Critical Power vs FTP: Which Threshold Metric Should Drive Your Training?
5 min read
Critical Power and FTP are not the same threshold metric. Learn when to use CP, when to use FTP, and how to switch zones without blurring your training choices.
On this page
Photo by Rizal Setiya on Unsplash.
CP is model-based; FTP is a practical single-number estimate. Use CP when varied maximal efforts and power-duration context drive the next choice.
Critical Power and FTP both help you set training load, but they do not ask the same question. CP comes from a power–duration model, while FTP gives a field-friendly anchor for zones and progress checks. Because the grounding source is a PubMed search rather than one named study, this refresh keeps physiology claims narrow and focuses on sound coaching use.
Critical Power, or CP, is a model-based threshold drawn from several maximal efforts across different ride lengths. FTP, or Functional Threshold Power, is a practical single-number estimate often tied to a hard steady effort.
The difference matters because a threshold metric should shape the next session, not just fill a dashboard. Use CP when you can feed the model with sound power files and want richer guidance for efforts above threshold.
Use FTP when you need a simple field test, clear zones, and less analysis overhead. If your late-season racing is done and autumn base work is starting, FTP may be enough until harder work returns.
Pick CP if you have several clean maximal efforts across durations.
Pick FTP if you need one clear anchor for zones.
Do not switch labels mid-block without resetting zones.
Use the same power source for key tests.
The best metric is the one that turns your current data into one clear training choice.
Keep one threshold metric, pick the one that fits your data and coaching goals.
Photo by Karen Pancho on Unsplash.
CP comes from the power–duration curve, which maps how your best power falls as time gets longer. The model also pairs CP with W′, a finite work capacity above CP.
FTP is more of a coaching tool than a full model. It gives you one number that can set zones, guide workouts, and track whether training is moving the right way.
That makes CP richer and FTP easier. For field testing basics, use a clear threshold field test before you reset zones from one hard day.
If you need FTP-specific steps, a well-paced FTP test is the cleaner first move. If your lab and field numbers disagree, review why threshold tests diverge before changing every workout.
CP needs several best efforts, not one file.
FTP needs repeatable setup and steady pacing.
W′ helps frame work done above CP.
Zones must be rebuilt when the anchor changes.
CP gives a system model; FTP gives a usable single-number dial.
Critical Power (CP) is derived from a power-duration model using multiple maximal efforts and maps to an intensity that separates heavy a…
Data analysis is free, forever. The Dynamic Coach is Pro. Connected to Garmin, Strava, Whoop, and intervals.icu.
PubMed-indexed work supports the broad split between CP as a power–duration model and FTP as a practical field estimate. The two ideas overlap in coaching use, but they are not the same construct.
The open question is not whether CP and FTP differ. The harder question is which one gives better long-term training choices for a given rider, test setup, and event type.
That uncertainty should make your method tighter, not more complex. Keep physiology claims modest, keep tests clean, and avoid treating one threshold file as a verdict on fitness.
If progress slows, your threshold did not disappear; the training system around it may have drifted. Review block order with threshold and VO2max sequencing before you add more intensity.
Treat CP and FTP as different tools.
Do not compare them as identical labels.
Keep test setting as stable as possible.
Use peer-reviewed claims with care.
Literature supports the conceptual difference; which trains better long-term remains context-dependent.
Use CP if you can collect reliable maximal efforts across more than one duration and will use W′ for interval planning. Use FTP if you need a simple test that you can repeat without heavy modeling.
If you are unsure, keep FTP for the current block and gather better maximal efforts in the background. Then move to CP only when the added detail will change sessions, pacing, or race choices.
One tactical email with training ideas and product updates. No spam — unsubscribe anytime.
Keep reading
- Norwegian Threshold Method for Amateur Cyclists: Lactate-Guided Training Without a Lab (Updated) — PubMed does not confirm a branded Norwegian Threshold Method. Use this conservative field protocol for lactate-informed threshold cycling without a l...
- Four-Minute vs Five-Minute vs Eight-Minute VO2max Intervals: Choosing the Right Duration — Compare 4-, 5-, and 8-minute VO2max intervals for cycling, then use a one-week test protocol to choose the right duration for your goal.
- Training for Masters Cyclists (40+): Age‑Adapted Strategies for Sustainable Performance — Practical, science-based training for cyclists 40+. Learn how to preserve VO2max, fight sarcopenia, prioritise recovery, and use adaptive plans to ke...
Switching threshold metrics changes zone borders, session targets, and how you read hard intervals. That is why a metric change should be handled like a training change, not just a settings update.
CP-based sessions can aim work around the modeled threshold and the limited capacity above it. FTP-based sessions are simpler, using percent-based zones that many riders and tools already know well.
If you use over-unders, be clear about which anchor sets the load. A block of threshold over-unders will feel different when the line moves.
For high-intensity work, match the metric to the workout aim. A needs a sound anchor, but it also needs good recovery input.
This rule also fits the autumn handoff from racing toward base. Keep the system calm, then add sharper CP-based work when the plan again asks for hard repeats.
Stay with FTP when simple zones solve the problem.
Move to CP when surge cost shapes performance.
Collect clean maximal efforts before modeling CP.
Use one primary metric per block.
The next move is not a new label; it is the metric your data can support today.
Keep the threshold that your measurement system can sustain.
Rebuild zones after any metric switch.
Tag old rides with the old anchor.
Log rest, fueling, heat, and power source.
Make the first week after switching slightly easier.
The training system around your threshold must stay consistent for decisions to remain valid.
CP earns its place when performance depends on repeated hard work, not only one smooth effort. Think short climbs, race surges, attacks, or hard exits where effort cost builds across the ride.
The model is only as good as the efforts behind it. Poor pacing, stale fatigue, or mixed power sources can bend the curve and make the next session less clear.
Use CP when you will act on the extra signal. If it will not change the workout, pacing plan, or recovery choice, FTP may still be the better tool.
Use CP for surge-heavy racing demands.
Use CP when duration-specific pacing matters.
Skip CP when test files are weak.
Retain FTP if CP will not change training.
FTP is the better first move when the goal is steady progress, clear zones, and repeatable checks. It is easy to explain and easier to keep consistent across a normal training month.
That simplicity has value. A rider who tests well, records context, and trains from one stable anchor often makes better choices than a rider who keeps swapping models.
If you want a wider base for power work, review how FTP anchors training. Then hold the number steady long enough to learn from the sessions it shapes.
Week 0 — Decide and baseline. Pick the metric using the decision rule. If choosing CP, collect several maximal efforts across short, middle, and longer hard durations under well-rested conditions. If choosing FTP, perform a standardized field test and record conditions.
Weeks 1–2 — Adjust the block. Keep planned intensity targets, but trim total load while the new anchor settles. Do two sub-threshold sessions and one long aerobic ride, then log perceived effort and interval quality.
Weeks 3–4 — Reassess and lock. Check whether recent sessions match the chosen threshold. If FTP is the driver, repeat the same field protocol when needed. If CP is the driver, add clean hard efforts only when they improve the model.
Before rewriting zones again, decide whether the change reflects fitness, test noise, fatigue, or setup. One clean decision beats a string of small guesses.
CP is model-based; FTP is a practical single-number estimate. Use CP when reliable power-duration data and repeated hard efforts shape training. Use FTP when you need simple zones, repeatable testing, and a stable anchor for the next block.
CP can give more detail across durations, but only when the input efforts are clean. FTP is less detailed, yet often more useful when you need one stable training anchor.
Yes, but use one primary metric for the block. Let the other metric answer a narrow question, such as pacing or test comparison, not rewrite every session.
Most newer power users should start with FTP because the test is simpler and the zones are easier to apply. Move to CP when racing or coaching needs make the extra data worthwhile.
They do not have to match because they come from different methods. Check pacing, fatigue, power source, and test setting before you change the plan.