MLB Pitch Speed Tracking

InstaFace

The Ultimate One
SoSH Member
Sep 27, 2016
21,754
Pittsburgh, PA
An article I was browsing through this morning had this italicized bit that caught my eye:

For Lindor, the StatCast guys report that he hit just one home run last season off a pitch as fast as the 94.1 mph fastball from Dyson (Note: MLB has changed its tracking system this year, and pitch velocities are going to be an estimated 0.7 mph or so faster).
...and I almost spilled my coffee. wait, WHAT?! I don't remember hearing anything about this. "oh, whoops, our system was FUBAR last year, all your assumptions about your pitchers losing velocity? yeah those were BS". Likewise, "trust us, we've got it right this time!"

Can any of our PitchFX mavens here clarify this? Can we trust our velocity data? Help me, @Jnai , you're our only hope.
 

absintheofmalaise

too many flowers
Dope
SoSH Member
Mar 16, 2005
23,328
The gran facenda
Fangraphs had an article on this recently.

This year, however, with Statcast officially replacing PITCHF/x in the big leagues, the data being pulled publicly is now from the Trackman radar. While PITCHF/x velocity numbers were reported at a defined point along each pitch’s trajectory —usually at the 55 foot mark — so that velocity didn’t have to be calculated at every x/y point along the pitches path, Statcast outputs the highest velocity that Trackman records along the flight of the pitch, which due to physics, is going to be immediately after the ball leaves the pitcher’s hand.
 

charlieoscar

Member
Sep 28, 2014
1,339
Re: Fangraphs article blurb posted by absintheofmalaise: While PITCHF/x velocity numbers were reported at a defined point along each pitch’s trajectory —usually at the 55 foot mark — so that velocity didn’t have to be calculated at every x/y point along the pitches path, Statcast outputs the highest velocity that Trackman records along the flight of the pitch, which due to physics, is going to be immediately after the ball leaves the pitcher’s hand.

Now you need to know the point at which each each pitcher releases the ball (stride/length of arm). A ball leaving a pitcher's hand at 95-mph will reach home plate sooner if the distance is 54 feet than if it is 55 feet.
 

Jnai

is not worried about sex with goats
SoSH Member
Sep 15, 2007
16,123
<null>
Well, someone necro'd this thread, so I'll chime in. Sorry, I missed this a week or so ago.

The gameday standard was *50* feet, not *55* feet, as incorrectly noted by Dave in that article. 50 feet was chosen because it matched the speeds reported by scout radar guns when they were testing the system. Most places on the internet reported 50 feet.

My website, BrooksBaseball.net, has always used 55 feet, by extrapolating the 9p fit that is given for model pitch trajectory. It continues to use 55 feet.

The MLBAM data feed now contains two separate but related pieces of information:
1) A release speed, which is the speed at some arbitrary point that the system detects is very close to the pitcher release and
2) A 9p fit with y=50ft.

These are both perfectly related to each other, as we describe here:
http://www.baseballprospectus.com/article.php?articleid=31529

So, the velocity data from this year - at least on my website - are directly comparable to the number from previous years, at least for this particular change. Besides the fact that the website is unaffected, I will note that the average release point is very close to 55' across pitchers (of course individual pitchers have individual release points).

Please note that there are still lots of other changes that happened, which have created new exciting movement artifacts and other sorts of junk that take our very conservative algorithms time to identify and correct.

It's also worth noting that we continue to hand review pitch classifications. The MLBAM classifier has been maybe not the best at certain points this year.

One ongoing issue has been with automated game scraping. My parser has been tossing a lot of data as "bad" because it's missing an sv_id, which used to be the unique identifier for pitches, but that identifier is now being overwritten (and sometimes, discarded completely) when they reprocess games. Who knows. We're trying to get it sorted out.
 
Last edited:

CarolinaBeerGuy

Don't know him from Adam
SoSH Member
Mar 14, 2006
9,419
Kernersville, NC
Sorry to hijack a bit, but for a few years my brother has been repeating a line about Nolan Ryan being proven to be the hardest thrower ever because of the way velocity was measured in the past versus how it's currently measured. Any truth to this? I think it's BS, but he swears it's true.