Welcome, Guest. Please Login
SLINGING.ORG
 
Home Help Search Login


Page Index Toggle Pages: 1
Send Topic Print
Video frame counting for speed - kind of works (Read 106 times)
bandit111964
Descens
***
Offline


Trust GOD’s plan for you!!

Posts: 205
Boothbay Harbor, Maine
Gender: male
Video frame counting for speed - kind of works
Aug 22nd, 2026 at 8:44pm
 
Well, it would obviously work much better if I used better software and hardware instead of try to frame count using an iPhone and free websites.

Before today I had guessed I was slinging in the mid 90’s mph. 

Today I tried to throw hard while still being accurate.  Frame counting the video afterwards and then adjusting the average speed to starting speed, showed pitch speeds of 94 mph to 100 mph. 

Might be wrong, but sounds reasonable.

Was happy with accuracy too. Even tossed a few baseball at the end to compare them to the tennis balls.  Was a fun day!

Feel free to check the velocity. Throwing distance was 60 feet, 6 inches from the fence….measured with a tape.

https://youtu.be/geD5Z6l8omM?is=hCk4fVUxU46CtvMl


Back to top
 
 
IP Logged
 
erricrice
Funditor
****
Offline


reformed spaghetti enthusiast

Posts: 507
Clearwater, FL
Re: Video frame counting for speed - kind of works
Reply #1 - Aug 22nd, 2026 at 10:28pm
 
I've never done any deep research on velocity loss for tennis balls, but I'd be a bit wary of anything estimated by AI. Tennis balls are pretty lossy projectiles so it certainly could be a lot but I feel like 30% is a bit much even by those standards. Good throws regardless! It gets a lot more fun as you get up there in speed and get your technique dialed in. And even more fun if you have somebody walk up with their dog and you launch their dog toy into the stratosphere Grin

My math here, in case I missed something:

16-17 frames depending on the throw
.5333 seconds (16 frames) / 60.5 feet = 113.6 fps
113.6 * 1.4666 = 77.5 mph average speed

Best option like you said would be to measure yourself with a proper radar gun - there are some cheap-ish($80) hand models you can get that do a decent job without a ton of cosine error, but the best options are $200+ see this thread: https://slinging.org/forum/YaBB.pl?num=1783016068



Back to top
 
 
IP Logged
 
bandit111964
Descens
***
Offline


Trust GOD’s plan for you!!

Posts: 205
Boothbay Harbor, Maine
Gender: male
Re: Video frame counting for speed - kind of works
Reply #2 - Aug 23rd, 2026 at 12:00am
 
Definitely having fun now that I can mix slinging my baseball background.  Ready to pitch to a real batter…tennis balls of course as still a little wild. Just got to find someone brave and dumb enough!

Thank you so much for checking!  Totally agree with average speed. I count 16 frames and get 77.5 mpg too. 

I don’t see hard data either on tennis ball speed reduction. I don’t know it is correct but I feel like it might be.

Intuitively I would imagine the tennis ball is designed to slow significantly for safety reasons of course but I think also very likely for playability too.

A tennis court is essentially the same length as baseball pitching distance. The problem is the tennis ball is coming off the server’s racket at 120 mph vs a MLB pitcher thrown ball speed of 95 mph. If the tennis ball did not slow down significantly it would be un-hittable I would think.

So reduce the tennis ball speed from 120 mph to 95 mph in 60 feet distance. That is 21% reduction in speed in 60 feet.  Honestly 95 mph is still pretty hard to hit.  A 30% reduction from 120 mph serve would mean trying to hit a 84 mph ball….which is more reasonable to accomplish.

So I don’t know 30% speed reduction is true….but seems like it kind of could be.

Anyone have the real answer for how much a tennis ball slows in 60.5 feet distance?
Back to top
 
 
IP Logged
 
IronGoober
Interfector Viris Spurii
*****
Offline


...and now, No. 1, the
larch...

Posts: 1983
California
Gender: male
Re: Video frame counting for speed - kind of works
Reply #3 - Aug 23rd, 2026 at 12:34am
 
This was the velocity drop that I got from mapping out position with a camera (from the side) using somewhat calibrated images.  Take it with a grain of salt, but the drop in speed is quick. (this is given in m/s)

Attached is a video of the test. I placed 2 tennis balls on the ground at 5m apart and slung over the top and in line with both of them (to try and keep the distance/pixel accurate). I used the spatial calibration of the distance between tennis balls for the entire image. It seemed to be pretty accurate because my calculation said release speed of 56 m/s and the radar was 55 or 57... I don't recall, I just remember it was 1m/s off of the camera calculation.
Back to top
« Last Edit: Aug 23rd, 2026 at 12:02pm by IronGoober »  

test.mp4 (1855 KB | 1 )

John R.
 
IP Logged
 
FloatingFlint
Descens
***
Offline


Slinging Rocks!

Posts: 132
Germany
Gender: male
Re: Video frame counting for speed - kind of works
Reply #4 - Aug 23rd, 2026 at 4:53am
 
If we have an average velocity of 77 mph (~34.5 m/s) over the 60ft 6" (18.44 m) and assume a terminal velocity of the tennis ball of 25 m/s, then this is the velocity drop over the thrown distance. The initial velocity is a bit over 40 m/s which is about 90 mph.
If the terminal velocity of the tennis ball is more at 21 m/s, as some sources suggest, then the velocity drops even more which means that your initial velocity was a little bit higher.
Back to top
 

... and chose him five smooth stones out of the brook, and put them in a shepherd's bag which he had ... (1.Samuel 17,40)
 
IP Logged
 
bandit111964
Descens
***
Offline


Trust GOD’s plan for you!!

Posts: 205
Boothbay Harbor, Maine
Gender: male
Re: Video frame counting for speed - kind of works
Reply #5 - Aug 23rd, 2026 at 8:03am
 
Thanks so much for the help!!

Flint:  i was calculating a lot of the throws at 94 mph.  That agrees with your comment the speed could be a bit more than 90 mph if terminal velocity changes.  3 of the tennis balls were brand new and the others were used from 1 to 3 months. “Fuzziness” and drag differed I am sure and terminal velocity possibly could change.  So this helps show my calculations are not completely unreasonable.

Irongoober: 56 m/s is 125 mph!   Obviously it is a fact you are way more skilled than me.  If I am mid-90 mph (maybe 100 mph) that would be consistent with the expectation your speed is way higher than my speed.   So maybe if I claim a higher terminal velocity and add a little speed to Flint’s calculations, and the result is my small addition still remains very far from your speed….well, that helps show perhaps the numbers aren’t horribly inaccurate.
Back to top
 
 
IP Logged
 
IronGoober
Interfector Viris Spurii
*****
Offline


...and now, No. 1, the
larch...

Posts: 1983
California
Gender: male
Re: Video frame counting for speed - kind of works
Reply #6 - Aug 23rd, 2026 at 12:05pm
 
oops. the graph I put up disappeared. I fixed it.

@Bandit I was just trying to point out that the drop in velocity is very fast, nothing about our relative speeds.
I get about 30% drop (over 20m) between the release speed and average speed, so that is in line with what you calculated. You could have been hitting 100mph with an average speed of nearly 80mph.
Back to top
 

John R.
 
IP Logged
 
bandit111964
Descens
***
Offline


Trust GOD’s plan for you!!

Posts: 205
Boothbay Harbor, Maine
Gender: male
Re: Video frame counting for speed - kind of works
Reply #7 - Aug 23rd, 2026 at 2:55pm
 
Thanks for the graph.  So nice to see actual data instead of just relying on what Google tells me.

It is interesting the tennis rules control tennis ball speed indirectly via size, mass, and bounce due to the differing playing surfaces. There is no direct rules saying how fast a tennis ball can go or how fast a ball should decrease speed due to drag.

Fortunately we are some slinging data nerds who know how to generate our own data!

Thanks so much for the help!

Back to top
 
 
IP Logged
 
erricrice
Funditor
****
Offline


reformed spaghetti enthusiast

Posts: 507
Clearwater, FL
Re: Video frame counting for speed - kind of works
Reply #8 - Aug 23rd, 2026 at 3:08pm
 
Wow this is great data! Crazy that it drops speed so much faster than I thought. If only racquet balls were a little heavier and cheaper, they'd be a much better practice ammo.

That's three different sources with effectively the same outcome, pretty well confirmation there barring some much more in-depth tests which are probably not worth the effort(and will probably show the same result).
Back to top
 
 
IP Logged
 
Page Index Toggle Pages: 1
Send Topic Print
(Moderators: joe_meadmaker, Morphy, vetryan15, Chris, Kick, Rat Man, Curious Aardvark)