Key Takeaways
- Critical power (CP) is the power above which the body's responses to exercise can no longer be stabilised. W' is the work you can do above CP before you are spent[1]. This calculator fits both from 3 to 6 maximal efforts.
- There is no standard test. The lengths of your efforts, how many you do, how much rest you take and which equation is used all change the answer[3]. Repeat the same protocol if you want to track change.
- CP is not FTP. In two studies of trained cyclists, CP came out higher than a 20-minute FTP estimate, by 7 W on average in one[4] and 16 W in the other[5], with wide gaps for individuals.
- Use the results inside the 2 to 15 minute range the model is normally fitted to[2]. Do not read them as a power you can hold for an hour.
What Critical Power Is
The link between how hard you ride and how long you can last is hyperbolic. For cycling it reduces to two numbers: CP, the power the curve levels off toward, and W', the work you can do above CP[1].
Together they predict how long you can last above CP[1].
CP is described as a fatigue threshold. Below it, the physiological responses to exercise can be stabilised.
Above it, they cannot, and exercise ends when W' is used up[1]. Chorley and Lamb describe how long a rider can stay at CP as somewhat unpredictable[2], so do not assume you can ride at CP indefinitely.
How the Calculator Works
Each effort gives a pair: its average power P and its length t in seconds. The two-parameter model says P = W' ÷ t + CP. If you plot P against 1 ÷ t, the points fall on a straight line.
CP is where the line crosses the power axis, and W' is its slope in joules[2]. We fit that line by ordinary least squares.
The same model can be written as a hyperbola or as a line of work against time. In theory all three give identical results. They agree only when every trial is truly maximal and the fit is perfect, so with real data each gives slightly different values[2].
McGrath and colleagues compared one hyperbolic and two linear versions in highly trained riders[5]. We use the power against 1 ÷ time version.
Example (illustrative numbers)
The intervals are 95% confidence intervals from the scatter of your points around the line, using a t distribution with the number of efforts minus 2 degrees of freedom. They do not include error from the protocol, such as an effort that was not maximal.
Choosing Your Efforts
| Choice | What the sources say |
|---|---|
| Length | Time-to-exhaustion trials are usually 2 to 15 minutes. Shorter trials give a higher CP and lower W' than longer ones, while the fit quality (R²) stays similar[2]. |
| Number | Typically three, and up to seven[2]. This tool takes 3 to 6. |
| Recovery | Testing on one day with 30 minute and 3 hour recoveries gave lower W' than testing on separate days with 24 hours between[2]. |
| Format | Constant-power trials to failure are the conventional method. Time trials have been used in the field, but their validity has raised concerns[3]. A 3-minute all-out test is another option, though it may overestimate CP and needs an earlier incremental test[3]. |
One study used time trials of 12, 7 and 3 minutes with 30 minutes of passive rest in between[4]. We pre-fill those lengths as a starting point, not as a validated protocol. Give each effort your best, and enter the average power you held, not the peak.
Reading Your Results
Compare your numbers only with results from the same protocol. A change in effort lengths alone can move CP and W'[2]. The table of your efforts shows how far each one sits from the fitted line, so you can see which effort pulls the fit most.
R² is shown for information only. Chorley and Lamb report that trials of different lengths can give different CP and W' values with similar R²[2], so a high R² does not show that the test was a good one.
CP and FTP are different numbers
FTP here means 95% of your best 20-minute power[4]. In 17 trained cyclists, CP from three time trials averaged 256 W against 249 W for FTP. The mean gap was 7 W, but the 95% limits of agreement ran from −19 to 33 W[4].
A second study of highly trained athletes found 282 W for CP (from one of its linear models) and 266 W for FTP, a mean gap of 16 W[5]. Both teams concluded that the two should not be used interchangeably[4][5].
So this page does not convert CP into an FTP.
Limitations
- The model is described as remarkably robust, but with a number of shortcomings[6]. One is that the equation sends power toward infinity as time goes to zero[2], so it cannot describe very short efforts. We keep the prediction table to 2 to 15 minutes, the range the model is normally fitted to.
- Results depend on the protocol and on whether each effort was truly maximal. The testing protocol has not been standardised[3]. Self-paced time trials have given higher CP estimates than time-to-exhaustion trials[2].
- This calculator fits one line to all your efforts. Chorley and Lamb note that a best-fit approach cannot exceed a rider's true limits, so it will underestimate CP, W' or both[2].
- Three efforts leave one degree of freedom, so the 95% interval can be very wide. A fourth effort narrows it.
- We accept efforts of 30 seconds to 60 minutes and 50 to 2000 W to catch typing errors. Those limits are not from a source. We warn when any effort falls outside 2 to 15 minutes.
- There is no FTP estimate or training-zone plan here. The studies above do not support treating CP as a fixed percentage of FTP or the reverse, and no source gives zone percentages built on CP. For ride loads, see the Cycling NP, IF and TSS Calculator.
- This is a training reference, not medical advice. All-out efforts are demanding, so check with a doctor first if you have a heart or other health condition. For a race-day estimate from power, try the Cycling Race Predictor, or work out speed and pace with the Bike Pace Calculator.
Frequently Asked Questions
Sources & References
- Critical Power: An Important Fatigue Threshold in Exercise Physiology. Poole DC, Burnley M, Vanhatalo A, Rossiter HB, Jones AM. Medicine & Science in Sports & Exercise. 2016;48(11):2320-2334 (2016)
- The Application of Critical Power, the Work Capacity above Critical Power (W'), and its Reconstitution: A Narrative Review of Current Evidence and Implications for Cycling Training Prescription. Chorley A, Lamb KL. Sports (Basel). 2020;8(9):123 (2020)
- Methodological Approaches and Related Challenges Associated With the Determination of Critical Power and Curvature Constant. Muniz-Pumares D, Karsten B, Triska C, Glaister M. Journal of Strength and Conditioning Research. 2019;33(2):584-596 (2019)
- Relationship Between the Critical Power Test and a 20-min Functional Threshold Power Test in Cycling. Karsten B, Petrigna L, Klose A, Bianco A, Townsend N, Triska C. Frontiers in Physiology. 2021;11:613151 (2021)
- Do Critical and Functional Threshold Powers Equate in Highly-Trained Athletes?. McGrath E, Mahony N, Fleming N, Raleigh C, Donne B. International Journal of Exercise Science. 2021;14(4):45-59 (2021)
- The critical power and related whole-body bioenergetic models. Morton RH. European Journal of Applied Physiology. 2006;96(4):339-354 (2006)

Certified Personal Trainer & Sports Nutritionist
NASM-certified fitness and nutrition coach with over 10 years of hands-on experience helping people build strength, lose fat, and live healthier lives. Specializing in gym-based workouts with a strong focus on lifting technique, biomechanics, and practical exercise science. Through FitLifeRegime, sharing the tools, tips, and insights that have worked for hundreds of clients — helping you start your own fitness journey with confidence and clarity.
Learn more about Manish
NASM Certified
Certified Personal Trainer & Sports Nutritionist from the National Academy of Sports Medicine
Disclaimer: This tool is for informational and personal journaling purposes only. Results are estimates based on published formulas and may vary based on individual factors such as genetics, training history, and measurement technique.