waDS 101: Measuring Minor League Batters’ Performance Growth
It’s not surprising that it can be difficult to measure how minor league batters grow and develop from season to season. A wide variety of park factors, an (allegedly) ever-changing baseball, and other variables make it tricky to measure growth even at the major league level. Now, consider the expanse of those factors in the minors and it’s apparent that the challenge is even greater.
There isn’t really a (publicly available) way to measure a minor leaguer’s growth across multiple seasons other than comparing their stats from one year to the next. Comparing numbers from rookie ball through AAA is inherently flawed due to the many confounding variables I mentioned earlier, and because the players working their way through the minor leagues are constantly adjusting, changing, and growing.
When I initially started playing around with seeing which minor leaguers improved the most year to year, I ended up creating a very bad and very complicated (and much worse) version of wRC+. However, after exporting 52 .csv files from Fangraphs and several pivot tables, I birthed my second child1: weighted-adjusted development score – or waDS for short.
Feel free to spell it out and call it W-A-D-S, or pronounce it so it rhymes with “pods”. Despite the pretentious name and fancy acronym waDS is relatively simple: it gives each player a number that shows how much they improved or regressed in their offensive output from season to season while accounting for different leagues AND levels across all of minor league baseball.
Components
The statistics that are factored into a player’s waDS number are:
- Pitches per plate appearance (P/PA)
- Walk rate (BB%)
- Strikeout rate (K%)
- Swinging strike rate (SwStr%)
- On-base percentage (OBP)
- Isolated power (ISO)
- Weighted on-base average (wOBA)
Each of these stats are weighted differently according to a player’s league and level within the minors and then combined and adjusted to create a single number. For example, a player improving their wOBA moving from the Cubs low A affiliate to the Red Sox AA affiliate has their number adjusted and weighted differently than a player in the Pirates system who also improved their wOBA while played in high A in back to back seasons.
The higher the number, the better, but it’s best to view a number’s corresponding percentile rank because it provides more context. For instance, Guardian’s prospect Bo Naylor has a waDS of 541.42. Is that good? Yes, but you wouldn’t know that without his percentile rank; Naylor’s waDS is in the 98th percentile. What does this even mean? It means that Naylor saw an elite amount of growth (better than 98% of all other qualified minor leaguers) in his offensive production from 2021 to 2022.
A few other things to note:
- Sample sizes used are a minimum of 150 PA, and players must have at least two individual samples from the last two seasons
- Age is not factored in to any of my equations because I don’t want to penalize a player for being a late bloomer (it wouldn’t be very growth mindset of me to do that), BUT some old dogs really don’t learn new tricks so only players under 30 years-old get waDS
- All data comes from Fangraphs, and all calulations were performed in collaboration by me and Google Sheets.
Benefits
- waDs allows us to see which players truly improved their offensive output while accounting for varying run environments
- Players who are already at the top still get rated highly for being able to maintain their high level of offensive production thanks to a separate player score component I added to the waDS leaderboard
- Stats are included and weighted from the past two seasons, so a player who is normally good at getting on base but saw their production dip in that area as a result of a different run environment or random variance would not be unfairly punished
Limitations
- Not all players have the appropriate samples and sizes to qualify, so not every minor leaguer is ranked. However, stats from all minor leaguers are accounted for when determining league averages and weighting
- Age is not heavily factored in to a player’s output at different levels, but the weighted averages of each league include mean and median ages so they’re accounted for in a way that doesn’t allow for a 28 year-old playing in high A to have a better number than everyone else
- Players who had qualifying samples accross multiple levels in one season have their largest single sample compared with their largest single sample from the other year
I’ll continue to update and adjust things as I learn more and eventually adapt to include data from 2023, so follow my Twitter to stay up-to-date. Use the link below to access the waDS Leaderboard!
1wEZR, a cool pitching stat, is my first born – check my Twitter for info on that. But there’s also a leaderboard here if you’re ready to jump in.
