Staring at a ride file full of numbers? Bike power data looks complicated, but a handful of metrics do most of the work: FTP, normalized power, intensity factor and training stress score.
This page sits inside our wider guide to training with power, and focuses on reading the numbers after the ride.
| Item | Value |
|---|---|
| FTP estimate from a 20-minute test | Average power for the 20 minutes, less 5 percent |
| Example: 250 W test average | 238 W FTP |
| Normalized power for three equal blocks at 100, 200 and 300 W | Approximately 239 W, against an average of 200 W |
| Intensity factor example (210 W NP, 280 W FTP) | 0.75 |
| TSS for an all-out 60-minute workout | 100 TSS |
| Variability Index for steady riding | 1.05 or less |
The figures come from TrainingPeaks’ power terminology guide. Definitions and examples can change, so check the source before you rely on them.
Start with the basics: watts and average power
A power meter reports watts, the rate at which you push the pedals. Unlike heart rate, power responds quickly to changes in effort and measures mechanical output directly, although temperature changes can affect meter accuracy.
Your cycling computer will usually show several versions of it:
- Instant power: the reading right now, which jumps around a lot.
- Averaged power: a smoothed figure over a few seconds, easier to pace by.
- Ride average: total work divided by time, which hides surges and coasting.
Ride average is the number most people look at first, and it is also the one that misleads most often. Two rides can share an average and feel completely different.
What is FTP and why does everything depend on it?
Functional Threshold Power, or FTP, was popularized by Hunter Allen and Andrew Coggan in their book “Training and Racing with a Power Meter”. TrainingPeaks describes it as the maximum power you can maintain through an hour’s effort without fatiguing.
A wrong FTP skews your power zones, IF and TSS. If yours is a guess, retest before you trust the rest.
A common field method is the 20-minute test. Per the same guide, your average power for that 20 minutes, less 5 percent, approximates your FTP: a 250-watt average gives 238 watts.
A structured alternative is covered in our ramp test guide, which suits riders who dislike a long solo effort.
Zones: turning FTP into target ranges
Zones split your power into bands, each as a percentage of FTP. Joe Friel’s TrainingPeaks guide gives this six-zone cycling set, which he cites from Allen and Coggan:
| Zone | Share of FTP |
|---|---|
| Zone 1 | Less than 55% |
| Zone 2 | 55% to 74% |
| Zone 3 | 75% to 89% |
| Zone 4 | 90% to 104% |
| Zone 5 | 105% to 120% |
| Zone 6 | More than 120% |
These bands come from Joe Friel’s quick guide to setting zones; zone models can change, so check the source before you rely on them.
Other coaches and apps use different zone counts and labels, so check which model your software applies before comparing notes with a friend. The idea is the same: easy riding below threshold, hard efforts above it.
Zones also give you a shared language. Instead of “I rode hard”, you can say most of the time went into the second zone, with two efforts in the fourth. That makes it far easier to plan a week, repeat what worked and spot when you are going too hard on easy days.
Normalized power: why steady and stop-start rides differ
No ride is perfectly steady, and surges cost more than the average suggests. Normalized power (NP) adjusts for that. TrainingPeaks gives an example: 100 watts for 20 minutes, 200 for 20 and 300 for 20 still averages 200 watts, but the NP is approximately 239 watts.
The same guide says to watch NP on your cycling computer rather than average power. Our normalized power page covers how the calculation works.
The Variability Index (VI) compares the two. It measures how smooth your output was, and a steady ride such as a triathlon should have a VI of 1.05 or less. A criterium or a hilly group ride will sit higher.
Intensity factor and TSS: how hard was it, and how much?
Intensity factor (IF) is simply the ratio of normalized power to your FTP. With 210 watts of NP and an FTP of 280 watts, the IF is 0.75.
The same ride later in the year, after threshold power had risen to 300 watts, would score an IF of 0.70. That is a useful sign: identical work now sits lower relative to your ability.
Training Stress Score (TSS) adds duration. TrainingPeaks describes it as a single value for the effective load of a session, and you earn 100 TSS for an all-out, 100 percent, 60-minute workout.
- Check that your FTP is current.
- Look at NP and IF to judge how hard the ride was.
- Use TSS to compare a short hard day with a long easy one.
A simple routine for reading your bike power data
Before the ride, decide what it is for: an easy spin, a steady endurance block or a few hard intervals. Afterwards, check whether the numbers match the plan, not whether they are big.
An easy day that shows a high IF is a sign you rode too hard, however good it felt. A hard day with a low IF may mean the efforts were too short or the rests too long.
Write one line in your training notes per ride: the goal, the normalized power, the IF and how you felt. After a month, that log shows more than any single chart.
Reading your data without over-reading it
One ride tells you little; trends tell you more. Compare similar rides over weeks and note whether the same watts feel easier or your IF at a given pace falls.
Keep a few cautions in mind:
- A stale FTP skews zones, IF and TSS together.
- Different head units and apps can calculate smoothing differently, so compare like with like.
- Numbers do not replace how you feel. Poor sleep or illness matters even when the watts look fine.
Context matters as well. Wind, heat, a fresh set of tyres or a new power meter calibration can each shift the numbers slightly, so treat small changes as noise and look for patterns that repeat across several rides.
Pick one or two metrics to watch, ignore the rest until they answer a real question, and build from there.
Frequently asked questions
Is average power or normalized power better?
Show answer
For variable rides, normalized power better reflects the effort. TrainingPeaks advises viewing NP on your cycling computer rather than average power.
How often should I retest FTP?
Show answer
There is no single rule in the sources used here. Retest whenever your fitness has clearly changed, since IF and TSS rely on the figure.
What is a good IF?
Show answer
It depends on the ride. TrainingPeaks’ own example is a long early-season training ride that scored an IF of 0.75, and because IF is a ratio to FTP, a higher value means higher normalised power relative to FTP.
Does 100 TSS always mean the same thing?
Show answer
It equals an all-out 60-minute workout, but how tired you feel also depends on sleep, fuel and recent training.
