is similar to a lower seamed ball, but that the percentage of time with a large wake goes up with seam height.

I’ll also note that it is very easy to see and feel the differences in these two balls. If you are an MLB pitcher in the playoffs, and a home run will kill you right now, better get a high-seam ball, Sparky.

Related Post