Does Getting on Base Really Matter?
The other day, I was watching Moneyball, one of my favorite films. My favorite scene is when Billy Beane is in the room with a bunch of scouts saying the team should sign a bunch of questionable players. His response, as I'm sure most of you know, to all the criticisms he receives is "He gets on base." This got me thinking, how important is it to get on base?
Using Fangraphs, I collected a bunch of data from last season, focusing on the team as a whole rather than individual stats. My goal was to see which aspects of hitting (OBP, OPS, wOBA, etc) had the largest impact on the number of runs the team scored. Some of the stuff I found seemed obvious, other stuff was quite surprising.
For each statistic I ran a linear regression between total runs(y) and that statistic(x). I have reported the slope and correlation coefficient for each one:
- AVG: slope=4.76(r=0.81)
- OBP: slope=4.96 (r=0.89)
- SLG: slope=2.91(r=0.90)
- OPS: slope=2.04(r=0.94)
- wOBA: slope=5.00(r=0.94)
- BB%: slope=4.18(r=0.48)
- ISO: slope=2.87(r=0.60)
- HR: slope=1.47(r=0.53)
- SB: slope=-0.08(r=-0.03)
It comes as no surprise that wOBA has both the strongest correlation and the most runs per percentage point of all the statistics I looked at. If anything, all this means is that wOBA truely is a great way to measure offensive value. In addition, the fact that OBP has a slightly larger impact than AVG indicates that getting on base via BB, HBP, Error, FC, etc. does create more run-scoring opportunities. However, the correlation between BB% and total runs was quite weak. Now that I think about it, this may be because a team can walk less and hit more and still recover their OBP, so this is really no different than AVG having a lower r-value than OBP. I'm surprised that SLG and OPS have a smaller impact than AVG and OBP, but I think this just exposes the weakness of those stats compared to wOBA. The fact that ISO has such a weak correlation with total runs further emphasizes that hitting for extra bases does not tell you as much as simply getting on base. I figured ahead of time that using HR and SB would provide weak correlations since they are counting stats, but it was interesting to see there is no correlation whatsoever between stolen bases and total runs scored.
I guess I didn't really learn all that much from doing this, but I can say the simple act of getting on base is more important to scoring runs than the business of how one gets on base (OBP vs AVG or BB%), and that wOBA is an awesome and useful statistic. So I guess Billy Beane was right, if a player gets on base, he's worth having on your team.
I'm new to Twins Daily and this is my first blog post! I love numbers but I also love the other aspects of the game, so as I continue to publish I promise not everything will be quite this dry
Thanks for reading and feel free to share your own thoughts and analysis!