It seems like I've been writing about nothing but running lately, so it's past time for a change.
Fortunately, the smart folks at Swim Smooth just posted a great entry in their "Feel for the Water" blog about shoulder pain. They contend that shoulder problems can be caused by having a thumb-first sideways hand entry into the water, due to inward forearm rotation. While this is true in a narrow sense, I believe the sideways hand entry is a symptom, not a cause. Let me explain.
There is far more going on with swim-related shoulder problems than just internal forearm rotation. In particular, if you look at the normal range of motion of the shoulder joint, there is only one point in the freestyle stroke where a hard limit of motion is encountered, and that is during the forward reach, just before/as the arm enters the water, when the arm passes close to the head.
If you keep the shoulder joint open, the upper arm goes wide, meaning the elbow must bend to bring the hand back in front of the shoulder, which makes forearm rotation to a sideways hand entry not only easy, but hard to avoid. Fixing the hand rotation doesn't fix the underlying cause. That is to say, the hand rotation is a symptom, not a cause: The underlying cause is a lack of arm extension.
However, extending the arm fully can result in impingement of the rotator cuff, which then causes the arm to move away from the head, 'opening up' the shoulder angle to reduce rotator cuff impingement. So I would say that a sideways hand entry is an indication of avoiding shoulder impingement. That is, there may be damage long before the thumb position becomes evident.
To get the hand to naturally enter flat, and to ensure maximum reach, the elbow must be nearly locked before the arm enters the water. This must be done in a way that prevents shoulder problems.
How is this possible? Aren't they contradictory goals? Not at all: This is best done by 'throwing' the arm out front, so the associated shoulder rises, combining this motion with a strong final push as the other arm completes its stroke, letting the opposite shoulder drop. Angling the shoulder upward opens the shoulder joint, freeing the arm to become straighter prior to entry, which in turn encourages the hand to face the water at entry. Try it yourself: Extend your arm straight up over your head as far as possible. You can't accidentally point your thumb the wrong way when your elbow is locked.
In other words, never swim with both shoulders 'square' to each other in the water, especially as an arm is entering the water. That's not to say that both shoulders are never square to the body: What it means is that when both shoulders are square to the body, one of the shoulders had better be out of the water! The shoulders become square to the body during the middle of the stroke, a situation that should not be present at the end/start of the stroke.
Some swimmers have trouble raising/dropping one or both shoulders, which in turn easily causes shoulder problems during the stroke. This is easily measured: Standing on a hard surface with the back against a wall, extend one arm straight up with palm forward (back of the hand touching the wall), while extending the other arm straight down with palm against the wall (back of the hand against the back of the thigh), with the forearm traveling slightly behind the hip and next to the buttock. Lock both elbows and try hard to get the hands as far from each other as possible (my lats cramp when I hold this position too long), being certain to keep both feet firmly on the floor. Mark the position of the tip of each middle finger on the wall (I use pieces of tape). Repeat for the other side.
Measure the distance from each of the four marks to the floor. The difference between the heights of the marks for the left arm is the maximum possible swim stroke length for the left side, similarly for the right side. The difference between the heights of the two upper marks is one measure of lateral asymmetry, as is the height difference between the lower marks. Asymmetry is not uncommon, and may indicate a mechanical asymmetry (not uncommon, especially for broken bones set poorly). If there are no known mechanical issues, then asymmetry can indicate a loose shoulder joint or a latent muscle problem.
If there are no known bone or shoulder issues, then asymmetry likely indicates a lack of flexibility and/or a lack of strength in the muscles used to perform the stretch. The easiest way to address such issues is simply to repeat the stretch on a regular basis, and also as part of your pre-swim warm-up.
The next opportunity for shoulder trouble comes during the catch, as the hand descends and the elbow bends, which forces the humerus to rotate within the shoulder joint (not a problem for swimmers having a bent arm prior to entry). Swimmers for whom this motion feels awkward or difficult may 'drop the arm', moving the upper arm toward the front of the torso, which can cause the forearm and hand to either go too deep or to cross the center line, depending on the degree of elbow bend.
For me, a good remedy is to focus on getting the elbow bent while the arm is still extended, forcing the elbow out and slightly down, before applying full power with the lats. I do this using the biceps for a brief moment to pull on the forearm, and the pectorals to pull down the arm, which combine to produce power at the hand and ensure a strong catch. As soon at the upper arm has rotated enough, the full force of the lats can be applied through the mid-stroke.
From its raised position during the entry, the shoulder will naturally drop as the lats contract to provide power through the mid-stroke. When the lat contraction nears its end, it is time for the arm to 'get long' again. Just as we emphasized 'throwing' the arm out front to open the shoulder angle, we must also emphasize 'pushing' the arm 'long' at the end of the stroke, to pull the associated shoulder down, helping open the angle for the opposite arm. I do this by giving the lats a reason to stay contracted by continuing to generate power by contracting the triceps to lengthen the arm through the end of the stroke.
There is another opportunity for shoulder-related trouble at the end of the stroke that primarily affects swimmers who under-rotate: If the swimmer is chest-down in the water (not rotated), then the arm must travel behind the plane of the torso to leave the water, which is an awkward (but not harmful) motion. With a good, strong rotation, the arm will rise closer to the lateral plane of the torso, well within the shoulder's normal range of motion.
But doesn't this amount of torso rotation at the end of the stroke force us to do a dreaded scissors kick? It would if the spine were straight, so it is important to twist a bit at the waist to keep the hips closer to level. As the end of the stroke approaches, instead of thinking about 'lifting the arm', I instead momentarily think about 'dropping the hip'. In my stroke, as my triceps do their final strong contraction to straighten the elbow, I'm also thinking about bringing my hip toward my elbow, contracting the associated oblique and QL, which in turn causes the pelvis to twist.
If this is too much to think about during a stroke (which is often the case for me), then I simply don't kick when my hips aren't level. The cost to levitation and propulsion will be minimal, though the new kick pattern may take some getting used to.
From another perspective, a scissors kick is a good thing! It means you are rotating enough to breathe and make it easy to lift your arm at the end of the stroke. Conversely, the absence of a scissors kick should encourage taking a close look at the rest of the stroke, especially where shoulder limits are involved.
As an ocean swimmer, I don't get a wall push every lap to let my arms and shoulders recover from abuse (or to add abuse by bringing my arms tightly together over my head). It is a wonderful thing to be able to do thousands of sequential strokes without interruption. In the open water, it quickly becomes clear when your form falters: You go off course. Sometimes very badly so. In the ocean there's no black line to watch to keep yourself on course despite poor form, something I've watched many pool swimmers do.
I often swim with a GPS tucked under my swim cap (a Garmin Forerunner 305), and upon downloading the data after a swim it is amazingly easy to see when my stroke is balanced and consistent, versus when it isn't. Of course, we can look up to do a 'spot' to check our course while swimming in the open water, but the need to do so should be minimized, since, even when done perfectly, it is an extremely inefficient maneuver.
I have a drill I do while swimming straight distances of 1 km or longer: Swim a number of strokes, then do a spot to check my bearing. If I'm on course, increase the number of strokes until the next check. If I'm off course, decrease it by an amount proportional to the error. I'm now always able to swim at least 15 strokes between checks, and have managed 50 when the wind is low and the water is particularly smooth. In ocean triathlons I check every 8 strokes, mainly because of traffic.
In a pool, take away the lane markers, then rapidly swim the length of the pool with your eyes closed. I recommend doing this at night with the pool lights off, to prevent cheating. Will you be inside the lane when you finish? (Put swim noodles or helpers across the water to warn you to stop before hitting the wall.) For bilateral breathers, drifting off course is not only a sign of stroke or strength asymmetry, but it can also indicate subconsciously favoring of one shoulder over the other.
I seldom breathe bilaterally in the ocean. How do I stay on course for so many strokes? I'm afraid my answer is ridiculously simple: I swim as if I will breathe on each side for every stroke! That means I always rotate enough to get one eye clear of the water, even if I'm not breathing on that side.
Over-rotation? I think not. Why should a breath stroke differ from a non-breath stroke? Shouldn't a breath stroke be just as efficient as a non-breath stroke? I believe I'm doing precisely the motion needed to avoid shoulder damage while swimming powerfully and staying on course (an important way to be efficient in open water). When I'm wearing a wetsuit, it is also the motion needed to prevent chafing on the back of my neck. As the torso rotates, so should the head, but not the hips (if possible).
The downside is that so much head motion can lead to motion sickness. For me, this faded with practice, and the benefits have been substantial. It also works better with a long-distance endurance stroke. In the pool, the extra head rotation may make it harder to stay in the lane, though this shouldn't be a significant problem if the stroke is truly symmetric.
Bottom line, the motions needed to avoid shoulder injury can simultaneously give you a more powerful and more efficient stroke. Prevention is better than cure. Waiting to see a clear symptom, such as shoulder pain or a mis-pointed thumb, is asking for injury. A long, forceful reach and good rotation not only limit the risk of shoulder injury, but also maximize stroke length, which lets those of us with shorter arms close the gap with the longer-limbed Phelps and Thorpe clones.
My mantra during my long swims is: REACH! in a stretch, CATCH! using the biceps, PULL! with the lats, then PUSH! with the triceps while twisting the hips. And while the hand is wet, the fingers are pointed straight down. After about 250m my body internalizes the overlapping instances of the mantra, and I get to enjoy the day, returning to active stroke monitoring only when doing a spot, and when fatigue or waves makes my mantra falter.
I didn't come up with either my stroke or my mantra on my own. The mantra, and the notion of 'throwing' the arm forward, are both due to attending free seminars by elite athlete, exceptionally fast swimmer (Phelps clone) and TrainingBible coach Jim Vance, who dispenses additional wisdom in his blog. He also has a store where you can buy some of his recorded seminars. I was taught to swim in late 2008 (at age 52) by Bobbie Solomon, a remarkable person and great coach.
I started studying shoulder motion after I awoke with a sore shoulder the morning after one of my 2-mile ocean swims. The interactions between joints, muscles, and the demands of sport-related motion are complex, and it took me a while to figure out not only what was going on, but also what I needed to do about it. Everything fell into place when I attended a second seminar by Coach Vance, after which I was swimming faster than ever, without discomfort (much as I've done with running, but far less difficult).
I'm certainly not a fast swimmer, but I'm able to sustain 1:40/100 yards for hours in a calm water, until I run out of energy to burn. I do this without any shoulder problems.