I don’t remember why, but I was really curious about batted ball data earlier today, specifically regarding park factors, and I went on a long session of data collection. I don’t think I answered my original question (since I forgot it) but it led me to do a way-too-early look at the batted ball data from 2015 through today and if the Humidor was having any effect.
I collected the data into four buckets:
- Dbacks @ Home
- Dbacks Away
- MLB @ Chase
- The MLB as a whole
Since Statcast only goes back to 2015, that limited my date range from 2015 to 2018. For all four of those buckets, I compiled the following:
- Average flyball exit velocity
- Average flyball distance
- Average line drive exit velocity
- Average line drive distance
- Hard% on flyballs and line drives (combined)
This post is going to be primarily a data dump and a means to start some conversation. I’ll give my take at the end.
Some interesting quirks here. Notably, the drop in flyball exit velocity and distance for opponents at Chase Field whereas the Dbacks have maintained a very similar gap between home and away across the past three seasons. The other interesting thing I see here is that the MLB average flyball distance has been rather flat from 2017 to 2018 while Arizona has seen an improvement. However, I would chalk that up to the improvements we’ve seen out of Ketel Marte, Daniel Descalso, David Peralta, and AJ Pollock as well as the rise in John Ryan Murphy. I think May really shrouded the fact that we’ve had several players making offensive improvements this year.
Line Drive Data
Again... we see a small shrinkage for non-Arizona teams at Chase Field (e.g. the gap between “MLB @ Chase” and “Whole MLB” has shrunk) but still a very large noticeable gap between the Dbacks @ Home and the Dbacks away.
The distance chart is fascinating, though. In the era of increasing launch angles and increases in exit velocity (as seen in the first LD chart), the average distance for line drives has dropped over the last four years. I don’t have an answer for this - this doesn’t make sense per physics.
Now I’m confused. Statcast just showed us that the Diamondbacks have been hitting the ball considerably harder at Chase Field than anywhere else, yet the Hard% shows that the team has been hitting more “Hard%” on the road?
It’s also interesting to note that Chase Field (“MLB @ Chase”) has maintained an almost even gap between the league average (“Whole MLB”) over the past four years.
I think our data is still very noisy - dependent on the teams we’ve faced so far to date. But my personal take is that I don’t think the Humidor is having much of an effect, at least per the batted ball data.
I subscribe to the “juiced ball” theory from 2016 that led to the home run spike and the inevitable outcome was that we saw a “peak” for offense in 2017. Home runs are down considerably this year and offense, as a whole as dropped (last year the average MLB OPS was .750; this year it is currently at .722). I think this drop in offense might be mistaken for some effects of the Humidor.
Regarding the Humidor: what exactly is its purpose? The Humidor is designed to reduce offense; by lowering the overall velocity of the ball, the overall distance of line drives and fly balls would lessen. Lower distance = less power.
So keep that in mind when you look at the above charts: are we seeing a measureable decline in exit velocity or distance for balls at Chase Field? Personally, I’m not seeing much that’s convincing me, at least not yet. The gaps for exit velocity on fly balls and line drives have narrowed for Chase Field as a whole when compared to the MLB average; however, they haven’t shrunk for the Dbacks home versus away. And the overall average distances haven’t really changed, either. The hard% chart also supports this theory, though I have my questions for its calculations at the moment.
Please don’t take my analysis above as fact. I’m still in a “wait-and-see” period as I think the data is still noisy. I think we need more teams to come through and help balance the data out with more players. But there is one other thing to consider:
I think, so far, it would be a safe assumption that the current park factor for Chase Field should still be well over 100, seeing as the average exit velocities, distances, and hard% are all above league average at Chase Field. Maybe the results haven’t quite been there yet (thus why we see a 97 park factor at ESPN currently) but I think the batted ball data will prove out in the end.
But regarding the park factors, keep in mind that we’re mostly looking at April and May data. I did some Googling on Park Factors and came across this article that researched how park factors vary from month to month (note that they analyzed DraftKings +/- instead of Park Factors, though it appears to be a similar system). They observed that Chase Field was very positive in April, May, and September, but was much closer to neutral in June, July, and August, albeit still slightly positive.
So, if you’re interested, please take a look at the data and come up with your own conclusions. I think there is a lot more to this subject than what I posted here and I am very open to other ideas, opinions, and methodologies. I still want to see a full season’s worth of data before I draw my final conclusions, though I have my suspiscions as I noted above. Let’s discuss this further in the comments!