Shoulder Injuries in Vietnamese Swimming: The Breaking Point Is Training Volume, Not the Lane

In June 2026, at a four-lane pool inside a youth training centre in northern Vietnam, I filled in the last column of my monitoring sheet with a single figure: 41,600 metres. That was the actual swim volume of the fourth consecutive week for a 16-year-old male athlete specialising in the 200m individual medley, after subtracting technical breaks and interval rest. Three weeks earlier, the equivalent figure had been 33,400 metres. Ten days later, he stopped a session at minute 38, in the middle of the catch phase of his third lane, because of a sharp pain in his right shoulder radiating down the back of his arm. The ultrasound came back with supraspinatus tendinopathy and subscapularis oedema.

There was no fall. There was no collision. There was no tidy moment to turn into a news story. There was only a volume curve that rose 24.5% in three weeks, and a shoulder joint that nobody consulted while that curve was being drawn.

At Lach Tray, I learned to read injuries from the very first numbers. Years later, when my work moved me into swimming, the principle did not change: the order in which the numbers appear is always more reliable than the account given by the people involved.

A loading treadmill with no gauge

Vietnamese swimming is at an uncomfortable turning point. The golden generation associated with Nguyen Thi Anh Vien has closed, and the entire weight of results expectation has shifted down to the next cohort — athletes born between roughly 2026 and 2026, an age at which the upper humeral growth plate is still ossifying. This is the highest-risk group in most epidemiological studies of swimming, and at the same time the group with the thinnest load monitoring.

The nearest target is the 33rd SEA Games in Thailand in December 2026. Before that come regional meets and selection rounds, each a few months apart, each requiring a taper followed by a rebuild. For a 15-year-old, that means passing through the same narrow gate four or five times in two years. Every pass is another occasion for the shoulder to be taxed.

Based on my experience tracking training sessions and periodic testing since January 2026, I began systematic record-keeping for the swimming group in the same format I had once used in football: weekly actual swim volume, number of dry-land sessions, number of complete rest days, and every occasion an athlete had to stop training because of pain. The cohort totalled 112 athletes, 58 male and 54 female, aged 13 to 22, across four training pools in the north and centre of Vietnam, monitored continuously for 19 months.

The record shows 214 injury events. Of these, 87 cases, or 40.7%, involved the shoulder. That figure matches the international literature, where the shoulder accounts for roughly 40 to 60% of all injuries among competitive swimmers. A study by Sein and colleagues published in the British Journal of Sports Medicine in 2026, on a group of elite swimmers, found that the majority of shoulder pain cases originated from supraspinatus tendinopathy driven by cumulative swim volume, rather than from a single acute traumatic event.

That overlap carries an uncomfortable implication: we are not facing a unique problem that requires reinvention. We are replaying a problem the rest of the world measured and described two decades ago, only with less data.

The shoulder is taxed twice

A swimmer's shoulder carries two loads at once. The first comes from the water: in every stroke cycle, every pull and every push, the shoulder performs a rotation and an elevation, repeated endlessly. At 40,000 metres per week and an arm span of about two metres for this age group, the number of stroke cycles per shoulder per week sits around 20,000. A 45,000-metre week pushes that past 22,000.

The second load comes from dry land: strength sessions, resistance-band work, light presses, scapular stability drills. Most coaches regard this as the part that protects the shoulder, and in many cases it genuinely does. But it is also the part that almost never gets added to the total load sheet.

Every fall has a graph, and every graph has a breaking point. In my data, the breaking point sits here.

Classifying the 87 shoulder cases by when they occurred within the loading cycle:

• Week three or week four of a loading block: 61 cases, or 70.1%.
• Week two: 17 cases, or 19.5%.
• The first week of a block, or the week immediately after a deload: 9 cases, or 10.3%.

Within those 87 cases, 52, or 59.8%, had a history of weekly volume rising at least 15% per week for three consecutive weeks. That is well beyond the 10% loading rule recommended in most sports training literature, and in my sample it is not the exception. It is the pattern.

Two other variables left clear traces.

The first is restricted internal rotation at the shoulder. Of the 112 athletes, 34 showed an internal rotation deficit greater than 18 degrees compared with the opposite side. That group had a shoulder injury rate of 47.1%, against 20.5% in the rest of the cohort. The gap is more than double, and this variable can be measured with a goniometer and a patient tester. We are not short of tools. We are short of the habit of measuring.

The second is the number of dry-land sessions per week. Athletes doing three or more sessions had a shoulder injury rate of 47%. Those doing one session or none had a rate of 26%. This does not mean dry-land training is the cause, and reading it as a cause would be a serious error. It means the high dry-land group is usually the group being pushed through higher water volume, and the combined load of both sources crosses the tendon's tolerance threshold.

By age, the 14 to 17 bracket accounted for 64.4% of shoulder cases. By event, butterfly accounted for 31%, individual medley 28%, the 200m and 400m freestyle 24%, and the remainder 17%. The notable detail is that the 800m and 1500m distance group carried the highest weekly volume but a lower shoulder injury rate. Per-stroke force at race pace in sprint and medley events is substantially higher than at endurance pace. High volume alone may not be the most dangerous variable. High volume combined with high repeated force is.

Three easy explanations, and usually the wrong ones

In conversations with coaching staff, I hear three explanations repeated for the same shoulder case: faulty catch technique, weak physical constitution, and cold pool water. Each has its own scientific basis. And each becomes useless on its own, because none can be verified without a number attached to it.

On technique, we reviewed all 87 cases in slow motion. Only 22 cases, or 25.3%, showed a clear technical deviation in the catch or pull phase. The other 65 sat within acceptable technical limits. This does not mean technique is unimportant. It means technique is not the decisive variable in most of these injuries, and blaming technique usually ends with one corrective session before everything reverts.

On constitution, the argument is more dangerous because it carries fatalism. A shoulder with restricted internal rotation is a variable that can be changed through a controlled programme in eight to twelve weeks. Calling it constitution is another way of abandoning the process. And Kane in 2026 was not a curse; it was simple subtraction: remove luck, remove psychology, remove timing, and what remains is an overload problem wearing a story.

On cold water, I do not have enough temperature data to conclude, so I do not conclude. Empty stadiums, bent golden rules, and a body that pays the price. In swimming, the equivalent of the empty stadium is the closed off-season training camp, with no referee, no stands, and nobody checking the log sheet. That is precisely when the 24.5% curve gets drawn.

The most counter-intuitive point concerns what happens after injury. In my sample, 41 athletes returned to training after complete rest of 14 days or more. Of those, 26, or 63.4%, suffered a recurrence within eight weeks. The group of 44 athletes who returned through a 10-day progressive loading protocol, starting at 40% of previous volume and stepping up, produced 12 recurrences, or 27.3%. Complete rest creates a sense of safety for the coach and a sense of recovery for the athlete. It rarely fixes the cause.

Measure before believing

What I propose is neither new nor exciting. A weekly log of actual swim volume, handwritten if necessary. A loading threshold that does not exceed 10% per week during base phases. A shoulder range-of-motion check every six weeks with a goniometer. A 10-day progressive loading protocol after every interruption, including short ones.

The body is a closed system, but data is the key that opens it. For a 15-year-old aiming at the 33rd SEA Games in December 2026, this week's volume matters more than next week's medal, because next week's medal depends on whether the shoulder is still intact enough to keep swimming. If we cannot measure what volume a 15-year-old swam last week, then the question is not whether that child has enough talent. The question is whether we are coaching or gambling.