View Full Version : Summary: Baseball throw vs. tennis serve injuries

01-01-2008, 11:52 AM
Here is a summary of the responses we got to our question concerning the
differences in injury rates between baseball pitching and tennis
serving. This question was posted to experts who are not biomch-l
subscribers as well as the biomch-l list. We are posting all responses
here. Thanks to everybody who responded.

Here is the question again:

"[A friend] has noticed that the frequency of shoulder injuries among
elite tennis players is way less than that of his elite pitchers, even
though they all have the same kinematic sequence (serving and pitching
are virtually the same) and their practice number of serves equals that
of the pitchers pitch counts. He feels that this may be due to the
simple fact that the tennis player does not let go of the racquet and
the pitcher lets go of the ball? Does this explain the stress to the
arm or is there another main reason? "

======================summary or responses=============================

>From Roger Fredericks:

I would think that the reason baseball pitchers have more problems than
tennis players is because ALL of the baseball pitchers throws are
overhand with the arm being elevated vertically, with much more
frequency than the tennis player. Yes, the tennis player's serve is
also overhand (with the arm vertical), but they get more balance from
the repetitive stress because they're also hitting millions of balls
with the sidearm motion that they employ in their ground games. I would
think that as a result of doing both overhand and sidearm motions with
the shoulder girdle, the tennis player would have less repetitive stress
on the shoulder girdle than the baseball pitcher, who isn't constantly
just making overhand deliveries. In either case, I still believe that
their hips - along with their shoulders - have to become "tamed" after
their sessions to restore the muscles back to function,

That's my theory.

>From David Ostrow:

I wonder if it has something to do with the centrifical vs. centripetal
force differences that holding a racquet and releasing of a ball....is
there any data on these forces on the rotator cuff at the moment of
release or impact? Somehow, that seem to me to be the plausible
explanation. There could be extra long axis stress on the rotator cuff
via the distraction of the humerus, that I envision upon releasing a
ball vs. the recoil and weight that a racquet puts on the shoulder in
terms of centripetal forces....

I suppose high speed filming could help visualize if there were more LAD
on the humerus with pitching vs. tennis.

Is it also possible that even though the kinematic sequences are the
same that the actual motion and loading are not...

>From Peter Mackay:

Initially it seems that the release of the ball requires more eccentric
counter-force with the rotator cuff but let's look a little deeper!

>From Mike Boyle:

1- Tennis players are not going downhill ( the mound and the downhill
action increases the loads and correspondingly increases the eccentric
force needed to decelerate.
2- Tennis players hit backhand as well as forehand. This may act to
strengthen the posterior musculature in a way that a pitcher could not.

Just two thoughts.

>From Randy Myers:

Does the weight of the racquet equal the weight of a baseball? I'm not
sure, but all of the tennis pros I trained had well developed shoulder
muscles on their serving arm - I assumed it was because they were
carring around a racquet and overtime it served a like a light weight
and got them stronger.

Baseball pitchers develop skills at a similar age, but aren't lifting a
racquet (weight) overhead on a frequent basis. I also think he's onto
something with releasing the ball. Although the shock of hitting a
tennis ball may increase the stability of the shoulder joint. Just some
of my thoughts.

>From Ramsay McMaster:

I think it has a lot to do with position of the cervical spine and head
righting reflexes
ps plus triceps proprioceptive conrtol

>From Dr. Byrd:

For pitchers, the difference may be the mound. Adding the gravity force
of the body to the velocity pushes the shoulder close to the point of
breakdown (or beyond... oops!). That is partly why they need four days
to recover between outings, unless they see limited pitches as a
reliever. Every day position players may have shoulder problems more
comparable to what is seen in tennis. I am not sure, but I could check
the stats with some of the ATP docs, if you like.

The idea of less deceleration is intriguing. Since the time interval
for deceleration is much shorter than acceleration, the force curve is
much steeper, explaining some of the rotator cuff injuries that occur
due to tensile failure during the deceleration phase of throwing, as it
contracts eccentrically to slow the arm. In tennis deceleration is
largely accomplished by contact with the ball. Of course, it is not
that simple, but may be partial plausible reasoning on the subject.

>From Paul Hurrion:

Despite the kinematic sequence (serving and pitching being the same) and
their practice number of serves equals that of the pitchers pitch
counts, the key issue is how the arm decelerates in order to create the
power in the throw or delivery. The same situation can be said for
Javelin (injury prone) and Hammer throwers. Despite the hammer being
heavier (Hammer 7.5kg - Javelin 800gm) there are no shearing forces at
play during the activity. It is these shearing forces that rip the body
apart and do the damage. Hammer throwers release the hammer whilst in
motion and often air-borne. There are no deceleration forces in the
shoulder joints during release, hence they can throw and throw and throw
and do very little damage to there body. However, the Javelin release is
very similar to baseball, it is the stabilization required to aid the
deceleration of the shoulder joints that creates the force to propel the
800gm javelin that creates the strain... A tennis player swings though
the ball, not at it, again like the Hammer thrower with very little
deceleration in the shoulder joint. They are often air-borne at the
point of contact.

>From Gray Cook:

A few questions came to mind since it sounds like the work load is the
same. I wonder about the frequency between pitches and serves. Pitches
are successive in longer packages than serves. There is also lots of non
overhead activity between each package of serves. I also wonder about
arm speeds - are they equal?

I totally think that Tom's observation has some merit. At first glance I
thought there might be something else but then I started thinking about
neurological stuff. Check this out:

Since the tennis player continues to hold onto the racquet and the
centrifugal force on the racquet increases in the arc - a few things
happen that could cause significant "automatic" protection to the
shoulder of the tennis player that my not be afforded to the pitcher.

1) The traction created by holding the racquet my create a reflex
stabilization effect on the scapular stabilizers and cuff muscles. Both
compression and traction stimulate the joint receptors creating enhanced
reflex stability within skeletal muscle stabilizers.

2) The need to hold onto the racquet and the increased traction force on
the racquet in the arc cause a reflex response in the grip. The reflex
grip activation my stimulate improved shoulder stability. Over gripping
is often used in weight training to increase strength and stability
(atomically). We often see grip strength reduced in compromised shoulder
positions and assume it is some kind of reflex inhibition. The same
thing may occur with reflex stimulation of the grip causing some form of
facilitation in tennis that cannot occur with a pitch.

3) Basically a complete release (pitching) my be more stressful to the
shoulder than an incomplete release (tennis). In tennis the load of the
ball is released but the load of the racquet is maintained and may also
assist in help with deceleration.

One or all three of these my serve to help the tennis player - Pardon
the pun!

>From Dave Donatucci:

I have reviewed this question with some of the tennis coaches at the
academy and have an answer.

First according to one of the coaches he does not feel the incidence of
injury is less. He feels there are a lot of overuse injuries on the

second the motion of the serve (Biomechanic if you will) is differnt
than a pitcher. The new pattern of serving is similar to throwing a
baseball or football straight up in the air. Meaning shoulders are on a
60 degree plane with the lead shoulder facing upward (trophy pose), the
hand with the racquet is scratching the back of the person, and upon
hitting the ball the racquet hand is pronated and the racquet is dropped
downward pinting to the ground. Of course the body moves as well but the
shoulder movement is again almost an upward motion, with a racquet drop
at the top and excessive pronation of the hand. The upper arm does not
create the torque or stress on the Ulnar collateral ligament as that of
a pitcher because the hand stays above the shoulder and is not bent
behind the body. The upper arm stays at 90 degrees.

The main differnce is that the arm in tennis finishes above the head and
the body lowers it back to the ground, if done correctly. wheras in
pitching the shoulder musculature has to decelerate the arm.

>From Rob Mottram:

Pitching vs tennis shoulder injuries.

Trying to think about what the differences are between the motions even
though 3D shows they're fairly similar. Dr. Jobe has EMG activity for
pitchers....how about the tennis serve?

Other thoughts...

1. Tennis players don't ONLY serve... whereas pitchers repeatedly do
the same motion. So, I would guess it's a frequency - over use issue.
2. Does striking the tennis ball with a racquet help slow the arm down,
vs the stress and strain of the deceleration of the arm after releasing
the ball, making it lighter, being hard on the posterior shoulder?

3. How about throwing down hill from a mound vs the tennis player being
off the ground and swing slightly upward...making it not gravity


The following responses are from BIOMCH-L - the biomechanics and
movement sciences listserve

Arnel Aguinaldo MA ATC
Director, Center for Human Performance
Assistant Professor, Biomechanics
Department of Exercise and Nutritional Sciences San Diego State

This is a very interesting question that I've been asked many times.
Although we've studied hundreds of pitchers, we have yet to study the
tennis serve so my answer here is partially anecdotal and is my no means

As you mentioned, the kinematic sequences between the tennis serve and
baseball pitch are very similar. In fact, you can make the argument that
most athletic movements that require a sequence of bodily rotations (ie,
soccer kick, football pass, tennis serve, baseball pitch, etc.) have one
common denominator, which is the transfer of momentum from one segment
to another is needed to produce a net effect (ie, throw the ball, serve
the racquet). Some have quoted the "summation of speed" principle to
describe this behaviour, albeit a bit oversimplified, in my opinion.

However, in observing the tennis serve vs. a baseball pitch, there are a
couple ofvmechanical differences, despite their similarities, which
could explain thevdiscrepency in injury rates. One obvious difference is
the mound on which avpitcher has to throw the ball. The regulation mound
is 10"
off the ground
and can play significant role in influencing potentially injurious
stresses at the shoulder and elbow. Although there is currently limited
data on the relationship of joint stress vs. mound height, it is
theoretically possible that raising the height and slope at which the
player delivers a pitch can cause higher rotational torques compared to
that of flat-ground throwing.
The kinematic sequence surronding a pitch is initiated by front foot
contact, in which the reaction force created by the front leg landing on
the front of the mound either from the wind-up or the stretch kicks
start the chain of angular momentum transferring up the body through the
pelvis, trunk, arm, and ultimately, the ball. Since this reaction force
is the net sum of gravitional and transient forces, the height at which
the pitcher "falls" down the mound will partially define the potential
energy that kicks starts this kinematic sequence.

Another notable difference is that the axis about which the trunk and
the arm rotates during a pitch differs to that of the tennis serve. For
the sake of argument, most pitchers with a 3/4 arm slot or lower rotate
their trunks about an axis perpendicular to the transverse plane whereas
tennis players serving primarily rotate about an axis perpendicular to
the sagittal plane. In our most recent MLB study, we found that players
with lower arm slots (ie, sidearm deliveries) exhibited higher valgus
loads at the elbow, even with less elbow flexion, compared to those with
3/4 or overhand arm slots. As a side note, it's been previously
suggested that pitchers who kept their arms "straighter" typically were
less prone to elbow valgus injuries. While it would seem that less elbow
flexion would reduce the moment arm about the long axis of the humerus
(and perhaps subsequently minimize elbow valgus loading), the moment arm
about the trunk's axis of rotation as it squares up increases with elbow
extension, causing a bending moment at the arm that peaks in the early
part of acceleration, particularly in deliveries with lower arm slots.
Since a tennis player would be less likely to serve in the same plane of
rotation as a pitcher, I imagine he/she would be less prone to the same
stresses, both in magnitude and loading rate, at the arm as a baseball
pitcher. (And I played on my JV tennis team in high school so I would
know!). However, it would be interesting to comparatively study the
mechanics of the two to substantiate or refute this notion.


>From John Dewitt:

I don't think that tennis players snap their wrists like a pitcher
throwing a curveball - this snapping motion should place more stress on
the elbow and shoulder

From: Maggie Mae Zerger:

Tennis players hit forehand and backhand. This allows for posterior
development that is lacking in throwing overarm athletes. Overarm
athletes have imbalanced shoulders caused by hyper anterior development
and weak posterior development.

>From Gordon J. Alderink, PT, PhD:
Associate Professor
College of Health Professions
Program in Physical Therapy
Grand Valley State University

Of course, we can speculate "till the cows come home"...it's possible
there are fewer shoulder injuries in tennis because they are generating
torques with an instrument that has a sizable mechanical advantage.

>From Glenn S. Fleisig, Ph.D., Smith & Nephew Chair of Research American
Sports Medicine Institute:

I am out of the office on vacation, so I am doing this from memory. My
colleagues and I have compared the biomechanics of the tennis serve to
the baseball pitch. Here are the citations:

Elliott B, Fleisig, R, Nicholls R, Escamilla R. Technique effects on
upper limb loading in the tennis serve. Journal of Science and Medicine
in Sports 6(1):76-87, 2003

Fleisig GS, Nicholls RL, Elliott BC, Escamilla RF. Kinematics used by
world class tennis players to produce high-velocity serves. Sports
Biomechanics 2(1):17-30, 2003

As I recall, the tennis serve has greater ball velocity, but less
shoulder rotational velocity and torque. I believe that the less
velocity was due to the substantial extra inertia in the distal segment
of the kinetic chain. In other words, the tennis racquet is, in a
sense, an extension of the hand. The tennis player has far more inertia
than the baseball player does in resisting wrist flexion, but the
increased distance of the segment produces greater linear velocity (of
the tennis racquet head, compared to the pitcher's finger tips).

That's the best I can do for now. Happy Holidays,

- Glenn

>From Michael Feltner:

don't know that we have a complete answer to this question based on the
literature and research. My hunch is that baseball is unique in that
the pitcher must decelerate his/her arm after release and bring its
velocity to zero using his/her own musculature, ligaments, etc. In
volleyball, tennis, or other "overarm" striking sports, the impact from
the arm/racquet contact with the ball provides a force that slows the
rotation and velocity of the arm and prevents the muscles/tissues of the
performer from providing all of the forces necessary to stop the arm's
movement. While injuries will always be multifactorial in nature, the
combination of the rapid rates of elbow extension and shoulder internal
rotation coupled with the need to stop these movements after release
primarily by eccentric actions of the elbow flexors and external
rotation muscles at the shoulder, respectively, likely increases the
injury risk in baseball compared to other sports with a similar movement

Just my $0.02

>From Adam J. Bartsch, M.S.
Cleveland Clinic Spine Research Laboratory:

One would suppose the much larger weight of the baseball (145g vs. 55g)
and ball rotation induced by the hand-elbow-shoulder as opposed to
racquet angle would create more torsional and shearing stress on the
shoulder and elbow joints to the baseball pitcher.

One would also question the statement "...even though they all have the
same kinematic sequence (serving and pitching are virtually the
same)..." as being valid? It seems as though this might be an overly
broad statement to apply to two very complicated kinematic operations.


>From Michael Yessis, Ph.D
Professor Emeritus, CSUF
President, Sports Training, Inc.:

I believe a comparison between the baseball pitch and tennis serve is
not 100% possible. The same kinetic change is involved for the basic
body movements, with differences in speed and amount of joint action,
etc., but the arm actions can be very different.

For example, in the tennis serve, the arm is extended overhead and is
fully straightened when contact is made. In baseball pitching, we do
not see the same position. The ball is usually released with a bent
elbow or with an extended arm more to the side of the body rather than

The tennis serve is more limited in regard to wrist and forearm actions,
usually flexion and pronation, while in the baseball pitch, we see these
two actions in the fastball. But when throwing other kinds of pitches,
there may be ulna flexion or variance of wrist flexion (and in some
pitchers, supination).

Thus, to determine which actions cause shoulder injury, it is necessary
to study more closely the exact position and wrist and forearm actions
that occur. For example, I have found that most shoulder injuries in
the tennis serve occur when contact is made with a bent elbow. While
in the baseball pitch, most shoulder/elbow injuries seem to occur when
slinging the ball more out to the side rather than overhead.

===============end of summary======================