View Full Version : Sprint Biomechanics

Mel Siff
04-26-2002, 08:37 AM
Would anyone care to comment here on an interesting discussion on sprint
biomechanics which recently took place on another list?


Dr. Peter Weyand recently presented his theories on sprint biomechanics in
Lisle for a day long seminar. He has an extensive background in track and
field. He was a very good collegiate athlete (1:54.56 in the 800 and 14:41
for 5K), and he served as the head track coach at Duxbury High School in
Massachusetts. I mention because some have been critical of Dr Weyand for
not "understanding" the track or speed community, which is inaccurate.

Some also questioned the runners he has been working with, felt he was not in
touch with truly elite athlete, and noted that slight differences which are
inconsequential to an individual with a max velocity of six meters per second
are huge for runners whose top end speed is over twelve meters per second. I
can assure you the elite athlete was very much at the heart of the study.

Dr Weyand presented detailed Power Point slides of his research with Michael
Johnson, when he had Michael hooked up to FitSense monitors at the Penn
Relays. For the Harvard team, the critical factor is that the force in
question is mass-specific (Favge/Wb) so it is the force in relation to body
weight. Now, it is possible that the fast guys like Johnson could be doing
something mechanically odd that would allow them to blast without greater
forces, but Peter believed this was very unlikely.

However, he at first thought this might indeed be the case with Johnson,
simply because he looks so funny running. As a result, the JAP study team
stayed away from a video analysis of Johnson in the earlier phases of the
study, only to do the video analysis on him later.

What was the result? He conformed right on in terms of the expected Favge/Wb
(gained from the ratio of tstr/tc) for his top speed. In a nutshell, Johnson
stayed airborne for a larger fraction of the stride more than anyone they
tested or the other guys they did video on (Greene, Bolden, Fredericks), and
his top end speed was faster.

Data on MJ confirmed their hypothesis regarding ground support forces.
Further, it became clear that, if the fastest athletes in the world could
swing faster, they would. There's a limitation here: the limit or barrier to
speed occurs when ground impulses (in this case = vertical force x ground
contact time) decrease to provide the minimum time the runner needs to
reposition the limb. Further increases in speed would decrease contact times
and vertical impulses. These reductions would decrease the time available to
reposition the swing limb.

What have I come away with? Well, it is clear to me that much of the
locomotion research that's been done at the Concord Field Station for the
past thirty years has never reached the coaching community. Those who
attended the seminar feeling that Peter just didn't understand top flight
sprinting I hope left with a far different opinion. Here is one of the
points from the seminar:

"I would suggest beliefs in the training value of pawing, clawing, and
dorsi-flexing are not well supported by the contemporary scientific
understanding of either the evolution of the musculoskeletal system for
locomotion or established principles of the mechanics of running. Runners
such as ostriches and race horses can obviously outperform the fastest human
runners with great ease. However, in many cases the limbs of these animals
preclude movements equivalent to human dorsiflexion. Quarter horses have
covered 400 meters in less than 20 seconds. Clearly, they accomplish this
without any conscious effort to modulate form or any training in how to do
so. Generally, horses, big birds, and virtually anything that runs and has
been studied conforms to general mechanical and energetic patterns, and all
the experiments implementing deviations from those patterns indicate that
function is compromised."

In simple terms, the body loads the limb, not the other way around. All the
pawing, clawing that could be done have no effect on the net requirement for
force or the force that is ultimately generated.

The pawing and clawing ideas are also at odds with the spring model. A spring
by definition is a passive system - it only gives back what goes in, and the
idea that slapping the leg backward and down faster via training indicates or
at least implies that this is an active process achieved with motor control.

Many may have misinterpreted or confused issues related to contact time,
contact length, impulse, aerial time, etc. In a nutshell, here's the point:

"More force -> more aerial time -> longer, more casual swing - > the runner
is not required to use his minimum swing time until a greater speed is
reached -> his top speed becomes faster."

Again, it's hard to do justice to the material through e-mail postings,
especially since many have not read the complete study, and none have had the
opportunity to visit the Concord Field Station to review the work and the
procedures as I did.

Therefore, I thought the clinic was a great opportunity to shed some light on
an issue important to our sport.

As I said to another list member:

The research presents the possibility, and perhaps probability, that swing
mechanics, like ground mechanics, are already optimized, particularly in
trained runners.

If this is the case, then we should reconsider whether the focus of sprint
training should be on strength training, drills, etc aimed at
improving/increasing ground force in relation to body weight, or on coaching
the form of leg swing mechanics.

In my opinion, the most valuable study and data set to address this would be
pre- and post-training data on a large body of sprinters over a period when
they get appreciably faster. This data set could address the basic and still
unknown question - what mechanical factors change as sprinters get into
better speed shape and therefore sprint faster?

The possibility exists that swing times do decrease and that this might be an
important part of training-induced increases in sprint performance. This
finding would certainly make sense to a whole lot of current speed

However, for now I'd have to agree with the locomotion experts, and say that
there is just too much compelling evidence to the contrary.

If active power were the mechanism of limb repositioning rather than a
largely passive process involving energy transfers and tendon springs, then
the Harvard team would have found that sprinters with faster muscle fibers
and more power would reposition their limbs more rapidly - with no other
complicating factors.

In the above scenario/interpretation differences in fiber speed would
correspond one to one with minimum swing times. Clearly this is not the
reality. What the data shows is that minimum swing times are similar because
the repositioning process is largely passive and therefore independent of
fiber speed.

I've been a high school track coach for the past twenty-eight years. My
intention in all of this is to simply further our craft, and to try explore
any new and exciting concepts that might really help me to get this "sprint
thing right," at least before I retire.



I read with interest your statements. However, I still have a major problem
understanding where any of the findings in the Harvard study have application
to running training. I have no problem with the creation of ground forces.
This is, of course, important and must occur for a runner to have any type of
speed. However, I would like to see some practical recommendations utilizing
the "findings" that they uncovered.

For example, the results on Michael Johnson are interesting but not
outstanding. The question that I would like to see answered is this (as I
posted in my statement last week) - if those who run faster take as long to
reposition their legs as those who run slower, do not the faster ones cover
more ground? Thus, even though the time is the same, the distance that each
stride covers is greater and as a result, speed is greater. Total time for
repositioning however can be the same as they discovered. If they closely
checked the angles in the hip and other joints, we would see a greater range
of motion to show that the legs traveled through a greater distance even
though the amount of time was the same.

I disagree very strongly with the statement that the training value of pawing
is not supported by scientific understanding, etc. If they studied the
animals that were mentioned, they would see that the success to their running
speed was due to the pawing action. It is truly amazing how they can miss
this very obvious point. I would agree however, that the dorsi flexing is
not well supported. As far as I am concerned, this happens automatically and
is not a conscious action by the athlete. Also, to say that the animals
accomplished the running without any conscious effort - well, so do humans.
In the learning, however, you have some thinking, but not in the final
performance, it is automatic. Through evolution, man no longer perfects his
running and other skills - this must be learned!

Now here is the $64,000 question. It is stated that the body loads the limb,
not the other way around, although, I can't see how a limb can load the body.
But, it is the limb that makes contact with the ground. In order to have
ground reaction forces the limb must make contact with the ground. Thus, if
pawing is not an action used to generate this force, what is? Is it a
passive touchdown with the foot? In this case, touchdown should be well in
front of the body, not directly underneath the body, which, by the way, is
accomplished by pawback. If it isn't, please describe the actions that occur
with the legs to get into this position at the moment of contact.

The pawing action is not at odds with the spring model. If it is at odds
with it, then the spring model used is strictly a vertical phenomenon; but,
you do not run up and down, you run horizontally. It seems the passive and
active processes are confused here. It is an active process, but once the
foot makes contact with the ground then the spring model takes over, not
prior to. There are two distinct actions here, not only one. It should also
be noted that on the landing, where there is creation of tension in the
muscles and tendons, the return is not passive. It must be a volitional
contraction exhibited in the push-off. A spring will automatically return to
its original shape, but the body will not. It must be made to do this.
Thus, it is very active spring model not a passive one.

It is also interesting how they never brought out what the unique feature of
Johnson is. He does something much better than other runners to create the
arched bow effect during the push-off and it is not force production -
although it is related.

I can appreciate you being impressed with the study and the findings from
visiting there and from the clinic; however, I have not seen any sound
explanation of the results. Instead, I see some glaring omissions.


Dr Mel C Siff
Denver, USA

To unsubscribe send SIGNOFF BIOMCH-L to LISTSERV@nic.surfnet.nl
For information and archives: http://isb.ri.ccf.org/biomch-l