Decoding Swimming Injuries: The Load Table, the Swimmer's Shoulder, and the Fracture Points Nobody Sees

At Lach Tray, I learned to read injuries from the very first numbers. But it was only when I left the football pitch to sit for hours in front of a swimming lane's dataset that I realized there are sports where the human body is not defeated in a single moment, but worn down across thousands of identical repetitions. A professional swimmer performs somewhere between 9,000 and 16,000 arm strokes per shoulder in a single heavy training week. Multiply that across a year, and the shoulder joint becomes the most common fracture point in the aquatic sport.

Every fall has a graph, and every graph has a breaking point. In swimming, that graph is not drawn by collisions, but by volume. And the breaking point, in more than half of recorded cases, sits in the shoulder. That is not a curse, not the price of talent, but an overload equation that can be solved by subtraction — if we are willing to read the right number.

Context: A sport that runs on repetition

In football, I grew used to injuries being tied to an event: a tackle, a turn, a loss of balance in the 88th minute. In swimming, no such moment exists. The injury arrives on an ordinary Tuesday morning, mid-set, when the swimmer feels a slight twinge in the shoulder but still completes the full 8 km on the schedule. By Thursday, the pain is impossible to ignore. By Sunday, a medical examination records a supraspinatus tendon injury.

The problem is that swimming is one of the sports with the highest rates of overuse injury among all Olympic disciplines. Epidemiological studies published in Western sports-medicine journals over the past two decades agree on one point: roughly 60 to 80 percent of injuries in swimmers are overuse injuries, not acute ones. That means most injuries do not happen because of a single mistake, but because a process was violated over a long period.

The training volume of an elite swimmer far exceeds an outsider's intuition. A national-team swimmer may cover 40 to 70 km per week during the accumulation phase, equivalent to running a long distance every day with no rest for the joints. On top of that, they add 6 to 10 hours of dryland work per week — stretch cords, weights, high-volume upper-body work. And most of that dryland volume lands directly on the shoulder joint.

In Vietnam, the context has its own nuance. Swimming is a sport where we hold high medal expectations at regional level, yet dedicated sports-medicine infrastructure and staff for swimming remain thin compared with the major Asian centers. A Vietnamese swimmer often has to reconcile international-level training volume with locally-resourced recovery. That gap is exactly where injuries breed.

Core analysis: The load table of a shoulder joint

Let us start with the number most people skip. In a standard freestyle stroke cycle, the swimmer's shoulder passes through four phases: catch, pull, push, and recovery. Each phase places the shoulder in a different position, and the pull phase — shoulder internally rotated and elevated — generates the greatest compressive load on the supraspinatus tendon. A swimmer covering 8 km of freestyle per day, at roughly 18 to 22 stroke cycles per 50 meters, performs about 3,000 to 3,500 cycles per day, meaning about 10,000 overhead shoulder elevations per week per side.

That is the number I want you to remember. Ten thousand. Each one compresses the supraspinatus tendon under the acromion. Each one is a micro-injury. A healthy body can repair these micro-injuries during rest. But when volume exceeds recovery capacity, the sum of thousands of small micro-injuries becomes a major injury.

Numbers are silent, but their sequence always tells a story. The sequence here is clear: a swimmer's shoulder injury does not appear in a single heavy swim, but at the intersection where volume rises and recovery time falls. In the monitoring data I collected across training cycles, a repeated pattern emerges: six weeks of load increase before a major meet, accompanied by cutbacks in light dryland and recovery sessions, tends to produce a cluster of shoulder injuries concentrated within two to three weeks after the meet ends.

I call it the post-peak payment effect. The body does not complain under peak load, because the central nervous system can draw on reserves during a short competition phase. The body complains when peak load ends and reserves run dry. Injuries appear after the race, not during it.

The table I build for a swimmer's shoulder looks like this, reduced to measurable variables:

First, weekly swimming volume in km. This is the most basic variable. The warning threshold in my data sits at an increase of more than 15 percent over the previous four-week average. A 10 percent week is normal training. A 25 percent week is a call for the medical team to step in.

Second, the ratio between freestyle and non-freestyle distance. Breaststroke and butterfly load the shoulder differently from freestyle. When a program increases the share of these strokes during accumulation, shoulder injury risk rises, because the attack angle of the supraspinatus tendon changes. This is a detail that simple monitoring tables often miss.

Third, upper-body dryland volume. This is the most underrated variable. A high-intensity stretch-cord session can place compressive load on the shoulder joint equivalent to many hours of swimming, yet it is not logged in the pool diary. Many shoulder injuries I tracked did not originate in the water, but in the weight room, only surfacing in the first swim session afterward.

Fourth, recovery time between sessions. This is the hardest variable to measure and the most decisive. The body repairs tendons not while training, but while resting. When two heavy sessions sit within 24 hours of each other, repair capacity drops sharply, and injury accumulates faster than recovery proceeds.

The body is a closed system, but data is the key that opens it. The four variables form a closed loop. You cannot raise one without paying somewhere else. And injury is the price of deliberately breaking that balance over a sufficient length of time.

The breaststroker's knee: A forgotten breaking point

If the shoulder is the most common breaking point in swimming overall, the knee is the characteristic breaking point of breaststroke. The breaststroke kick — knee rotated outward and flexed with a lateral drive — places stress on the medial collateral ligament and medial meniscus. Repeat this thousands of times in a week, and you have an injury with its own name in sports medicine.

In my data, the breaststroker's knee follows a different pattern from the shoulder. Shoulder injuries appear gradually and give warning. Breaststroker's knee injuries tend to appear more suddenly, often during a phase of increased kicking volume or when a swimmer moves from recreational breaststroke to competitive breaststroke with high-amplitude kicks. And here is something few notice: a breaststroker with a knee injury can often continue swimming freestyle and backstroke without pain, which leads them to underestimate the severity of the damage.

In one monitoring cycle I ran, a young breaststroker felt knee pain during a kicking set, but because she could still swim freestyle pain-free, she continued full-intensity training for two weeks. By the time the pain no longer allowed her to perform a kick at any intensity, the injury had progressed to tendinitis and prolonged synovitis, requiring four months of recovery. Had the knee data been read correctly in week one, the rest period might have been ten days.

This is the point I want to stress methodologically: an injury is not only a medical event, but a chain of decisions. Every decision to keep training through a warning signal adds a link to the chain. And the longer the chain, the greater the final cost.

Backstroke, butterfly, and the burden on the lower back

A third breaking point I often see but rarely discussed is the lower back. Butterfly and backstroke place the lower back in constant extension. Butterfly in particular compresses the lumbar spine during the body-wave phase. A competitive butterfly swimmer may perform hundreds of body-wave cycles per session, and the lower back pays for each one.

In the data I recorded, lower-back pain in swimmers is not rare. It is often mistaken for dryland-induced back fatigue, and therefore ignored. But when imaging is performed on butterfly swimmers with persistent symptoms, the rate of structural abnormalities detected — mild disc degeneration, facet joint changes — is not small. This is a silently accumulating injury, and it produces no clear symptoms until it has reached a difficult-to-reverse stage.

Here, my experience tracking many training cycles yields a simple but hard-to-accept conclusion: most swimmers with lower-back injuries are not injured because they swim too much, but because they perform upper-body dryland work without proper control. A poorly executed sit-up, repeated hundreds of times, can cause more damage than a 6 km swim.

The ten-day protocol: When patience becomes a weapon

During the pandemic period, I once proposed that a club adopt a ten-day progressive load protocol for athletes returning after a break. The coach refused, wanting immediate results at the first meet. By the end of the cycle, the non-compliant team lost a significant share of its roster to injury, while the compliant team remained intact.

I tell this story not to praise myself, but because it illustrates a principle anyone working with sports data must accept: a load-progression protocol is not a slowness, but an investment. In swimming, a ten-day protocol for an athlete returning after a break can look like this.

Days one to three: volume at 40 percent of normal, low intensity, freestyle and backstroke only, no upper-body dryland. The goal is not training, but reassessing water feel and latent pain points.

Days four to six: volume at 60 percent, introducing short interval sets, adding breaststroke and butterfly at low doses. Light upper-body dryland, focused on movement control rather than load.

Days seven to nine: volume at 80 percent, rising intensity, beginning short speed sets. Upper-body dryland returns to half of normal.

Day ten: volume at 100 percent, not yet at maximum intensity. Then a week of stable maintenance before returning to the usual program.

Notably, this protocol does not lengthen the rest period. It only lengthens the time to reach peak load. And in my data, an athlete who goes through the ten-day protocol has a significantly lower injury probability over the next three months than one who jumps straight into the old program. The difference is not fitness, but whether the body was given time to adapt to a load from which it had been detached.

The contrarian angle: The culture of endurance and the trap of haste

A common belief exists in swimming, and it is present in many countries, not only Vietnam: injury is part of the path, and the strong are those who endure. The saying sounds like a mental quality, but when checked against data, it is a mistake in risk management.

The issue is not that athletes should avoid all pain. In elite sport, fatigue and mild pain are part of the process. The issue is that the culture of endurance often fails to distinguish normal physiological fatigue from the signal of structural damage. And in an environment where reporting pain is seen as a sign of weakness, athletes tend to stay silent until a small signal has become a major injury.

This is why I always emphasize data over narrative. A swimmer may say the pain is "just normal fatigue," and that may be true. But when range-of-motion metrics drop, when measured stroke force falls, when recovery time between sessions lengthens, there is an objective signal that the story is not as simple as the telling. Data does not replace the athlete. Data only ensures that the decision to keep training is made on a foundation of full information.

And here is the second contrarian point. When an athlete is injured and returns, the most dangerous thing is not the injury itself, but the speed of return. The re-injuries I tracked did not usually occur because the original injury was too severe, but because the athlete returned to full load too soon relative to the adaptive capacity of the damaged tissue. In swimming, tendon injuries are especially dangerous here, because tendons recover more slowly than muscle and show no clear symptoms until re-injury has occurred.

I once compared two groups of athletes returning from shoulder injury. The first returned to full programming within two weeks of pain resolution. The second went through a four-week load-progression program. Over the following six months, the re-injury rate in the first group was significantly higher. The specific number varied by monitoring cycle, but the trend was stable. Rushing back does not save time. It only shifts time from the initial recovery phase to a longer re-injury phase.

Empty stands, the golden rule bent, and the body paying the price. In several periods I monitored, periodic medical screenings were cut for various reasons, and those periods had the highest injury density. Not because training volume increased, but because the monitoring mechanism was loosened. When no one watches the number, the number still runs; it is just that no one reads it.

What actually changes the picture

If there is one lesson I have drawn from years of working with injury data in both football and swimming, it is this: the picture does not change through a grand invention, but through patiently maintaining a small process. In swimming, the following three changes have greater impact than any new drill.

First is a structured pool diary. Not a notebook of feelings, but a table recording volume, intensity, stroke ratio, and shoulder and knee status on a simple scale. This takes a few minutes a day but creates a data series that allows a breaking point to be detected before it manifests as injury.

Second is a load-increase rule. No weekly volume increase exceeding 15 percent of the previous four-week average. This rule needs no technology, no expensive equipment, only discipline. But in practice, it is often broken under meet-schedule pressure.

Third is separating dryland and water training in monitoring. This is the change I consider most important, because it fixes a major blind spot in common awareness. Many still think swimming injuries come from the pool. My data says the opposite in a significant proportion of cases: they come from the weight room, and only surface in the water.

Career impact: What cannot be recovered

A shoulder injury is not just an interruption. For a swimmer in a development phase, it is a loss of time that cannot be regained. In a four-year cycle toward a major stage, a six-month shoulder injury can mean losing an important competition season, losing a qualifying spot, losing a phase of building a fitness foundation. And for an athlete at the peak, it can be the end of the ambition to reach the top.

But the deeper impact lies on the other side of the career. A shoulder joint injured repeatedly in youth does not return to its original state. It becomes a shoulder with a history, and that history follows the athlete for life. In long-term data, athletes with multiple youth shoulder injuries tend to shift toward strokes that demand less of the shoulder, or to shorten their elite competitive career. Their career picture is shaped by decisions made in weeks no one recorded.

That is why I believe in data not because data is clever, but because data does not forget. Human memory is selective. A swimmer's memory tends to record successful competition days and erase training days that carried warning signals. But the data table does not erase. It simply sits there, waiting for someone to read it.

Hai Phong, water, and non-repeating milestones

I have said before that Hai Phong, a port city, and three milestones in my career taught me that injuries never repeat identically. Each athlete is a system of their own, and each injury is a unique configuration of the same overload equation. No universal formula applies to all. But one principle is universal: injury does not begin in a moment, but in a sequence.

And if injury is a sequence, it can be cut at any link. Cut early, and it is easy. Cut late, and it is hard. Cut after the sequence has completed, and there is nothing left to cut.

That is all an injury analyst can say to a coach at the start of a season. You can choose to read the number, or you can choose to wait until the body reads it for you. But the body reads the number in its own way, and that reading is never gentle.