How Much Clubhead Speed Is Generated by Forearm Rotation?

[Warlock]

From purely an academic standpoint, I've never been able to figure out if the rotation of the forearms (mainly the lead forearm) can be used to significantly increase clubhead speed in a real golfer's swing.

I've looked at Dave Tutelman's website (link below) and there seems little mention of M_Arm (ie. torque rotation of the arm/forearm) and it's relation to clubhead speed generation.  Dr Sasho MacKenzie's model (with torque generators for body segments) was optimised for clubhead speed and he produced these set of graphs below.

Modeling the Golf Swing

The area under the M_Arm graph is 'torque impulse' which I assume will mean a change in the angular momentum of the club because the arm torque generator is directly linked to the club via the wrist joint.  Does anyone with a bit of scientific background (hopefully better than mine, as I did physics a long time ago and have forgotten lots) know whether this could cause a significant increase in clubhead speed by impact?

I also read this other article by Dr Sasho MacKenzie (I've attached the .pdf ) titled 'Club position relative to the golfer’s swing plane meaningfully affects swing dynamics'.

It seems to be the same model being used for various simulated swings but within the text he says the following:

----------------------------------------------------------

"The clubhead speed at impact for Sim3 (44.1 m/s) was 24% higher than that generated
during Sim4 (35.5 m/s; Table II). The optimization algorithm found the muscle
coordination pattern that resulted in the highest clubhead speed at impact while also
ensuring the clubface was square to the target line. The clubhead speed for Sim3 (44.1 m/s)
was 22% higher than that generated during Sim1 (36.2 m/s), which indicates as to how much
active rotation of the forearm from a muscular torque can contribute to clubhead speed"

-----------------------------------------------------------

A 22% increase in clubhead speed due to active musculature rotation of the forearm seems a huge amount. Is it actually feasible for a real human golfer to create that much of an increase in clubhead speed?

Sasho MacKenzie2012Clubpositionrelativetothegolfersswingplanemeaningfullyaffectsswingdynamics (1).pdf

Edited September 9, 2022 by Warlock

[Big Lex]

I read the whole paper. Pretty interesting to read. The swing models and the "Darwinian" approach to creating the model golfer is just wild. 2000 generations of golfers in 50 minutes 🙂

I have about the same physics background as you, so I'm not sure I know the answer to your question. But that 22% number was a difference in model-generated swings, and was comparing a swing with zero forearm rotational torque to one with some amount of it. Is that a realistic representation of the range of how we use our forearms in real swings? I don't know. The model is supposed to be based on validated parameters for how we can move our muscles, though, and the model should have rejected any swings that were not humanly possible.

One way to look at it is that adding a torque late in the swing should have a proportionally larger effect than earlier in the swing. The club head is already moving pretty fast by mid downswing, and an effective arm torque, being later in the kinematic sequence, could have an unexpectedly large effect on club head speed, because it's further accelerating something that's already moving fast. 

[mdl]

Isn't the forearm rotating just enabling the wrist to torque the shaft lever (undoing lag)? As in, the force at the sweet spot added simply by rotating the club is incredibly small compared to that added by releasing lag (maybe like 18-19x given ~2" from hosel axis to sweet spot versus ~36-38" for iron shaft). So unless you credit some of that force to forearm rotation because rotation is required for us to undo lag with a square club face, the forearm rotation is adding very little.

Edited September 9, 2022 by mdl

[Big Lex]

3 hours ago, mdl said:

Isn't the forearm rotating just enabling the wrist to torque the shaft lever (undoing lag)? As in, the force at the sweet spot added simply by rotating the club is incredibly small compared to that added by releasing lag (maybe like 18-19x given ~2" from hosel axis to sweet spot versus ~36-38" for iron shaft). So unless you credit some of that force to forearm rotation because rotation is required for us to undo lag with a square club face, the forearm rotation is adding very little.

Hmm. I don't know.

If the forearm is supinated when the wrist is fully uncocked, then yes the rotation only turns the club face a little.

But if the lead forearm supinates while the wrist is still cocked/"lagged", the club head gets an accelerating torque.

At the top of the backswing, it seems like in many expert swings, the lead forearm pronates a little, which "flattens" or shallows the plane of the club when viewed from behind. If you imagine supinating the left forearm at the top, the club head would go above the plane and be steep and I can't think that happens. 

But maybe it happens a little later in the downswing? 

I know the motorcycle move of the lead HAND also flattens the shaft, and I don't know if expert golfers do BOTH moves to shallow the club, one, or the other. 

I'm going to see if the GEARS data on tour pros shows forearm supination and I'm going to see if it's happening early in the downswing. 

[Warlock]

3 hours ago, Big Lex said:

Hmm. I don't know.

If the forearm is supinated when the wrist is fully uncocked, then yes the rotation only turns the club face a little.

But if the lead forearm supinates while the wrist is still cocked/"lagged", the club head gets an accelerating torque.

At the top of the backswing, it seems like in many expert swings, the lead forearm pronates a little, which "flattens" or shallows the plane of the club when viewed from behind. If you imagine supinating the left forearm at the top, the club head would go above the plane and be steep and I can't think that happens. 

But maybe it happens a little later in the downswing? 

I know the motorcycle move of the lead HAND also flattens the shaft, and I don't know if expert golfers do BOTH moves to shallow the club, one, or the other. 

I'm going to see if the GEARS data on tour pros shows forearm supination and I'm going to see if it's happening early in the downswing. 

 

If you look at the M_Arm graph it starts building torque above zero around P5 (left arm horizontal). I can imagine that the MOI of the club about the longitudinal axis of the lead arm is very high at that point so any small lead forearm/arm torque might not have much effect on moving the club off plane (ie. toppling the club). Although the MOI might be rapidly getting less as the angle between club and lead arm starts getting larger.  

The increase in clubhead speed due to forearm rotational torque must be happening much later in the downswing when the MOI has decreased. But won't that still cause the clubhead to move off the 'functional swing plane' ?

[iacas]

My off-the-top, gut reaction is that it adds a little, but of course the primary function is to help square the face.

The back of the left hand is generally facing out at ~A6, and the face is often nearly in the same direction, and then must rotate about 70-90° by impact A7, so… since the shaft isn't inline with the lead arm that whole time (the wrists aren't fully ulnar deviated with the lead arm and shaft inline in all directions), it adds a little speed, but… again, not the primary function. And minimal speed gain given everything else used to accelerate the clubhead.

[SEMI_Duffer]

FWIW: The guy that started the on-line training academy I'm using (degree in biomechanics) addressed forearm rotation at some point during back-swing training with a comment something like "I don't see what this adds other than a complication" and recommended not doing it.

 

[iacas]

11 minutes ago, SEMI_Duffer said:

FWIW: The guy that started the on-line training academy I'm using (degree in biomechanics) addressed forearm rotation at some point during back-swing training with a comment something like "I don't see what this adds other than a complication" and recommended not doing it.

Sounds like he needed to study harder?

Do you know what a swing would look like with no forearm rotation?

I don’t want to get off topic, so let’s just shut that line of discussion down now as the fallacy it is.

[SEMI_Duffer]

Probably pretty silly :)  Quite possibly I'm misunderstanding something?  He was talking about a purposeful club face manipulation during the back-swing.

I'll just sit down and shut up.

Edited September 10, 2022 by SEMI_Duffer

[Warlock]

On 9/10/2022 at 2:55 AM, iacas said:

My off-the-top, gut reaction is that it adds a little, but of course the primary function is to help square the face.

The back of the left hand is generally facing out at ~A6, and the face is often nearly in the same direction, and then must rotate about 70-90° by impact A7, so… since the shaft isn't inline with the lead arm that whole time (the wrists aren't fully ulnar deviated with the lead arm and shaft inline in all directions), it adds a little speed, but… again, not the primary function. And minimal speed gain given everything else used to accelerate the clubhead.

So are you saying that the club squaring mechanism is probably as per Dr Sasho MacKenzie's explanation below with little forearm musculature effort required?  His research for optimising clubhead speed with his model generators both squared the clubface at impact but with a 22% increase in clubhead speed.

Doesn't that mean a real golfer could theoretically also do the same?

 

Edited September 11, 2022 by Warlock

[iacas]

18 minutes ago, Warlock said:

So are you saying that the club squaring mechanism is probably as per Dr Sasho MacKenzie's explanation below with little forearm musculature effort required? 

No. Nor am I not saying that.

I am saying what I said - that the primary function is to square the face.

The golfer in that image has the CM and shaft of the club below the hand path.

[Warlock]

44 minutes ago, iacas said:

No. Nor am I not saying that.

I am saying what I said - that the primary function is to square the face.

The golfer in that image has the CM and shaft of the club below the hand path.

Not disputing that the primary function is to square the clubface but if there was a way of also increasing clubhead speed by 22% wouldn't that be an added bonus?  Although I suspect it would require impeccable timing to do both. 

[iacas]

5 minutes ago, Warlock said:

Not disputing that the primary function is to square the clubface but if there was a way of also increasing clubhead speed by 22% wouldn't that be an added bonus?  Although I suspect it would require impeccable timing to do both. 

Increasing it 22% over what?

My point is that you're going to rotate your forearms regardless (because you have to square the face). So I'm not interested in swings where you "don't" do that and comparing it to swings where you do.

Also, the difference between swings 2 and 3 were not "rotating the forearms versus not" or something, but instead…

Quote

The combined moment of inertia of the left forearm, hand, and club, relative to longitudinal axis of the forearm, was nearly identical for both simulations over the first 0.1 s of the downswing (Figure 4d). The moment of inertia then dropped quickly due to the ulnar deviation at the wrist bringing the center of mass of the club nearly in-line with the forearm. Ulnar deviation was delayed considerably for Sim3 in comparison to Sim4; hence, the associated delay in the timing of the decrease in moment of inertia for Sim3.

[Warlock]

26 minutes ago, iacas said:

Increasing it 22% over what?

My point is that you're going to rotate your forearms regardless (because you have to square the face). So I'm not interested in swings where you "don't" do that and comparing it to swings where you do.

Also, the difference between swings 2 and 3 were not "rotating the forearms versus not" or something, but instead…

This is what is says in that research article I attached.

--------------------------------

"The clubhead speed for Sim3 (44.1 m/s) was 22% higher than that generated during Sim1 (36.2 m/s), which indicates as to how much active rotation of the forearm from a muscular torque can contribute to clubhead speed"

"Sim1 started with the forearm angle set to -10 degrees (Figure 3a), which placed the center of mass of the club 7.3 cm below the golfer’s swing plane (Figure 3b)."

"For both Sim1 and Sim2, the forearm torque generator was set to zero for the entire downswing. In other words, the golfer model was not capable of actively supinating the forearm to square the clubface for impact"

"Simulation 3 (Sim3) was considered the reference condition, as the downswing swing was initiated with the shaft perfectly within the golfer’s swing plane and the optimization was conducted with all the torque generators available to supply energy to the system"

-------------------------------

Doesn't this mean that SIM1 was still able to square the clubface without any active musculature rotation of the forearms?  The forearms must have been rotated by the angular momentum of the club created by that passive torque effect. 

 

 

 

 

 

Edited September 11, 2022 by Warlock

[iacas]

56 minutes ago, Warlock said:

Doesn't this mean that SIM1 was still able to square the clubface without any active musculature rotation of the forearms?  The forearms must have been rotated by the angular momentum of the club created by that passive torque effect.

SIM1 had much earlier ulnar deviation, which I think accounts for almost if not all of the 22%.

When the club is nearly inline with the lead arm, the MOI to rotate it drops significantly, requiring much less muscle activity.

[Warlock]

8 hours ago, iacas said:

SIM1 had much earlier ulnar deviation, which I think accounts for almost if not all of the 22%.

When the club is nearly inline with the lead arm, the MOI to rotate it drops significantly, requiring much less muscle activity.

Yes, that seems to make sense.

Here are the MOI graphs of SIM3 and SIM1

So am I correct in assuming that in SIM3, because the MOI of the 'lead arm/club' unit is smaller about the rotational axis of the torso for a longer period, the 'lead arm/club' unit could therefore be angularly accelerated more by the torso torque generator until the 0.2sec mark  (when the lead wrist starts to uncock)?  That this increased angular acceleration could account for most of that 22% increase in clubhead speed?

Actually, on reflection the above doesn't make sense because the clubhead speed on the bottom graph is about the same for SIM1 and SIM3 at the 0.2 sec mark.

Edited September 11, 2022 by Warlock

[iacas]

54 minutes ago, Warlock said:

Actually, on reflection the above doesn't make sense because the clubhead speed on the bottom graph is about the same for SIM1 and SIM3 at the 0.2 sec mark.

The change in radius may account for that.

Or better said the difference. An uncocked wrist is going to have a much wider radius and thus can match the clubhead speed of a shorter radius with a higher radial speed.

[Warlock]

59 minutes ago, iacas said:

The change in radius may account for that.

Or better said the difference. An uncocked wrist is going to have a much wider radius and thus can match the clubhead speed of a shorter radius with a higher radial speed.

You mean velocity= radius x angular velocity?

An uncocked wrist will also increase the MOI of the 'lead arm/club' unit, therefore its angular velocity will decrease even as the radius increases.

This is getting more complicated