A baseball loving physicist looks at the knuckleball | Hacker News Reader

A baseball loving physicist looks at the knuckleball

baseballprospectus.com

16 pointsmkovach18 comments

I don’t know how to reconcile that with the video evidence:

https://images.app.goo.gl/CtJmayDTwVrz9xwi6

From the article:

> PITCHf/x is a video-based tracking system that is permanently installed in every MLB stadium and has been used since the start of the 2007 season to track every pitch in every MLB game. The system consists of two 60 Hz cameras mounted high above the playing field, with fields of view that cover most of the region between the pitching rubber and home plate

> Depending on the specifics of each installation, the pitch is typically tracked in the approximate range y=5-50 ft, resulting in about 20 images per camera for each pitch

So the raw data is from two 60 Hz cameras build some time before 2007, each taking 20 pictures of a small, fast moving object tens/hundreds of feet away? The error bars on that raw data must be significant, especially when a "knuckleball wobble amplitude" of just a few inches would already be very effective.

If I'd design that PITCHf/x system today, I'd add a handful of 80 GHz radar modules, and cameras on two more axis.

Anything similar for Basketball?

I vaguely recall a company with a rfid like ball and player tracking system that was seeing some adoption by colleges. Maybe someone knows more?

I know what you are talking about. I think they went bankrupt. I looked at that and I think there are few Chinese copies on Amazon.

I am looking into something with higher tracking requirements. Maybe this is the equivalent: https://www.catapult.com/sports/basketball

Yeah, based on the description of the system I’m skeptical that the “raw” data is all that good. The author makes some assertions about it, but he’s also working with a vanishingly small sample size. He makes no attempt to compare specific knuckleball pitches with apparent wild behavior to the captured data. He just pulls data from four games of two knuckleball pitchers.

Which is weird because the gif above does seem to follow the pattern the author himself mentioned about a very-slow-spin pitch being influenced by wind against the ball’s seams in different ways as it rotates.

Article is from 2012. MLB has since adopted a doppler radar system called TRACKMAN.

...which was replaced a few years ago, now using optical tracking from Hawk-Eye.

The maker of the camera system claims 0.5" precision; TFA calculates a mean 0.3" RMS deviation from an aerodynamic model.

Good news, the author of this study has addressed that pitch, using higher quality radar data. https://baseball.physics.illinois.edu/DickeyPitch103a.html

Verdict: It is indeed nasty stuff.

This is great, thanks!

For anyone interested:

The famous Knuckleball pitcher Tim Wakefield of the Sox recently passed from cancer last year, his wife Stacy also passed from cancer only 5 months after.

The Impossible Career Of Tim Wakefield - https://www.youtube.com/watch?v=bpZfKftTEIQ

Tim was part of the 2004 Red Sox that broke the 86 year long curse. The entire series is on Youtube and possibly one of the greatest baseball series of all time. Here's game one, enjoy:

https://www.youtube.com/watch?v=5rLWrD1ipi0

https://www.mlb.com/video/red-sox-pay-tribute-to-tim-wakefie...

Something I liked - "If you laugh, you think, and you cry that's a full day."

A guy threw a knuckle in tribute to Wakefield as an opening pitch this season, sorry I can't find it tonight but if someone could find it that'd be nice. Good on 'em.

Takeaway is roughly: those observing the pitch expect it to be moving in one direction, and it often is moving in a different direction, giving the illusion of a sudden change of direction. So the knuckleball is unpredictable, just not in the way that it appears to be.

The cloud of "where it ends up" is not important for establishing the difficulty of hitting a pitch. What matters is the correlation between the initial trajectory and its final location. This is the information that the batter has.

It is a bit more complex than that; what matters is the correlation between the initial trajectory and when the batter's vision believes the ball will go. The dry spitball (knuckleball) and the spitball work on the same concept (i.e., the lack of spin will cause the ball to react differently than pitches with spin).

I'd expect the change-up to be an equally-as-interesting subject for a physics analysis.

The change-up is pretty well understood, I think. Use the same apparent arm motion but grip the ball differently to create a release that’s slower and may have a slightly different spin.

A change up is simply a slower pitcher than a fastball, thrown with the same motion. The slower speed mean that ball, while looking like a fastball, will act differently.

Reply on news.ycombinator.com