Calculate your serve speed

[Chas Tennis]

Originally Posted by Mister drool

why by image and not by sound?

The thumb of a raquet hitting a serve is pretty loud, and if you put a phone (or something similar near the service box you are aiming, it could pick up the sound of when the ball hits the gound... from both noises it could calculate the amount of time the ball traveled the air up until it hit the ground. you type in the measurements of a tennis court, from the baseline to the service box, count the height of the net, and then do the part where im bad at - the maths.


no?

The article below on the kick serve says that the serve bounces so high because it hits the court at a large angle. Since the ball curves more it will travel a longer path. How would serve trajectory based on sound measurements affect the velocity calculation compared to a flat serve?

Rod Cross, the kick serve in "TW University" link above
http://twu.tennis-warehouse.com/lear.../kickserve.php

For comparisons, the standard measurement has become the radar measurement after hitting the ball.

[Mister drool]

I did not tought of that my good sir
I was only thinking flat serve all the way...
good point good point

[pvaudio]

Originally Posted by Chas Tennis

The article below on the kick serve says that the serve bounces so high because it hits the court at a large angle. Since the ball curves more it will travel a longer path. How would serve trajectory based on sound measurements affect the velocity calculation compared to a flat serve?

Rod Cross, the kick serve in "TW University" link above
http://twu.tennis-warehouse.com/lear.../kickserve.php

For comparisons, the standard measurement has become the radar measurement after hitting the ball.

This is exactly the reason they measure this way. Trajectories and various spins vastly alter simpler radar methods. As a simple example using a hand-held radar, if you have a car traveling around a curve, the only point that you should measure the car's velocity (as speed and velocity are not the same) is when it as close to dead-on ahead of you (assuming you're standing in the track and somehow will not be killed by a speeding race car). If you measure it at any other time during the curve, the reading you will get is of its velocity which is going to be lower than its speed (what you really want to know). Put simply: how fast it is going around a bend is not the same as how fast it is approaching your fixed measurement point. The latter is slower. Since a serve can travel in a severe arc, measurements right off the stringbed (in the car example, when you're staring right at the headlights and take the reading) are the most accurate measure.

[3fees]

I'll do it the easy way,,,one person stand at the net with a stop watch(good to the tenths of seconds-interpolate if in between tenths) and clicks it at moment of impact and then clicks it again as the ball crosses the net(fast clicking) take V=D/T
With a calculator in scientific notation input 39'/x seconds multiply by the factor 1.467 to get miles per hour or mi/hr.

examples:
39/.40 sec ~ 143 mi/hr
39/.50 sec ~ 111 mi/hr
39/.60 sec ~ 95 mi/hr
....
39/1.0 sec ~ 57 mi/hr

The factor 1 x 60 sec/ minute x 60 minutes/ hour= 3600 seconds = 1 hour
5280 feet = 1 mile
5280 ft/ 3600 sec = 1.467 miles/ hour



Credit were credit is due for this....

[golfer]

can't you do this with a camera side view of serve. Take the frame at impact and the time it takes to cross the net and use that distance. That way human error is gone. I know bowling is slower but you can get axis tilt revs and ball speed by putting a strip of tape on the ball and recording the throw.

[tennisplayer_85]

hey guys, just updating you on my progress!
So i took last week off to work fulltime on this app! I am really happy with the result so far and hope i can commit it to the app store by the end of this week or at most beginning of next week.

@3fees: I have looked into this method but human error is a big issue here. With a 30 fps camera, selecting the frame the ball is hit and lands, the error can be 10 miles off. I dont think that a human is more precise than 30fps.

Solution: 60fps.

In the App i am using this approach and it works very well. There will always be a slight error in accuracy but it is reduced to the very minimun possible. And i guess we can live with a measurement that is about 2.5 miles off.

please keep on letting me know what you think at:
http://www.tennisservespeed.com/
Because this really gives me a boost to work hard and make this happen.

I'll let you guys now as soon as i am ready!

[paul_tennis]

Great idea, I definitely will get this app for my iPhone.

Good luck with the app!

[v-verb]

Originally Posted by paul_tennis

Great idea, I definitely will get this app for my iPhone.

Good luck with the app!

Me too!

[tennisplayer_85]

Hi guys! I submitted the app to the AppStore just now!
It still needs to be reviewed by apple, dont know how long that takes. But soon it will be available !

Its called, who could have guessed: "ServeSpeed".

I'll let you know as soon as it really is available. Just happy that i made it through this whole process

[BHiC]

Originally Posted by tennisplayer_85

Hi guys! I submitted the app to the AppStore just now!
It still needs to be reviewed by apple, dont know how long that takes. But soon it will be available !

Its called, who could have guessed: "ServeSpeed".

I'll let you know as soon as it really is available. Just happy that i made it through this whole process

What are the chances that you come up with an android version of the app?

The concept sounds really cool, but I don't have an Iphone

[The Meat]

Originally Posted by tennisplayer_85

Hi guys! I submitted the app to the AppStore just now!
It still needs to be reviewed by apple, dont know how long that takes. But soon it will be available !

Its called, who could have guessed: "ServeSpeed".

I'll let you know as soon as it really is available. Just happy that i made it through this whole process

How much will the app cost? Hope you get it approved!!

[SystemicAnomaly]

Originally Posted by sundaypunch

Are you going to tell the OP that Santa isn't real next?

WTF, no Santa? Heresy! Next you'll be promoting the idea of a heliocentric solar system. I'd be willing to bet that you are also a card-carrying member of the Round Earth Society (or do you guys prefer to be known as the Oblate Spheroid Terra Society?).
.

[tennisplayer_85]

hey guys, the app will cost 2.99. so if you can afford some tennisballs you should be able to buy this too. i could not make it free, the idea is if i do make some money with it to keep improving the app in future updates.

@BHic: the more people buy this app, the more chance i'll look into android too. I still need to do some more reseach into android though. how many android phones have a camera that can record at 60FPS?

@SystemicAnomaly: I also bought that app to try it and can confirm it did not work at all.

My app isnt perfect yet either but pretty accurate, definitely the best out there. I hope to achieve even better accuracy with some future plans i have using statistic serve data.

[raemuel9323]

to bad it isn't available in the uk, i would have bought it asap

[cluckcluck]

Check the site, looks cool. Problem is the instructions...there are none.
Should one stand behind the server? On the receiving end? At the net waiting for the ball to cross?

[SwissXV]

When I try to view it on iTunes from your website, it says that it is not available in the US. Why is that?

[paul_tennis]

Originally Posted by SwissXV

When I try to view it on iTunes from your website, it says that it is not available in the US. Why is that?

+1

I'm having the same problem...

[cluckcluck]

Originally Posted by SwissXV

When I try to view it on iTunes from your website, it says that it is not available in the US. Why is that?

+1....says not available in US.

[Avles]

Couple comments on your site, since the app doesn't seem to be available yet.

Your Info page says "more than 500 people up-voted this app at the Tennis Warehouse community." Does this mean that 500 people expressed interest in the app on your poll? If so, your page is misleading, because it implies that the Tennis Warehouse community has actually tried out your app.

Also, you have a tennis ball graphic and a little video graphic on the homepage that lead right back to the homepage. Not so good to have links to nowhere.

Also, I hope you'll eventually reveal the secret algorithm you use for calculating serve speed-- seems a little questionable to ask people to pay money for a tool which relies on calculations that are kept hidden. Just curious, how have you verified that your app is accurate?

[Avles]

One other thing.

If the numbers in your example screenshot are actually from the app, it looks to me like you are calculating the average speed between when the ball leaves the racquet and when it hits the ground, rather than estimating the initial speed of the ball when it leaves the racquet. Is that true?

If so, you should keep in mind that pro radar guns don't measure average speed-- they measure speed coming off the racquet (as mentioned further up in the thread). So your app will read considerably slower than a pro gun.