A linear relationship?

FrisbeeA quick update on the throwing research: I’m still waiting on the data. Getting each throw ready for analysis requires someone to go through the manual process of tagging which marker is the inside of the elbow, which is the heel, etc. For thirty or so markers. For about seventy throws per participant. Then they need to write and run a bunch of macros to pull out the stuff I’m interested in – spin, speed, how many degrees the wrist snaps, how much the shoulders turn, and about thirty other things.

You get the idea – the guys in the lab are facing a long task to get all the data organised before I can even start analysing it properly. And since the raw data can only be analysed on their system, there’s not much I can do to help.

But in the meantime, there’s something interesting I wanted to talk about. We found, when doing a couple of on-the-fly reconstructions during testing (and some back-of-the-envelope calculations) that the spin on the disc is pretty much linearly correlated to the velocity of the throw – i.e. your hucks have much more spin than your short passes. This research found the same thing (and gives, at least until we get our full data, a great deal more confidence in the result). I find it really interesting.

If spin is generated by actively snapping, using the muscles of your wrist, it shouldn’t matter too much what speed you’re moving the disc at. You should always be able to apply the same sort of snap, regardless of speed (though not regardless of acceleration).

On the other hand, if wrist-snap is created by a sudden deceleration of your arm, while simply allowing the wrist to carry on, then it would indeed seem as if a more high-speed throw would result in a sharper deceleration and thus more spin.

Or could it be to do with  the acceleration of the arm/disc just before the snap? Imagine that the muscles of the wrist are fighting hard to maintain a fairly neutral position while being pulled back by the inertia of the disc under acceleration, and then that acceleration suddenly ceases. That would create a much stronger snap than the muscles/tendons could achieve without a resistance to push against. [With your hand flat on the desk, try tapping a finger as hard as you can – and then try resisting your finger-tap with the other hand and suddenly letting go. It’s very different. I just did it and bruised my finger from hitting the desk too hard…]

But why would the acceleration be stronger for a long throw than a short one? Surely the point about a longer throw is that it has a similar acceleration, but over a longer throwing-motion, after a bigger backswing?

If not – if our short throws are generally accelerating less aggressively than our hucks – then we could all have vastly quicker releases on short throws by accelerating harder and having an even shorter throwing motion. There has to be an advantage in that for beating a mark or hitting a receiver earlier. Whatever our maximum comfortable acceleration is, determined by our strength and technique, shouldn’t we be somewhere near it on the majority of throws?

Or is the wrist only a bit-part player? Perhaps the disc is simply spinning off the fingers as you let go – which would tally very nicely with the data, in some ways. The speed – and hence the sharpness of the deceleration – would probably affect the spin in a linear way in this scenario.

Honestly, I don’t know. I personally expect that more than one of these things will contribute to the overall spin – certainly I can throw very short throws with just the wrist, but I can really feel the disc snap off the index finger when I throw a full-power backhand too. My coaching focus is generally on fighting to keep the wrist fairly neutral and using the resistance (like that finger-tapping exercise) created by the disc’s inertia to generate a quick snap when you run out of arm and stop accelerating. But so far, I can’t say for certain. Add your own thoughts below, of course…

What’s great is that sometime soon I’ll get some nice 400 frames/second data. We’ll see how many degrees the wrist actually snaps through impact, and how fast, and how that correlates with the amount of spin. We’ll see whether the disc then increases spin as it comes off the fingers too. We’ll know both the velocity and acceleration of the disc, wrist, and arm at 400fps through the release, for both short throws and long. We’ll see the differences in technique between people who get more spin and those who get less. We’ll see the differences when someone is told to throw for maximum spin instead of maximum velocity.

That’s pretty exciting, I think.

This entry was posted in General comment, Skills and Techniques, Throwing and tagged , , , , , , , . Bookmark the permalink.

13 Responses to A linear relationship?

  1. One of the things that a friend of mine and I have noticed (well, he figured it out, I’m trying get it working in my own throws) is that on his forehand, spin is imparted by him pulling his fingers in toward his palm. The wrist may impart additional spin, but has far more control over the velocity than the rotation. The shoulders and the core are still involved in the usual ways and I don’t think that this is unique to him, but if you watch the Rise Up forehand video Ben is not pulling the middle finger in until the disc is well out of his hand.

    My guess is that we’re probably fiddling with mechanics in a way that leads to bad habits…but it is so much easier to control the disc in the wind with the additional rotation.


    • More than one way to skin a cat, certainly… For myself, I tend (and coach) to get the most spin and power on my forehand from the fingers being pulled by the inertia of the disc and then snapping back to a neutral position, rather than starting neutral and pulling them into the palm. I find the tendons to be stronger and quicker acting than the muscles for this sort of elastic snap; but of course it’s not the only way to throw. I also have very very long thin fingers (I’m over 6’5″) and I do know a number of people with smaller hands who find it easier to keep the disc much more forward in their grip and do something more like you suggest.

      Also, those who have quite a palm-up grip will tend to rely on the fingers snapping back into place, whereas those whose palm faces more forward might be better placed to involve the (relatively strong) muscles that pull the finger into the palm. Hopefully we’ll see from the research data a little more about what might really be happening!


      • So either way, the fingers are snapping through the disc; I guess it wouldn’t be much of flick without that. The problem I occasionally encounter when trying to snap through the disc is that you can turn the throw over far too easily. Your response explains why I have more trouble doing this while trying to throw a big IO forehand. In regards to palm location, if I throw an IO my palm faces more upward than if I’m throwing anything OI (especially if I’m throwing something particularly bendy). In related news, I didn’t handle in college and I’m still shouldn’t be anything more than a tertiary handler/safety valve on any team with good throwers.


        • It’s interesting that you’re aware of a change of grip for the different curves. That’s not to say that grips don’t change – I certainly make adjustments for a high-release forehand, and a slight adjustment for a low-release backhand, and probably tiny adjustments on all different throws – but I’d have thought the majority of the change between O/I and I/O would be from the elbow (and maybe the shoulder, and even how much you bend at the waist).

          Again, though, we got all our research participants to throw both curves, for hucks and for shorter throws, so we’ll be able to look more closely at that too! What we couldn’t work out how to do was measure grip pressure and placement for each finger – any equipment (gloves, or something stuck on the disc) would change the throw too much. It would have been interesting to see just how much our grips do change.


  2. Steve B says:

    Was this correlation equal (within reason) across all participants? A quick thought… I assume the filming was done lab conditions which could make even experienced throwers lazy over short distances. Without wind I’m pretty sure I wouldn’t put the most spin I could on a short throw – it’s just not needed…


    • As for my data, I’ve no idea yet how the correlations will look. Very early days and about 4 data points… I wouldn’t have written this piece based only on what I’ve seen so far. And as for the data in the study cited above, I guess you can see their graphs for yourself – you know everything I do!

      Having noticed this pattern, we did ask some later participants to add some additional throws to the testing where they attempted to throw ‘touch’ passes, with a high spin/speed ratio, to see what would happen and how they did it. The study cited in the article also shows that it’s very possible to deliberately increase the amount of spin on short throws, but whether this is achieved by stopping sharply or by accelerating more aggressively or by some other means we can’t say until we get all our data. That cited study didn’t look at any of the mechanics of the throws.

      The wider point stands though – I could throw a short pass with more touch if I was trying to because of the wind or whatever; but I could also throw a huck with more spin if I wanted (I’d probably sacrifice some distance to do it). It’s still interesting to me that longer throws generally have more spin, and it’s very interesting that the correlation is as linear as it appears in that cited study.


  3. Dajo says:

    When I was doing my physical education studies I was just dreaming about doing a study like this but never managed to go for it. Really great to see you managed to go for it!
    I was wondering if you will also be looking at lower body factors affecting spin and velocity of the disc? Would be pretty interested in seeing the impact of foot placement and hip rotation to name a few. All the best with the analysis!


    • We will certainly have that information about the positions and rotations of the lower limbs – markers were on the hips, thighs, both sides of each knee, calves, both sides of the ankle, heels and toes. Foot placement is probably not the first place we’ll look for correlations – we’ll start with the wrist, arm, shoulders, and hips – but the data will be there if we can find a good way to interrogate it.


  4. Pingback: Light, Dark, and Cleats: 11/6 | Ultiworld

  5. Gabriel says:

    I would expect at least part of the correlation to come from the amount of width of the motion – in a huck you have more time to impart the spin, while for short passes you tend to favor a more compact movement. This may be related or amplified by the inertia of the disc. Most players start the flicking/snapping motion at the beginning of the throw, hence they try to impart spin on a disc that is at best is static, and they have to overcome the inertia of the disc to impart it any spin. However, the very few players that start imparting the spin only toward the end of the motion tend to have excellent hucks, and this may be a similar movement to a shorter pass with a lot of touch when you give it that last “extra push” as you release.
    It may be wrong, but at least it should be easy to tell if you can measure the angular acceleration during the throwing motion. On the other hand, if the above is correct it could be an easy way to learn how to huck more effectively!
    Anyway, I am looking forward to hear more about this! Thank you for the awesome job!!!


    • Thanks Gabriel. It will be interesting to see how and when people start to unwind the wrist. Certainly I would always coach not to start the snapping motion at the start of the throw – in fact, I’d coach not to cock the wrist until quite late in the throwing motion, and even then to only allow it to cock fewer degrees than you might think. But I guess we’ll find out more from the data!


  6. Simon Hwang says:

    Is there an update on this throwing research? A paper published somewhere or something?


    • Afraid not. All has gone quiet from the lab, and I never did get the data. For various reasons (confidentiality agreements, and the fact that we can’t put the analysis software on a computer that I can access myself) I’m dependent on them actually processing it.


Join the discussion...

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s