Measuring Player's "In the Park Power"
I began the idea for this post wondering if we can measure how hard balls are hit by guys with “under the radar” type power. That is, the hitters that hit the ball hard a lot but don’t have the awesome home run totals to show for it. For a good read, and very relevant to this post’s idea, I suggest checking out Ari Berkowitz’s Beyond the Box Score post asking “Can We Quantify Hitting the Ball Hard?”, in which he uses batted ball data to attempt to answer that question (since we don’t have easy to access hit f/x data that actually measures just that).
Initially, I attempted to figure a player's in the park power by calculating their slugging minus home runs and the at bats in which home runs were hit, but this produced too many players like Ben Revere posting high numbers, and I knew there wasn't any power in that bat. I needed to take the singles hitters out of the equation.
Most of you are probably familiar with the metric for hitters called ISO. For those that aren’t, ISO, or isolated power, is simply a player’s slugging percentage minus their average. This leaves you with a percentage showing only a batter’s extra base power.
|Hint: Paul Goldschmidt is really good at baseball.|
Most power hitters produce a high ISO with their ability to put the ball into the seats, and for good reason; if you can do that and do it often, you obviously have power and, at least somewhat frequently, hit the ball hard. But what about guys that consistently hit the ball hard, but only into the outfield for doubles, triples, and in the park home runs? Clearly they have some power too, albeit in a less noticeable way. Just for fun, I came up with my own simplistic metric to measure this. It’s called “IPISO”, or “in the park isolated power”, and it’s simply a player’s ISO without his home runs and the at bats in which he hit those home runs.
Looking at players from this season with at least 400 plate appearances, here are the top 10 hitters with the most “in the park power”. I’ve also listed the percentage at which they hit the ball for a line drive or for a fly ball into the outfield, and also their UBR (all stats per fangraphs):
|Player||IPISO||LD% (20.8 AVG||OFFB% (24.8 Average)||UBR (-0.1 Average)|
Some of these guys are unsurprising, like Goldschmidt, Trout, and Puig, but others weren't the prototypical "amazing athlete" types that I was expecting to see (no offense to guys like Lucroy, but I never would have guessed him to hit 54 doubles this year). Most of these hitters sport exceptional speed and/or made good base running decisions in 2014. But after the top ten things started getting a little more surprising.
|Player||IPISO||LD% (20.8 AVG)||OFFB% (24.8 AVG)||UBR (-0.1 AVG)|
For as down a year as Alex Rios had, he still managed to hit doubles and triples at the 14th most frequent rate out of all MLB (min 400 PA). Nori Aoki is another that jumps off the page, particularly because his line drive rate is just 0.2% above the ’14 MLB average, his OFFB rate is 13.2% below average, and his UBR was the worst in all of baseball by a landslide. I’d say it’s the BABIP gods at work, but Aoki only posted a .313 mark in that category compared to a .285 BA. Let's just say that if you were playing in some bizarro fantasy league that didn't use home runs, this would be your dream lineup. Or maybe if you were building a roster that played its home games at Petco Park?
How Do Guys with the Highest IPISO Produce Their Results?
I began to wonder which statistic best correlated with IPISO; was it the better base runners that stretch singles into doubles, and doubles into triples, that produced a higher mark? Or was it the guys hitting more fly balls into the outfield, or line drives, which were able to drive their IPISO up?
The simple answer based on the hitters with the top 25 IPISO is that there is no simple answer. Here is a look at some linear regression that my awesome girlfriend, despite being annoyed to death with statistics, helped me run:
|R squared values- IPISO & LD: .0667, IPISO & OFFB: .0073, IPISO & UBR: .1071|
From this set of data, and with R being as low as it is for each model, it appears that there is no direct correlation between hitting the ball in the park for extra bases and LD, OFFB, or UBR (although the latter seems to at least be the closest related, however slightly). It's important to note that we're looking at an incredibly small sample size, as this is just of the top 25 players per our "new" little metric, but the take away from this may be that there is no hard and fast rule to producing a good IPISO. The top performers are all known to have different skill sets, some more speed, others power, and still others solid contact. It may also be that creating extra base hits is a product of a combination of multiple skills.
Next week we'll look at a larger body of data from players across the league, and see if we can't find some correlation between IPISO, batted ball data, and UBR. We'll also look at whether there is a certain skill combination that produces higher IPISO more frequently than others.
Photo courtesy Not That Bob James per Flickr creative commons (Goldschmidt)