Weekly load and multisport training
How individual sessions add up into a weekly picture, how to handle an athlete who runs, rides, swims and lifts in the same week, and why power is not used to calculate load.
This page assumes you are already familiar with the individual session metrics: cardio load, muscular and structural load, and RPE.
The distinction that causes most of the confusion
Coaches working with multisport athletes usually get stuck in the same place, and it is not because load is hard to calculate. It is that volume, load and intensity behave differently when you combine sports — and only one of the three can legitimately be added up.
| Quantity | Question it answers | Combines across sports? |
|---|---|---|
| Volume | How much was done | No — keep it per sport, in that sport's own units |
| Load | How much total stress | Yes — this is the shared currency |
| Intensity | How hard | Per session — read its spread, not its average |
Treating all three the same way is where weekly planning for a triathlete goes wrong. The rest of this page takes them one at a time.
Volume
Volume is deliberately not unified, because no unification is honest. Four kilometres swum and four kilometres run have nothing in common, and even time misleads — three easy hours on a bike is not three hours of running.
| Modality | Primary | Also worth tracking |
|---|---|---|
| Running | Distance (km) | Moving time; elevation gain and loss |
| Cycling | Time | Distance; elevation gain; kilojoules where power is recorded |
| Swimming | Distance (m) | Time; stroke count where recorded |
| Resistance | Time | Tonnage (sets × reps × load) where you track it |
Track these separately and compare each against that sport's own history.
The moment one exists, someone will use it — and it will report that replacing a long run with a long ride changed nothing. It changed a great deal, as structural load shows.
Intensity
Two different things in the app are called intensity, and keeping them apart matters here.
The Intensity metric shown on a workout is derived from that session's cardio load, counting only the segments above the session's average heart rate so that rest intervals and warm-ups do not dilute it. It is time-independent, and it is calculated from heart rate for every sport, cycling included.
Intensity zones are what you prescribe against, and those can be built on maximum heart rate, heart rate reserve, or FTP — so a cycling zone profile can be entirely power-based.
Power and normalised power are read from the file, stored, and shown on the workout. They do not feed the Intensity metric or cardio load. If a ride's Intensity looks low for how hard it was, the heart rate data is what to check — not the power figures.
The common scale
Underneath the Intensity metric is heart rate reserve — where the athlete sits between their resting and maximum heart rate:
heart rate reserve = (average HR − resting HR) / (maximum HR − resting HR)
An athlete at 0.75 is working at three quarters of their usable range, and that means the same thing whether they are running, riding or swimming.
Percentage of FTP is an equally sound intensity scale, and for cycling it is a better one — it responds instantly where heart rate lags, and it is unaffected by heat, fatigue or caffeine. The reason heart rate reserve is the scale used across sports is not that power is inferior, but that it only exists for one of the four. See why power is not used for load below, which turns on the same point.
Do not average intensity across a week
A hard swim and an easy ride average out to "moderate", which describes neither session. Look at the spread instead — how many sessions sat in each band — which is the polarisation question actually worth asking.
Two cautions specific to multisport athletes
Heart rate in water runs roughly 10–15 bpm below land at the same oxygen cost. Horizontal body position improves venous return, water pressure assists circulation, cooling is far more efficient, and the mammalian diving reflex slows the heart directly.
Because cardio load weights intensity heavily, that gap compounds: swim sessions will show a cardio load below what the session felt like. This is a property of heart rate in water and affects every heart-rate-based system equally, so there is no setting that corrects it. Read swim cardio load as conservative.
Heart rate during resistance work spikes within a set and drops during rest, so a session average describes neither state. Duration × RPE is genuinely the better model here.
Which one gets used depends on the data, not on the sport. If the session arrives with heart rate data — recorded on a watch and synced — the heart-rate calculation is used, and the RPE rating will not override it. Read those numbers knowing they are a weaker description of a gym session than they would be of a run.
If the session has no heart rate data, which is the common case for manually added gym work, the RPE rating is what the load is built from. Unrated, it defaults to a low effort and quietly under-reports. So rating those sessions is the single highest-value input a coach controls.
Load — the shared currency
Because cardio load is built on heart rate reserve rather than on any sport-specific measure, it is directly comparable across every modality and can be added together. A run, a ride, a swim and a gym session all produce AU on the same scale, and summing them is valid.
This is the number to use for weekly totals, and everything in the next section is built on it.
Two athletes with identical weekly cardio load can be under completely different strain, because the modalities load different tissue. A swim and a run at matched cardio load cost the cardiovascular system the same and the skeleton nothing alike.
That is what muscular and structural load exist to show, and why the structural ramp below is tracked per sport rather than as a single figure.
Weekly load
These are the figures shown in the Reports section, and in period analysis.
Everything in this section — acute and chronic load, Relative Fatigue, monotony and strain — is computed from cardio load across all sports. There is no muscular or structural equivalent, and that is deliberate rather than an omission. See why the other two dimensions are handled differently below.
The one exception is the structural ramp rate at the end of this section, which is the structural dimension's own tool.
Acute and chronic load
Acute load is the recent training stress an athlete is carrying — a weighted average over the last 7 days. Chronic load is the base they have built — the same calculation over 21 days.
In Reports these appear as Fatigue over 7 days (TFF) and Fatigue over 21 days (TFF).
Both are expressed as an average per day, not as a window total, so they sit on the same scale and can be read against each other directly.
Recent days count for more than older ones. Training three days ago bears on today's readiness considerably more than training three weeks ago, so the average is weighted rather than flat — the most recent day carries full weight, and the weight tapers away toward the far end of the window.
Relative Fatigue (rTFF)
Relative Fatigue = acute load / chronic load, expressed on a scale of ten
This compares what the athlete is doing now against what they are trained for. A value around 10 means this week matches their established pattern. Higher means load is being added faster than they have adapted to; lower means detraining, or a deliberate taper. The commonly cited working range is 8–13.
The published "danger zone" thresholds for acute-to-chronic ratios have not replicated consistently outside the populations they were originally derived in. Read this number as a prompt to look closer, not as a threshold that dictates a decision. An athlete at 14 in a deliberate overload week is doing exactly what was planned.
Training monotony and strain
monotony = average daily load / variation in daily load
strain = weekly total load × monotony
Monotony measures sameness. A week of seven identical moderate days scores high. A week with genuine hard days and genuine easy days scores low — which is what you want.
Strain combines volume with sameness. High total load on its own is manageable; high total load delivered as seven identical days is the combination most associated with poor adaptation. This is why a hard week with real recovery days is safer than a moderate week with no variation at all.
Both are calculated across every sport the athlete did, which is the point — an athlete who runs Monday, rides Tuesday and swims Wednesday at similar loads is monotonous in load terms even though the sessions felt varied.
Structural ramp rate
ramp = structural load over the last 7 days
÷ average of the four preceding 7-day blocks − 1
This is the familiar "don't increase by more than 10% a week" rule, applied to structural load rather than to mileage — and tracked per sport, because tissue adaptation is movement-specific.
An athlete who moves from road to trail in spring keeps their weekly kilometres identical while their structural load rises sharply. A mileage-based check cannot see this, and that population is exactly the one it fails.
Replacing 30 km of running with a structurally equivalent volume of hiking keeps the whole-body total flat while running-specific tissue tolerance quietly declines. A single combined ramp figure would report no change. The per-sport breakdown is what catches it.
Why only cardio load
The acute-to-chronic comparison is not a general-purpose tool that we have simply not got round to applying to the other two dimensions. It is a cardiovascular construct, and its interpretation does not transfer.
The band comes from cardiac research. The idea that a ratio around 1.0 is safe and a high one is risky was established on cardiovascular training load. No equivalent threshold has been established for bone or tendon. Publishing a structural acute-to-chronic ratio would invite you to read it against a line that was never drawn for that tissue — a number that looks authoritative and means nothing.
Structural has a better tool anyway. The ramp rate above applies the ten-percent rule, which is a validated threshold for impact loading, and it is tracked per sport rather than as one whole-body figure. That is a stronger answer to "is this athlete building too fast" than a ratio would be.
Structural also recovers on a different timescale. It clears over days to weeks, where cardio load clears in 24–48 hours. A 7-day acute window against a 21-day base is calibrated to the faster of those. Reusing it unchanged on the slower one would describe neither well.
How to read the other two dimensions instead
Muscular and structural load have their own charts in Reports, and there is a way to read them that works considerably better than the obvious one.
Set the period to Week. All three loads can be charted by day, week or month. Daily is noisy — a rest day reads as zero and a long run as a spike, and neither tells you anything you did not already know. The weekly view is where a build, a plateau or a taper becomes visible.
Put one of them on the same chart as cardio load. Choose Cardio load as the main chart and Structural load (or Muscular load) as the additional one. The two are drawn against separate vertical axes — cardio load in AU on the left, structural load in SL on the right — precisely because the numbers are not comparable.
That is the whole technique: you are not comparing the values, you are watching the shapes. When the two lines rise and fall together, the athlete is loading both systems in proportion. When they separate, something has changed that a single number would have hidden:
| Pattern | What it usually means |
|---|---|
| Structural rises, cardio flat | Volume or terrain added without aerobic stimulus — impact bought without fitness |
| Cardio rises, structural flat | Intensity added on the bike or in the pool — usually safe, but race-specific tissue work is missing |
| Cardio holds, structural drops | Training moved off the feet. Deliberate around a niggle; a problem eight weeks from a marathon |
| Both fall together | A genuine taper or a break |
Weekly totals also appear in the calendar. The weekly summary panel beside each week shows all three loads for that week — cardio in AU, muscular in ML, structural in SL — next to planned and completed duration, so you get all three dimensions without leaving the calendar.
For the progression question specifically, use the structural ramp above. And whichever view you use, compare each metric against its own history — never against the other dimensions, which are different quantities on different scales recovering at different speeds.
Why power is not used to calculate load
A fair question from cyclists, and the instinct behind it is right — power is the best measurement in cycling. The answer has two parts, because power spans both of the quantities discussed above.
Power is both volume and intensity, depending on how you use it
kilojoules = average power (W) × time (s) ÷ 1000 → volume
intensity factor = normalised power ÷ FTP → intensity
Power multiplied by time gives kilojoules of mechanical work, which is a genuine and excellent volume measure for cycling. Power relative to threshold gives intensity. So power does belong in both columns — for cycling.
Why it cannot be the load currency
It measures output, not cost. Two riders holding 250 watts for an hour have done identical mechanical work and paid entirely different physiological prices. Load has to represent the price, because that is what determines recovery.
Only one of the four modalities has it. Running power is proprietary and model-dependent — two devices disagree about the same run — and swimming has no equivalent at all. A currency present in one modality out of four cannot be the one that makes them comparable, which is exactly the problem being solved.
It depends on a threshold that drifts. Power only becomes intensity once divided by FTP, and a stale FTP silently mis-scales every session after it. Maximum and resting heart rate move far more slowly and are easier to verify.
The practical recommendation
Use power within cycling — for prescribing sessions, setting zones, pacing efforts and tracking cycling-specific progression. It is more precise than heart rate for all of those, and it responds instantly where heart rate lags.
Use heart-rate-based load for anything that crosses modalities: weekly totals, Relative Fatigue, and comparisons between a swim block and a run block.
The two are not competing. They answer different questions, and the mistake is asking either one to do the other's job.
Getting accurate weekly numbers
Three inputs account for nearly every case of weekly load that looks wrong.
-
Set a real maximum heart rate in the athlete's Performance Data. Where one is not set it is estimated from age, and because cardio load rises steeply with heart rate reserve, a 10 bpm error moves every figure noticeably.
-
Set a real resting heart rate. It defines the bottom of the reserve range, so an error here shifts every session in the same direction.
-
Have athletes rate sessions with no heart rate data. Resistance work and indoor sessions fall back to duration × RPE. Rated, they are accurate. Unrated, they default to a low effort and under-report — which then pulls down the weekly total, the chronic load and the Relative Fatigue figure along with it.
Summary
- Volume stays per sport. There is no honest combined figure.
- Cardio load is the shared currency and adds across every modality, including strength work.
- Intensity is per session. Read its spread across the week, never its average.
- Acute vs chronic load tells you whether this week fits what the athlete is trained for. It is a cardio load figure — there is no muscular or structural version.
- Monotony and strain tell you whether the week had real variation in it.
- Structural ramp is the injury-relevant one for runners, and it is tracked per sport.
- Power belongs inside cycling. It cannot be the currency that joins four sports together.