The week before the week that breaks you

Ask anyone who has lost a season to an overuse injury when it started and you will get a date. The Tuesday of the tempo. The wet descent. The morning something in the shin said no.

The date you noticed pain does not tell the whole story. Your recent training and symptoms can add context, but a record of wearable data cannot establish when an injury began.

Injuries accumulate

Injuries have multiple contributing factors. Training exposure and previous injury can matter, but there is no single data pattern that explains every injury or reveals how much load a tissue can tolerate today.

A session may be where pain becomes noticeable. That does not establish a single cause.

Changes in your data are reasons to look more closely, not evidence that an injury is approaching.

What drift looks like

Your resting heart rate creeps up two or three beats and stays there. Not one morning. Nine of them.

Your HRV settles a bit lower and stops bouncing back on easy days the way it used to.

Your sleep gets slightly worse in a way you would describe as normal. You wake once or twice, you are still getting seven hours, and you file it under fine.

Your pace at a given heart rate slides. Ten or fifteen seconds a kilometre on easy runs, which you put down to the heat, or the route, or being a bit tired.

You start rating your easy runs a point higher than you used to, without noticing you are doing it.

And somewhere in there you mention to someone that your calf has felt tight for a while. Not sore. Tight. You stretch it and move on.

These examples can reflect ordinary variation, measurement error, illness, stress or training. Reading them together can suggest useful questions. It cannot identify an injury process or rule one out.

Why weekly totals cannot see it

A weekly total describes training dose. A workload ratio compares periods, but Trailwise does not use it to predict injury or to declare a safe training zone.

A weekly total cannot see that you hit the same 70 kilometres with two more hours a day on your feet at work, four hours less sleep across the week, and a cold you decided was allergies. Training load was flat. Total load was not.

It cannot see the shape inside the week. Seventy kilometres front loaded with two days off at the end is a different stimulus from seventy spread evenly, and both look identical in the total.

It cannot see the interaction, which is the part that actually matters. Sleep sliding on its own is a bad week. Sleep sliding while load climbs is a different animal, because the thing that was supposed to absorb the load is the thing that is degrading.

And it certainly cannot see the calf you mentioned once and then stopped mentioning.

The pattern lives in the relationship between streams over a couple of weeks. That is the shape of thing a person is bad at holding in their head and a system is rather good at watching.

When a change deserves attention

Trailwise has no validated seven-to-ten-day injury-warning window. Its flag lists risk factors and the available evidence. A 2025 observational study of runners linked unusually long individual sessions with injury rates; it did not validate Trailwise, establish causation or create a countdown to injury.

The next step depends on the symptoms and context, not on a fixed countdown.

Review the planned dose, including duration, intensity and terrain. There is no universal rule that cutting intensity first prevents an injury. Pain or an unresolved problem may call for stopping a session and getting advice rather than choosing a different workout.

Do not keep running just to preserve a routine or a stream of data. The appropriate activity and any return to training depend on the problem and, where needed, clinical advice.

Deal with sleep first if sleep is one of the streams that moved. It is the cheapest intervention available and it sits upstream of most of the others.

Review changes as the situation develops. Do not wait ten days to act on persistent, worsening or concerning symptoms.

And write the niggle down. “Tight calf, left, three out of ten, second time this month” is the difference between a pattern and a vague memory. The niggle you noted in week two is what makes the drift in week four legible.

Where this stops being true

Trailwise does not calculate your probability of injury. Signals can change for many reasons, and a clear flag does not mean low risk. Your symptoms and the limits of the available data remain part of every decision.

Some injuries give no warning whatsoever. You roll an ankle on a root. No dataset was going to save you.

And the signals are only as good as your coverage. If your device reports overnight heart rate and nothing else, there is simply less to read, and the honest move is to say so rather than pretend the picture is complete.

A runner balancing training, poor sleep and a recurring niggle may benefit from seeing that context together. The purpose is to make the next conversation clearer, not to claim that every problem can be caught before it happens.

A useful coach helps you decide what to ask, what to review and when to seek help. It cannot promise how the story ends.

Trailwise is not a medical device. Neither is it a replacement for a doctor or a physiotherapist. Trailwise flags what is worth a conversation. It does not diagnose, and it will tell you when it does not know.