ATL, CTL and TSB, the three curves on TrainingPeaks’ load chart, are powerful tools for following endurance training. But their value depends less on the numbers and more on reading the trends, the context and the physiological coherence of what is being done. This article teaches that reading and shows how the 9 / 7 / 5 model shows up clearly on the chart, without touching a single setting.
CTL (Chronic Training Load): the weighted average of daily load (TSS) over the last six weeks. It is your accumulated fitness.
ATL (Acute Training Load): the same average over the last week. It is your recent fatigue.
TSB (Training Stress Balance): yesterday’s CTL minus yesterday’s ATL. It is the balance between the two.
The first mistake: treating the chart as a judge
People new to TrainingPeaks often open the chart looking for a verdict: ATL is too high, TSB looks bad, CTL should be bigger. That is the core mistake: giving the chart the role of a judge. It doesn’t evaluate decisions. It only reflects mathematically what was done.
ATL, CTL and TSB are not targets, nor isolated signs of success or failure. They represent physiological processes with different response times.
The chart doesn’t decide.
It describes.
ATL: recent fatigue
ATL represents the functional fatigue built up over the last few days. It reacts quickly, both when load goes up and when it comes down.
| What to look at | What to avoid |
|---|---|
| How fast it climbs in denser blocks | Reading high ATL as a mistake |
| How quickly it drops when load is deliberately reduced | Comparing ATL between athletes |
| Whether the curve matches what was planned | Deciding based on a few isolated days |
An athlete starts a 9-day accumulation block. Volume and density build gradually. On the chart, ATL climbs steadily, with small spikes on the harder sessions and no meaningful drops. That response is expected and desirable: it shows method and physiology in line, not overload.
ATL is the echo of recent training, not a clinical diagnosis.
CTL: adaptation is read in weeks
CTL is the metric most often misread. It represents sustained exposure to training over time, not athletic ability or talent. It has inertia, it responds slowly, and the slope of the curve matters more than its value.
CTL is not a magic number. Someone with fewer hours a week will naturally have a lower CTL than someone training 20 to 30 hours. That doesn’t mean a mistake, a limitation or a lack of commitment.
Athlete A: 6 to 8 hours a week, consistent and well organized. Athlete B: 20 to 25 hours a week, with a long training history.
Both improve, respond well to training and execute the 9 / 7 / 5 model correctly. On the chart, athlete B’s CTL is much higher; athlete A’s is more modest, but stable and coherent. Neither is right or wrong: CTL reflects sustained volume, not merit.
CTL doesn’t measure quality. It measures load consistency.
TSB: context, not readiness
TSB comes from the relationship between recent fatigue (ATL) and accumulated adaptation (CTL). Many people read it as a synonym for being ready, and that is a mistake.
TSB doesn’t tell you if you’re ready.
It tells you if you’re less tired than you were.
At the end of the 7-day consolidation block, the normal picture is high ATL, CTL rising slightly and a clearly negative TSB. During the 5-day block, ATL drops fast, CTL holds and TSB approaches zero or turns slightly positive. That movement doesn’t create new adaptation: it lets the adaptation you already built show.
How 9 / 7 / 5 shows up on the chart
Applied well, the model is clear on the load chart with no change to the default settings.
| Block | ATL | CTL | TSB |
|---|---|---|---|
| 9 + 7 days | Climbs progressively | Grows slowly | Turns negative |
| 5 days | Drops clearly | Holds steady | Improves |
CTL’s lag is not a flaw in the math. It is exactly what makes the reading of the process safer. If the chart followed every micro-decision, it would lose its purpose.
For the chart to tell the truth
Well-defined zones. The chart only reflects training when the zones match the athlete’s real physiology. Outdated zones distort the load calculation and produce misleading readings. A beautiful chart built on the wrong zones is a lying chart.
Complete data. TrainingPeaks calculates with what it receives. Unlogged sessions, a forgotten watch, inconsistent sensors or incomplete uploads compromise the reading. The more faithful the data, the more accurate the reading and the smarter the planning.
Precision is not wisdom
Reading with precision is understanding what each curve represents. Reading with wisdom is not reacting emotionally to every change. Before any decision, ask:
- Am I looking at the trend or at a single number?
- Does this represent days or weeks?
- Am I adjusting training, or trying to fix the chart?
- What do the body and the context say alongside the data?
The ATL, CTL and TSB chart is not a command system. It is a reading system. Keep the default settings, respect physiology, look after data quality and build the maturity to interpret it.
The chart is a mathematical mirror.
The better the process, the truer the reflection.