The Rays Through Hit f/x (sort of)
Who, in all of major league baseball this year, had the hardest hit home run?
If you're like me, you probably thought of the shot Willy Mo Pena, he of batting practice legend, hit off James Shields, he of giving-up-home-runs-at-a-rate-nearly-sufficient-to-make-rational-men-abandon-DIPS legend.
But at 117.5 mph off the bat, that was only the ninth hardest hit home run this year. No, the absolute most viciously struck ball of 2011 was by one of the few heroes of yesterday's game, our very own Sean Rodriguez (118.4 mph!).
I once wrote that Roddy has a terrible batting eye, but Joe Maddon steadfastly maintains that he has immense offensive upside, on account of some very serious power. Seems like Maddon knows what he's talking about, at least concerning the power. I don't, of course, have access to hit f/x data, but ESPN Home Run Tracker (formerly Hit Tracker) is a fun tool. After the break, I'll try to make it do something for which it wasn't designed.
To my thinking, the important number in identifying power is speed off the bat. The distance of a home run depends on the angle at which it's hit, the conditions at the park, and a host of other reasons. But over the course of a season, most major league players should make solid hard contact at least a few times, and that just might be enough to figure out their raw power. To establish that there is in fact a relationship between how many home runs a player hits, and the average speed off the bat of their home runs, I divided all hitters into bins by their home run total in each year. (Note: I wouldn't expect the relationship to be very strong—there really is a difference between raw power and game power (poor S-Rod).)
I'm really not sure what was going on in 2008—maybe the massive tectonic shift in the baseball world as the Rays ascended to greatness threw everything off—but for the other years, the relationship seems pretty clear. If you're able to hit the ball harder when you catch it solidly a few times, you hit more home runs overall. Here are the Rays in 2011:
|
Row Labels |
HR Total |
Max of Speed off Bat |
Average of Speed off Bat |
|
Brignac, Reid |
1 |
96.9 |
96.9 |
|
Fuld, Sam |
3 |
101.8 |
99.2 |
|
Guyer, Brandon |
2 |
104 |
99.5 |
|
Chirinos, Robinson |
1 |
99.8 |
99.8 |
|
Kotchman, Casey |
10 |
106.3 |
100.3 |
|
Johnson, Dan |
1 |
100.4 |
100.4 |
|
Johnson, Elliot |
4 |
103.7 |
101.5 |
|
Zobrist, Ben |
16 |
109.1 |
102.2 |
|
Damon, Johnny |
14 |
109.4 |
102.9 |
|
Joyce, Matt |
18 |
109.7 |
102.9 |
|
Longoria, Evan |
28 |
108.2 |
103.0 |
|
Ruggiano, Justin |
4 |
112 |
103.0 |
|
Upton, B.J. |
21 |
112.8 |
103.4 |
|
Shoppach, Kelly |
8 |
109.2 |
103.6 |
|
Jennings, Desmond |
9 |
106.8 |
104.0 |
|
Jaso, John |
5 |
108.6 |
104.1 |
|
Lopez, Felipe |
2 |
105.9 |
104.9 |
|
Rodriguez, Sean |
7 |
118.4 |
107.1 |
What do I see in this table? Not much.
- John Jaso has a little bit more thump in his bat than I would have thought.
- While Desmond Jenning's has shown more power than most of us expected in his limited time up, it's not as if he's gotten his home runs on cheap swings.
- When B.J. Upton and Kelly Shoppach run into one, they hit it hard.
There's a lot of noise, though, along with a pretty obvious selection bias (hit tracker noted all long fly balls in 2007, but hasn't since). We know that Longoria has very good power, and that he leads the team in home runs by a fair margin, but his average speed off the bat is middle of the pack. Looking at his home runs individually, there are four with speeds off the bat under 100 mph bringing his average way down. Three of them came in Fenway, one in Yankee Stadium, both of which are pretty strong hitters' parks. Obviously there's a pretty large park effect that interacts with the selection bias. It will need to be adjusted for. Still, I think there's some information that can be found here if one is clever about how they slice the data. I'm including my spreadsheet of raw data and pivot tables from 2008-2011, if anyone else wants to take a stab at making something useful. Hit Tracker Data
18 comments
|
4 recs |
Do you like this story?
Comments
Awesome stuff, really cool Whelk.
First question off the top of my head: how does Hit Tracker calculate that initial velocity off the bat? Because it seems like such a shame that we only have this data for homers.
I love Casey Fossum. Now try and take me seriously. -- @steveslow
by Steve Slowinski on Sep 22, 2011 10:29 AM EDT reply actions
It considers where the ball landed, time in the air, and weather/atmospheric conditions.
It’s pretty slick. Complete explanation here.
by Whelk on Sep 22, 2011 10:42 AM EDT via mobile up reply actions
Yeah, I saw that.
But it doesn’t look to me like they specify how the speed off the bat is calculated. Is it calculated from other inputs, or is that something they’re independently measuring?
I’m assuming it’s the velocity is reverse engineered from other data points, but that makes me curious: what sort of inputs do you need to calculate it? Time in the air and flight angle? If it’s as “simple” as that to get an estimate of speed off the bat, why don’t we do it for more types of hits?
I love Casey Fossum. Now try and take me seriously. -- @steveslow
by Steve Slowinski on Sep 22, 2011 11:04 AM EDT up reply actions
Yeah, the only inputs are startpoint, endpoint, and weather.
The rest is calculated. Maybe it’s more complex than I give it credit for, but sure seems like it could be done for more balls. They did in 2007. My guess is the real buyers have hit fx so there’s just no demand.
by Whelk on Sep 22, 2011 11:26 AM EDT via mobile up reply actions
Oh, and time in the air.
I don’t know how to factor in weather, but weather neutral can be done with just high school physics.
by Whelk on Sep 22, 2011 11:29 AM EDT via mobile up reply actions
That was my thought too.
We could calculate time in the air and distance traveled easily enough for most balls…it’d just take a little bit of work and some MLB.tv reviewing. And if we limit matters to just the Trop, we wouldn’t have to worry about the weather. In other words, I’m curious if we could give it a try for some players.
If we would need to have angle off the bat as an input, that’d be a kicker….that’s a tough one to estimate based on TV.
I love Casey Fossum. Now try and take me seriously. -- @steveslow
by Steve Slowinski on Sep 22, 2011 11:37 AM EDT up reply actions
Hot Dome!
I hadn’t gotten my mind past the “in theory” part. Absolutely, let’s try to figure it out. I was just thinking about it in terms of no air, but when I get back to my desk, I’ll try to come up with the calculation for 72 degree, sea level, calm trop air.
by Whelk on Sep 22, 2011 12:42 PM EDT via mobile up reply actions
I thought of this when I woke up this morning.
If what I’m trying to measure is raw power, I really just want the best hit balls, and I mostly care about the measurements in relation to other players. The problem that we see in the Longoria example is that freak winds and small parks will include less well struck balls for some players, bringing down those players’ averages.
But hit tracker let’s you know how many ballparks a home run would have made it out in. If I limited the sample to hits that would have made it out in all 30 parks, that might zero in on “best hit balls,” at least for the people with enough power to get some type of sample, and clear up some of the noise.
by Whelk on Sep 22, 2011 10:52 AM EDT via mobile reply actions
I like the info and it's something we haven't really seen here before. Thanks for writing.
One suggestion I have would be to use only one decimal point in the Average column to clean up your presentation.
No, don't think that affects it
But speed off the bat isn’t being measured. It’s being calculated from the observed flight if the ball.
by Whelk on Sep 22, 2011 12:44 PM EDT via mobile up reply actions
Beautiful stuff. Love it.
A DRaysBay and FanGraphs writer from Cubs Stats and Twitter @BradleyWoodrum
What
The caller said the boy, after removing the bulb from its socket, left the building and threw the bulb on the ground. When the bulb broke, the caller said the boy screamed "not since they nerfed sunfire capes"
If you love Jesus Christ and are 100% proud of it, copy this and make it your signature!
by Top Gun Numba 1 on Sep 22, 2011 4:57 PM EDT up reply actions
How about looking at a player's avg speed off bat, generating an 'expected' # of HRs,
then comparing it to actual. May give a measure of luck and/or skill at turning hits into HRs. You would need to find the correlation b/t speed off bat (say, mph greater than 95) and # of HRs (or maybe a rate stat, like HR/AB).
Kila's slash for Apr 20 to May 4, 2011, right before he was sent down: .276 / .344 / .448
by SagehenMacGyver47 on Sep 22, 2011 7:59 PM EDT reply actions
That's of course the goal.
But from hit tracker, we only have the speed off the bat of actual home runs, so we really don’t know how hard hit all those non-HR fly balls are.

by 


























