I’ve had a couple requests for this topic. This is a fair bit more complex than anything I’ve covered in this column so far, but it’s also really cool stuff – hopefully I’ll be able to do the subject justice.

Twisting serves two purposes in upper-level routines. First and most obviously, twisting adds value to skills. For female gymnasts, a back layout is an A, but by adding a ½ or 1/1 twist, the gymnast can increase that value to a B. A back 3/2 or 2/1 is a C, 5/2 is a D, and so on (for male gymnasts a back layout is already a B, but the rest of those values are identical).

The second purpose is to allow the gymnast to more easily spot the landing. While it seems counterintuitive, there are some instances where twisting makes a skill easier to land consistently. One excellent example is in a Tsukahara on vault (performed here by its inventor, Mitsuo Tsukahara). A Tsuk, while not that difficult to rotate, is extremely difficult to stick, because the gymnast cannot see the landing until the last half-salto. At this point, the gymnast is rotating extremely fast, giving him a good deal less than half a second to see the landing and react. However, if the gymnast performs a Tsuk 1/1 or Kazamatsu (two similar but distinct twisting Tsuk variations – a Kaz can be seen here), he is able to spot the landing right around the beginning of the postflight salto, giving him much more time to judge his height and rotation and prepare to land. Therefore, while a Tsuk 1/1 or Kazamatsu takes longer to learn, it is for most gymnasts easier to stick once learned.

So how does twisting actually work?

First, a couple of terms:

TRANSVERSE AXIS: The axis of rotation runing perpendicular to the body’s length. Saltos rotate about this axis.

LONGITUDINAL AXIS: The axis of rotation runing through the body’s length. Twists rotate about this axis.

ANGULAR MOMENTUM: This is the amount of “rotating power” a gymnast has while airborn.

With that out of the way, let’s look at ways in which twist can be generated.

    1. TORQUE: This is where a gymnast begins to turn his upper-body before breaking contact with the floor. 

    2. OPPOSING ROTARY MOTION: This is a bit more complex, as it involves twisting without any net change in angular momentum about the axis. Without any external force, rotating any part of it in one direction causes the rest of the body to rotate in the opposite direction. This is why helicopters have a sideways rotor on the tail; without a tail rotor, the body of a helicopter will spin in the direction opposite the rotation of the blades on top. A gymnast can use this principal to twist by performing a hula-like motion with the hips in one direction, resulting in longitudinal axis rotation in the opposite direction. This is the principle a cat uses to rotate to its feet when it falls form a high place.

    3. BORROWING VIA TILT: If a gymnast is performing a salto, she already has a fair amount of angular momentum about the transverse axis. By the use of asymmetrical arm movements (ie moving an arm up on one side of the body while moving the other arm down on the opposite side of the body), she can cause her body to tilt slightly off-axis. This causes some of the angular momentum about the transverse axis to be “borrowed” to rotate about the longitudinal axis.

Coaches tend to discourage gymnasts from using torque or opposing rotary motion to twist, because both tend to interfere with other aspects of the gymnasts’ technique. A gymnast who tries to torque on takeoff will often end up sacrificing proper set technique on the salto, resulting in a low and often-underrotated salto. Similarly, a gymnast who deliberately uses opposing rotary motion to twist will be forced into a continual hula-mtion, which often makes the salto very difficult to control, and also just looks sloppy. A gymnast who twists by tilting, however, can generate twist while airborn, maintain the same body position throughout the twist, and completely stop the twist before landing. These three things make tilt the preferred focus of gymnasts learning to twist. In reality, there are a few occasions in which gymnasts must supplement their rotation with a bit of torque or opposing rotary motion, but we as coaches tend not to deliberately teach this – it tends to happen on its own anyway.

There is much more nuance to be discussed concerning twisting, and I’ve only scraped the surface. If anybody has any further questions about twisting (or anything else), feel free to ask.

This entry was posted in Uncategorized. Bookmark the permalink.

7 Responses to Twisting

  1. Damian Hilton says:

    I can’t believe someone hasn’t commented on this. Great breakdown, I coach a lot of former team gymnast. It amazes me how many of them don’t know the law of angular momentum. Really appreciate the simple, concise breakdown. I even start from the first Thermodynamic laws. …Energy cannot be created or destroyed only transferred from one type to another. When I explain to the students I can hear the less desirable(non-conserving kinetic) tumbling.
    Additionally, I suggest to my students that if you can do the skill, you can do the skill. Technique is clearly important. However if we are on only after a single twist then they could choose to have some less than desired technique. We are ultimately trying to set a strong foundation that if they are motivated to achieve much higher level skills that attention to physics is a must! …anyway this is clearly very cool to some of us!

    My question- I typically suggest students twist whichever way they cartwheel as they are more comfortable rotating that way regularly. I am a little confused by the “borrowing via tilt”, I believe setting arms wide and then decreasing your longitudinal radius to the side will make your rotation go faster(law of angular momentum), this does not fix “an object at rest stay at rest”. I think this where you are suggesting “borrowing” and want to make sure I am very clear. I want to make sure I explain things correctly also. I might teach the more desirable technique, but if I can’t explain the why …what good am I as a coach. Taylor is one of my students. She has done my requirements, that the focus has been a very high layout. She has shown some consistency in that. I measure actual height as where their hips are, as if I put a big red dot on their hip-bone. Taylor is a left footed/handed cart wheeler so I will suggest that she probably will prefer to rotate to the right. . i am confused by the “asymmetrical arm movements”. Are you suggesting that i should have Taylor set normally leaving briefly her left arm high and right arm wraps. Are you suggesting the old school diver type twists instead of the close body wrap? Please dumb it down for this guy.

    Please and thank you.

    • Damian Hilton says:


    • I think you’re asking two different questions here as one, so I’ll see if I can give an answer to both of them.

      Regarding “borrowing via asymmetrical arm movements:” used to have a really fantastic article breaking it down, but that site appears no longer to exist. Gerald George has an excellent book on the biomechanics of gymnastics in which he discusses it, called Championship Gymnastics; I highly recommend it.

      Here’s my best attempt at explaining it:
      The set should be identical to a normal salto. Once the arms are in the air, any sort of lateral asymmetry (ie moving an arm down on one side of the body, moving an arm up on the other) will cause the body to tilt slightly, so that the transverse axis is not exactly horizontal. The angular momentum generated on take off was about a horizontal axis, but now the body’s alignment is not exactly perpendicular to this original axis.
      If you can find any videos taken from either directly in front or directly behind a gymnast while they are twisting, you’ll generally notice that they are slightly off-axis in the rotation of their salto. This off-axis tilt is what causes the twist to occur. The gymnast can completely stop the twist by un-tilting his/her body.

      You can see this reasonably well in the following video analyzing Kohei Uchimura’s technique:

      As for twisting direction, I’ve heard coaches debate this for hours. Here are the key points:
      -A “left-handed” roundoff actually twists to the right
      -about 80% of elite gymnasts twist in the opposite direction on their roundoffs as compared to their other skills (ie an elite gymnast with a “left-handed” roundoff is more likely to twist to the left, even though this is the opposite direction)
      -There seems to be a bit of a nature-vs-nurture debate regarding twist direction. My understanding is that it’s like being right- or left-handed; that is, most people have a “natural” twisting direction, one which will always feel more comfortable to them. There are a number of ways to test for this. (George Hery did a brief video on this here: However, many coaches argue that it is completely the result of training, and you can thus have them twist either direction. It would take a neurological analysis of gymnast’s brains to say for sure, and I’m not aware of any such study having been done.

      My opinion is that it’s best to start off by determining a gymnast’s “natural” twisting direction, completely independent of which hand they lead with in their roundoffs. If, however, a gymnast seems to be equally comfortable (or uncomfortable) twisting both ways, I tell her to twist in the direction of her roundoff hand (ie a lefty tumbler will twist left, and a righty tumbler will twist right).

      Hope that helps!

      • One more thing: I don’t know if there really is a “best” position for the arms to be in once the twist has initiated. In T&T, they like to see arms by the sides. In diving, it’s one across the chest and one behind the head. In gymnastics, both arms across the chest seems to be the most common (though I do love good clean T&T-style twisting in a floor routine on the rare occasions when I see it).

      • You are correct that pulling the arms from out wide in towards the body will decrease the longitudinal radius and therefore make more efficient use of any existing angular momentum about the longitudinal axis; however, this does the gymnast no good unless she already has some amount of angular momentum about that axis to begin with. Zero is still zero, no matter how efficiently you use it.

        This is where either tilt or torque (or both) will come in; first the gymnast must generate (or borrow) the necessary angular momentum, then she can pull the arms in to decrease her radius and thereby increase her angular velocity.

        Hopefully this clears things up?

  2. Damian Hilton says:

    Thank you for all that! It is interesting but not surprising what you said about “elite”gymnasts and there twisting preference. It seems to be reasonable that which way you twist is inconsequential. It is clearly a regularly held belief and maybe just among old lower level coaches with regards to the cartwheel and preferred twisting direction being the same. Interestingly enough watching Kohei and the former eastern block young man’s(or is that spain on his chest?) bio-metric representations, that maybe the tendency to torque off the ground is greater for those who twist the same way they cartwheel. Although Kohei seemed to torque less in this one rep than the other young man in the representations. This could be part of the explanation for why the best in the world(at least 80%) do it this way. The rest of us….the other ;). I don’t know if you are current/former best in the world. Additionally, I would love to know where you got that statistic on elite gymnasts and their preferred twisting direction. I have often been annoyed with “elite” gymnasts “foo-fooing” the Round-off just so they could really get into where they feel they can gain greater velocity, the back-handspring. The people in the world that look up to them and try to model what they do with not even an 1/16th of the time they get in the gym can see arms wide, head out and assume they can be chinese power tumblers on a non-mechanical floor with this same technique. Wouldn’t be such a big deal if the chance for serious injury wasn’t so great. Let me tell you as a 6’3″ 240 pound guy if I tried alot of these things I would have never been successful, especially starting at 18. They seem to care very little about conserving the kinetic energy that they later alter to more vertical and less linear. Making it good enough. Off Soapbox. I suppose I should be a little more sympathetic considering how many hours they are training.

    For the sake of argument, I have long held that if we want to twist more, or along our longitudinal axis that we follow the law of angular momentum when considering velocity moreso, therefor meaning the smaller we can make our longitudinal profile or making more of our mass to the center and elongating our body for a smaller radius we will twist more in a shorter time. So always arms-in, over elbows-in twist tech.

    Like divers choosing the head and chest version, is the benefit too small(arms & shoulder mass being too negligible?) for us to really get a more satisfied result for coaches and athletes? I try to approach things in a way that if kids go to higher level coaches they have a very strong foundation to build on and that we will work very well with the time we do have. There is a clear priority list in what we choose to focus on as coaches. Don’t most coaches focus on getting as high as possible first and foremost?

    Thank you for taking the time to communicate with me and it seems I need to get the book you suggest. I have always tried to check my physics understanding with my atomic physicist friend, Dr. Laurent. If you feel I am lacking there please correct me. Sorry for being so chatty. Love the discussion and your take on what I believe to be true.
    Take care

  3. I got the Statistic on twisting direction from George Hery — I don’t know where he got it. I think I vaguely remember seeing the study itself published somewhere, but I forget where.

    Haha, as far as being former/current best in the world: at my peak, I was bottem-end elite. In other words, high enough level to compete with the national team guys, but not enough to have any real chance of beating them.

    I’m actually a pretty big fan of an arm offset in a hurdle going into a roundoff — ie, rather than having both arms up, one arm reaching forward and the other reaching up. Tony Gehman has a good explanation of this technique here:
    I am not, however, a fan of arms going out wide. I have always taught roundoffs as if they were done between two walls; as little lateral movement of the arms as possible.
    But as far as the importance of proper technique in roundoffs in general, I completely agree. But then, I also consider a roundoff to be the single most technically difficult skill in the entire sport of gymnastics. (I’ve even blogged about that:

    I used to emphasize height over all else on saltos, but less so these days. These days, I look at height as an indicator of good technique on most salto skills, but not necessarily as the ultimate goal. Coming out of a backhandspring, a gymnast has a finite amount of momentum. Some of this will be converted into vertical momentum, and some into angular momentum. In principal, you have to sacrifice a bit of height to get greater rotation. In theory, a mechanically perfect double or triple back goes lower than a mechanically perfect back tuck. Again, Dr. George’s book has a great breakdown of this.

    And there’s no such thing as “too chatty;” I would be perfectly content to chat about gymnastics all day.

    So you know; I don’t update this blog anymore (since Apex Gymnastics went out of business), but will likely be starting a new gym-related blog fairly soon.

Leave a Reply

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

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

Google+ photo

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

Twitter picture

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

Facebook photo

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


Connecting to %s