The Declining Rate of U.S.-Born Men's DI Golfers

A Data Deep-Dive

Recently, while reading a book on genes, I decided to pull some data on the height of golfers. About an hour later, I had accidentally designed a scraper of online men’s golf rosters. The scraper pulled all available parameters per player from the online roster-pages of DI golf teams.

After a somewhat extensive cleaning-process of the data, I found out I had an incredibly extensive dataset of collegiate golfers.

I pulled roster data directly from team websites, season by season, all the way back to 2005–06. The scrape produced 40,791 player–year rows, of which 9,310 are freshmen entries. About 86% of those include usable “From” information. This is enough to track shifts in where players are coming from, though it’s not a complete census. Coverage varies (some schools and years are missing, and several big programs like USC, UCLA, and Arizona State are missing from my dataset), but the data is still rich enough to reveal trends.
You can read more about the Data on the “Read Me” page of the report (linked at the bottom of this article).

Before we get any further - I’d love for you to Subscribe to this Newsletter. It’s free - and it would really help me out.
I have several posts lined up using the same dataset where we look at some other fascinating trends in college golf.
When you’re subscribed, you will receive these upcoming posts directly to your inbox for free. Thank you!

The Most Telling Graph

The graph above depicts U.S.-Born Freshmen on Men’s DI Rosters (from the dataset) as a percentage of all freshmen in the dataset for that particular year.
From 2005 through 2011, U.S. Born players accounted for more than 85% of the freshman class every single year.
Fast forward to the 2020s, and the U.S. Born rates have dropped to below 75%. This is roughly one less American on average per DI-team in the span of 15 years.

Here is a table of USA Count (U.S. Born Freshmen), Total Freshmen Count, and USA Share per year dating back to 2005.

Total Number of Roster Spots

So how does the declining rate effect total roster spots awarded to U.S. Born DI Golfers?
Well, first of all, according to the NCAA Participation Report, there has been growth in the number of DI Men’s Golf Teams in this period:

The only significant decrease in number of schools sponsoring DI Men’s Golf during this period was the COVID-year. In 2023, there were 302 DI teams.

You’d think this would lead to an overall increase in the amount of DI golfers, but roster sizes have decreased in this same time frame. Here are the averages from the scraped data I have. These averages line up closely to those reported in the NCAA report:

We see a steady decline with the COVID-year as an anomaly - followed by a lagging effect due to all the COVID-redshirts.

When it comes to number of DI athletes, these two drivers (more teams, smaller rosters) end up in basically a wash. The number of athletes have remained steady around 3,000 players for the entire period (Slight recent decline). See the column below - NCAA Report # of total Men’s Golfers in DI.

The drivers for the decrease in roster sizes can be several things. I don’t have data on this, but my hunch is that title IX considerations is a large contributor. Athletic departments are charged with having a representative student-athlete body of that to the entire student body at the school.

In order to avoid skewing the athlete count vs student-body, capping the men’s golf roster to a set maximum has been practiced at schools I’ve coached at.

Female enrollment to U.S. Universities have been higher than Men for an extended time. According to NCES’s Undergraduate Enrollment indicator, the 2021 enrollment, women were 58% of undergrads (8.9M) vs. men 42% (6.5M). I imagine a lot of athletic departments work towards having more female than male athletes in total.

Data Coverage

As a result of my scrape and cleaning, when I compare the dataset to the NCAA Participation report, I have roughly 60% “From” coverage in my dataset from 2010 till 2025. That means, I have roughly 1,800 of the 3,000 athletes every year identified with data about where they are from. Here’s a table with the coverage by year:

It’s not perfect, but a strong indication.

In order to find out how many fewer roster spots U.S. Born players are awarded, let’s assume the “From” data I have is representative for the “From” data I lack. This is not an unreasonable assumption as I have decent coverage (60%), the gaps in the data are spread across Power-4 and non-Power-4 schools, and I see other parameters line up to the NCAA report (like roster sizes).

When we multiply the U.S. Born vs. Internationally born share by year with the total athlete body (per the NCAA), we get this table:

This is an estimation. However, the numbers are quite striking. There are about 500-600 less U.S. Born DI Golfers since COVID compared to the the early 2000s.

With roughly 300 teams, that means that each team is carrying more than 2 fewer U.S. Born players than they used to.

(Note: I earlier published a visual representation of this logic with an error that overstated the International share. This was noted by one of the readers (Thanks Anthony!) The tables pasted above now accurately reflect the logic of estimating U.S. Born and International players by year.)

I knew we would see a change, but not quite of this magnitude.

I don’t see this trend slowing down right away. First of all, roster sizes will continue to decrease with the pending changes. Further, a lot of parts of the world remained relatively unexplored as far as recruiting. We’ll cover that in future posts.

Here’s the link to the report. More slides will be added to the report over the next few weeks in conjunction with posts on this Substack.
If you use it, please subscribe to this Substack and share with a few friends and/or on social media. Thank you!