wOBA modification pt. 2
After a little bit of work the last few days, I’ve got my modified wOBA stat assigned to a player in a database table. Given I had all of my variables, constants, and knew the equation to use, setting it up was pretty easy. In the end, here’s an example what my table looks like:
Next up I really wanted to compare what I had vs what FanGraphs had listed. I already had some tables set up from my previous little project in learning fWAR, one of which included a players wOBA. Joining the two tables was easy and I was able to get the top 10 gainers as well as the top 10 losers using this new method. Bare in mind: this is for the 2020 season only right now.
Some pretty big differences in there. Saw a .080 swing in wOBA from the biggest gainer and biggest loser. That’s the difference between “awful” and “great” and, when I started this, anticipated there would be a big gap.
So in all, I had 294 players here that I could match up. 60 players lots wOBA points, four had no change, and 230 gained. It looked like the average difference came out to be +0.013. Again, nothing that I didn’t already expect.
Now my next thoughts were to apply this to WAR and see what would happen. What I found here actually surprised me a bit more.
I had fewer players I could run this against. I believe this to be due to the defensive data that I had gathered previously. This go around it was 227 players. Here’s my top 10 and bottom 10 gainers here as well:
I knew there would be a difference. There really was no doubt in my mind in this. There had to — wOBA is a focal point in Batting Runs. What I wasn’t expecting was some to gain nearly a full win. So I decided to check out what the average net was for Batting Runs. The answer to that is 2.1 batting runs on average was gained.
0.013 wOBA = 2.1 Batting Runs. And in turn, 2.1 Batting Runs = 0.2 fWAR
Or..
For every 0.0065 wOBA points, a player gains 0.1 fWAR. At least for the year 2020. I suspect that won’t vary much year-by-year, though I will need to get more data added to find out for sure!
And with that — I hope to have my next update include that! From there, I think it would be fun to create a graph similar to my fWAR graph that shows the difference between them.
Thanks for reading!
