Measuring TopSpin!

Is there any machine or way to measure rpms on let's say a forehand. I read in men's vogue that Federer hit a 4400 rpm forehand, at 80mps. It was the highest reading ever recored by the researchers. It's the Men's Vogue with Clooney on the cover. Chime in! Thanks!

Kind Regards,
Dmitri
 

Kelly

New User
Quick check on tennisplayer.net says average Roddick serve was like 2600 rpm on first serves. Sampras' was like 2700!

Apparently a lot of the rpm on Roddick's serve is actually sidespin, and most on Sampras' is topspin. I haven't seen Sampras play though, I wish I could have.
 

Jonnyf

Legend
Kelly said:
Quick check on tennisplayer.net says average Roddick serve was like 2600 rpm on first serves. Sampras' was like 2700!

Apparently a lot of the rpm on Roddick's serve is actually sidespin, and most on Sampras' is topspin. I haven't seen Sampras play though, I wish I could have.
klkl kl

Yeah sampras was before my time ashame though. I have a hint if you like the older player but was too young for them find the Delta tour on tv/or go and see them.
I've gone to the delta tour final 2 years in a row now (like Maters Cup) and seen Mcenroe, Becker,Courier, Muster etc. and there is a rumour sampras might jooin
 
S

SageOfDeath

Guest
Pisolino227 said:
Is there any machine or way to measure rpms on let's say a forehand. I read in men's vogue that Federer hit a 4400 rpm forehand, at 80mps. It was the highest reading ever recored by the researchers. It's the Men's Vogue with Clooney on the cover. Chime in! Thanks!

Kind Regards,
Dmitri
I don't think anyone answered his original question..... I'm curious too.
 

JohnYandell

Hall of Fame
Hey Guys,

I tried to find that copy of Men's Vogue since that was our research about the spin rates--it must be at some newstand in SF somewhere...

But to the question--yeah you can measure your own spin but you need a $150K high speed camera--or you can rent the set up for only about $10K a week...

It's similar to the swing vision thing on CBS, but we sometimes point it at the ball to study spin as well as the players for high speed stroke analysis.

We have filmed Nadal as well. We'll be adding that data to the Advanced Tennis
data base (www.AdvancedTennis.com) over the next few months and I'll also be writing about it on Tennisplayer.net--there is some spin data and analysis of it there now.

John Yandell
 

!Tym

Hall of Fame
One thing though, isn't the problem with this quantitive research though that you may not get enough of a sampling to assess what truly is a player's spin range?

Players can hit all sorts of spin during a match, they might launch a moonball for example here and there when in trouble and that would skew the numbers greatly if you had a small sample, or maybe a random bad mishit or something would skew things as well. There are so many variables in a match, and the thing is I don't really doubt that any one player can hit an unusually high rpm or unusually low rpm on any given shot, even a series of shots, or even for one tournament.

I just feel like the most relevant way to assess "spin" is to simply ask the players what they think after having faced each other day in and day out for years. Poll them, they play and practice with each other day in and day out so they know what each other's strokes "feel" like to one another more intimately and accurately than anyone in my opinion, and that's not something that can be told from a small sample size of "quantitave" measurement in my opinion.

The problem I see with the high speed camera approach is that its just so expensive that obtaining a full sampling of players is not really possible, partiuclarly since players will change up how spin they're going for depending on surface and/or whether conditions or even their mood on a given day or who they're playing on any given day, maybe they're having a bad day, or holding back feeling the effects of an injury or soreness or something, maybe they're mentally distracted, who knows?

I just think that these numbers are cool and novel and all to hear, but in a way I feel like it's kind of just window dressing, a novelty number to reference, but not necessarily really all that relevant in reality since players don't play in a vacuum where all things stay constant. And besides they know that hey so and so's spin feels heavy, they don't need a number to tell them that. They know because they face each other every day and they have scouting reports on each other.

In my opinion, I think the money spent on this high cost analysis might be more usefully served by say trying to determine the REAL benefit of say different racket or string technologies, say trying to see if players hitting against a ball machine set to a constant pace, spin, and placement with a large sample number of balls hit with more or less spin on average with different string and tension and guage setups in their rackets, for example. How much is placebo affect, how much is imagined, how much is real?

That's where I think this technology would be more useful.

Sure if you had unlimted funding to follow players for an entire year, under a variety of conditions and surfaces, and player matchups, heck, even in practice situations, the difference between when they look like they're tanking a match or going all out in a match, the difference in their numbers between a "in the zone" day and a "in the dog house" day, etc., etc. AND sample the ENTIRE tour up to the top 100...THEN, I think these "numbers" would be soooo much more relevant and useful. But as it is, while it's cool to reference these "numbers" I can't help but feel it's a flawed approach simply because of the limited sample size.

Any system that by cost and trouble excludes 99% of the tour from being "studied" is to me a bit short-sighted. After all, how are we then to quantify what separates the "average" players from the "name" players? Is it the actual rpms of their strokes? Is it more than that, is it perhaps mental? What if we find that some journeyman claycourter based on a small sample size generates the same rpms, perhaps even more on one "magic" shot, than say Nadal? Does that mean his strokes are better than Nadal quantifitavily speaking? OR are we to determine that maybe if their "numbers" compute about the same, then maybe it's the intnagibles such as mental toughness and self-belief that separate the journeyman from the top tenners. Or perhaps, the small sampling "numbers" were just a fluke for this journeyman player, maybe he was just having his best match of the year this day or something, maybe the atmospheric conditions were just right, maybe he was inspired on this day by the meory of his late grandmother who passed away on this day, I don't know? Point being, that a larger sample size is needed, particularly a larger sample size under a VARIETY of conditions, surfaces, weather predicaments, players matchups (i.e. who you're playing, the "work" on their shots DOES have a direct correlation on the shots you're going to be generating).
 

MusterFan

New User
4400 rpm? Wow.

4400 revolutions/1 minute = 4400 revolutions/60 seconds = ...

... over 73 full revolutions in one second! I'm sure the data is accurate, but it just boggles the mind. I'm looking at the second hand on a wall clock and trying to imagine a tennis ball turning 73 times in the tick of a second. That is some serious spin!!!
 
!Tym, I agree with you, but I don't think anyone disagrees with you either. All these numbers, serve speed, rpm... are fluff that adds to marketing but none the less I find it very interesting.

But also you never know what findings the research will lead to ...
 

JohnYandell

Hall of Fame
First of all, there is no doubt that more data is almost always great, but Tym (is that you Bill?) is making statements that aren't factually based about what we have done so far, and the size of the data base we've accrued. So maybe we should start with some facts.

In 1997 alone when we did the first studies we captured over 700 events from about 20 players in live pro match play. Definitely fewer events per player means less accurate data for that player, but for an overall sense of the spin rates and the differences in the grips and spins--well it's all there and quite obvious in the data. And there to see on AdvancedTennis.com. So you might want to check that out before making too many more pronouncements about what the work is or its accuracy...

Since then we have had several more major data captures again involving hundreds of events. We haven't released the newest data yet, but to give just one example, when we filmed Roddick we captured over 20 forehands that ranged in spin from about 1500rpm to 4000 rpm and averaged about 2600rpm.

Federer? 20 events ranging from 1300rpm up to 4400rpm (that ball we also know was going 80mph). But his average spin was actually about the same or actually slightly higher than Roddick, about 2700rpm. We'll have data eventually on Safin, Hewitt, and Nadal as well.

I can say that if you take 15-20 radar gun speeds on the first serve of any pro player, you get a pretty accurate estimate of their speed range. I know this because we have done it and compared it to year long averages. I think the same thing is definitely true when it comes to spin. Your best criticism is that this data would different on different surfaces. How true if totally obvious! God willing, eventually we will look into that as well in the future. In fact we have already done some limited, controlled studies.

The point is that this work is part of an emerging picture that we have never had of the actual physical properties of the shots. Right now we are working to pair the spin data with speed and trajectory data that will tell us even more.

But even at this point I think it is more than interesting that Federer with a very moderate grip is actually generating as much or more spin than Roddick. That certainly goes against conventional wisdom, and what 99% of teaching pros believe--more extreme grip equals more spin. When we pair this info with the high speed stroke footage we can start to get a clue of not only why Roger's forehand is so nasty, but HOW he makes that way.

Another other issue to consider is the proportion of topspin and side spin in a given ball. Another study of the serve that was published in the PTR magazine compared Sampras to Rusedski and found a significant, quantitative difference in the heaviness of the ball at the time of the return based on the relative amount of topspin. More on that later on Tennisplayer.net.

When it comes to the application of high speed film to rackets, Tym, again you need to have accurate information (as opposed to no information...sorry) about what the technology actually is, what it's limits are, and what it can show before you are able to guide us with your advice about what would be more valuable in our research. 250 frames a second will tell you nothing about the rackets or strings. You need thousands of frames per second. And you need very close shots from specific angles. For technical reasons I won't get into here, that kind of filming is impossible in match play--although if you could get the players into a lab (and still had those unlimited funds) you might learn something. (Good luck with that one.)

I would agree that we have far from a complete picture of what's going on, but you know what, it is a start and we are seeing some things that are probably significant and have coaching applications.

My own belief, however, is that the role and influence of coaches and coaching information is overhyped anyway. Players are really the ones who guide and direct the direction of the game. We can't be rigid or dogmatic about what we believe or try to force on them--ie, "you have to match Federer's forehand rpms or you'll never play on tour." That's just silly and no player is going to listen anyway.

BUT to the extent we are making statements about players to players, more accurate information is preferable, wouldn't you agree? The first rule of being a doctor is "do no harm." I'm afraid that if this limitation was applied to tennis teaching, a lot of very well meaning coaches would still be out of business.

I personally think that the main value of this work still lies inthe high speed qualitative look at the strokes--that, again, had never been done. But we are making progress in many other areas. If we do ever get the complete picture of ball flight, speed, and spin that will be tremendously valuable as well as damned interesting.
 

ace94

New User
I was watching tennis and the commentators said that they did this with nadal's forehand and he hit with 5000 rpm!!!!
 

JohnYandell

Hall of Fame
Yeh I saw that too and emailed Mary Carillo. Apparently it's a different study--those aren't our numbers.
I have seen Nadal hit a 5000rpm topspin forehand, but would be very surprised if this was his average value--like twice Federer's. I've been trying to chase down the details on that one.
 

DarthCow

Rookie
Isnt there some sort of calculator on the TW website that can measure it?
Actually wait, i think im thinking about something else. Never mind.


Maybe some mathematician could base it on the angle of the racquet face on contact, with the top speed/acceleration of the racquet.
 

RanchDressing

Hall of Fame
Isnt there some sort of calculator on the TW website that can measure it?
Actually wait, i think im thinking about something else. Never mind.


Maybe some mathematician could base it on the angle of the racquet face on contact, with the top speed/acceleration of the racquet.
im no scientist but the incoming ball makes a huge difference on these things (yeah its obvious) but if i manage to get my stick to block back nadal's claycourt stroke, it will still have massive spin.
a mathematical formula may be a possibility only on like a self feed rally start ball.
but to be accurate the math dude would have to take into consideration the string tension the stroke, the amount of cover a players stroke puts on the ball, the ball's condition.... there are just to many variables to consider for it to be even close to a estimation...
for some people its difficult to understand that a fed forehand at 90mph has about asmuch topspin as a lower speed nadal forehand.
i wish there was a place (maybe at the SLO tw ! :) ) where you could have your shot be like measured for spin amount... for a fee ofcourse.
like a room with all the cameras and you bring in your stick, they give you a two toned ball and you hit it er somthing.... maybe charge 50-100bucks for a session of a few min er somthing... that would be sick....
 

Goldenwest

New User
Sorry to reawaken an old post, but I have the same question as the original poster.

I found one report on topspin from 2010 which used a Phantom camera capable of shooting 1,000 frames per second (fps).


Taking a quick look around Amazon, it looks like the Sony RX10 II offers 1,000 fps at a resolution that is only slightly below traditional HD. Cost is relatively inexpensive at $575.
Any practical reason, other than preserving the ability to zoom in without degradation, to not going with a cheaper set-up like this for doing motion capture, as well as rpm research? I'm planning on buying two cameras to video from the side as well as behind, so it is a big benefit to find the cheapest workable solution.
 

Shaolin

G.O.A.T.
Playsight courts measure spin, velocity etc. I've had my shots clocked. Pretty fun and helpful to use.
 

Demented

Semi-Pro
Playsight is using math to back calculate the spin. It uses the speed reading and trajectory to figure out where the ball should be if it was flat and then the actual location to determine how much spin would be needed to make that happen.
 

Goldenwest

New User
Wow, the amount of brain power on this forum is amazing. Thanks to everyone who responded.

John - What fps do you like to have video recorded at for topspin analysis? Is 1,000 fps useable or do you prefer a faster frame rate?
 

JohnYandell

Hall of Fame
Demented,
True--but in tests where we also filmed balls with high speed cameras, the spin numbers were within 20% or most but say 1 of 4 or 5 the number was way off.
 

Goldenwest

New User
Apologies that I can't find that article again. I did find on the ITF Tennis website that they used 1,000 fps for a study on racket and strings research, so that seems to confirm the previous article's assertion that the ITF was using 1,000 fps as a baseline in research. (I'm hoping this link works).


Glad to know 240 is plenty for estimating fps, that makes this process even cheaper (although that Sony is a pretty cheap camera).

Thanks again for the insights!
 

Demented

Semi-Pro
Yeah.. playsight will calculate wind as spin also if the ball leaves the sight of the analytic camera upwards then the reading will be way off.
 
Top