Tri ToolkitCalculators
Train › Guide

Training Zones Explained

Updated 2026-11-21

Training zones are ranges of effort built from one test per sport: Critical Swim Speed (CSS) for swimming, Functional Threshold Power (FTP) for cycling, and threshold heart rate or a VDOT score for running. Each estimates roughly the hardest effort you can sustain for a long time, but with a different test and different bands. That is why “zone 2” differs between apps, and why zones are a starting point, not a precise measurement.

Swim: Critical Swim Speed (CSS)

Test: after a warm-up, swim 400 m all-out, recover fully, then swim 200 m all-out. Record both times in seconds.

Formula:

CSS (metres per second) = 200 ÷ (T400 − T200)

Pace per 100 m = (T400 − T200) ÷ 2, in seconds.

Example: 400 m in 7:20 (440 seconds), 200 m in 3:30 (210 seconds). CSS = 200 ÷ 230 = 0.87 metres per second, or 1:55 per 100 m.

The idea of a “critical velocity” in swimming comes from Wakayoshi and colleagues (1992), who plotted distance against time across several swim trials. The quick two-trial version using 200 and 400 m was studied in their 1993 paper, which found CSS was close to the pace at a 4 millimoles per litre blood lactate level Moderate evidence. Later work suggests the 200/400 version tends to sit slightly above the true maximal steady state, and the studies were done mostly in trained pool swimmers.

Bike: FTP and Coggan’s power zones

FTP is defined as roughly the highest power you can hold for about an hour. Since few people want to test for an hour, common shortcuts are Coaching practice:

  • 20-minute test: FTP ≈ 0.95 × average power for 20 minutes.
  • Ramp test: FTP ≈ 0.75 × best 1-minute power reached in a ramp. This is how TrainerRoad and Zwift calculate it.

How well does this work? At the group level, FTP from the 20-minute test correlates well with laboratory thresholds. For individuals, results vary. Borszcz et al. (2019) found good average agreement with maximal lactate steady state but individual errors of around ±9%, larger in less-trained riders Moderate evidence. Inglis et al. (2019) found that maximal lactate steady state was on average about 93% of the 95%-of-20-minute FTP, meaning the formula overestimated it Moderate evidence. The ramp test factor is a platform convention with little published validation, and it tends to overestimate FTP for riders with a strong sprint Limited evidence.

Coggan’s power levels (% of FTP):

Level Name % FTP
1 Active recovery < 55%
2 Endurance 56–75%
3 Tempo 76–90%
4 Lactate threshold 91–105%
5 VO₂max 106–120%
6 Anaerobic capacity > 121%
7 Neuromuscular Maximal (no fixed range)

Coggan himself calls these “levels” to stress that the boundaries are not physiological switches.

Coggan power levels and race intensityCoggan power levels as percent of FTP: level 1 below 55, level 2 56 to 75, level 3 76 to 90, level 4 91 to 105, level 5 106 to 120, level 6 above 121. A 70.3 bike is usually ridden at 80 to 85 percent of FTP, in level 3. An Ironman bike at 65 to 75 percent, in level 2.1Recovery2Endurance3Tempo4Threshold5VO2max6+Anaerobic0%25%50%75%100%125%150%percent of FTP70.3 bike: 80–85%Ironman bike: 65–75%Level 7 (neuromuscular) has no fixed range and is left out.
Coggan power levels (TrainingPeaks) with typical race intensities from TrainingPeaks and Best Bike Split. Most racing happens in levels 2 and 3, not at threshold.

Race intensity sits lower than most people expect. A 70.3 bike is usually ridden at about 80–85% of FTP, the top of tempo, and an Ironman at about 65–75%, in the endurance band. Our guide on how to pace a 70.3 or an Ironman explains why, and the race pacing planner turns your FTP into target watts.

Run: Friel heart-rate zones

Joe Friel’s system bases run zones on lactate threshold heart rate (LTHR). Friel’s field test, which our calculator uses: run 30 minutes alone at the best effort you can sustain, and take your average heart rate over the last 20 minutes as LTHR Coaching practice.

Friel run zones (% of LTHR):

Zone % LTHR
1 < 85%
2 85–89%
3 90–94%
4 95–99%
5a 100–102%
5b 103–106%
5c > 106%

Friel’s bike zones use slightly different percentages, and your bike LTHR is usually a few beats lower than your run LTHR. Do not reuse one sport’s heart rate in the other.

Run: Daniels VDOT

Jack Daniels’ VDOT system starts from a recent race result instead of a lab-style test. The VDOT tables (first published by Daniels and Gilbert in 1979) link race times to a “VO₂max-equivalent” score, and from that score to five training paces: Easy, Marathon, Threshold, Interval and Repetition Coaching practice. The tables are built on regression equations from runners’ data. They are widely used, but they predict pace, not your personal physiology, and they assume you are trained for the distance you raced.

Why the numbers differ between systems

  1. Different anchors. CSS, FTP, LTHR and VDOT threshold pace each estimate “threshold” with a different test. Each test has its own error, and none exactly equals a lab measurement.
  2. Different band edges. Coggan has seven levels, Friel seven zones (with 5a–c), apps often use five. “Zone 2” in a five-zone model can cover a different range than in a seven-zone model.
  3. Heart rate lags and drifts. Heart rate rises with heat, dehydration and fatigue, while power and pace do not (more in our guide to racing in the heat). Heart-rate zones and power or pace zones will disagree in long or hot sessions.
  4. Sport-specific thresholds. Your threshold heart rate is different for bike and run, and swim heart rate is usually lower again.
  5. Day-to-day variation. Sleep, heat and fatigue move all these numbers a few percent.

The practical response: pick one system per sport, retest every 6–10 weeks or after a clear change in fitness, and use how the effort feels as a cross-check Coaching practice.

Use the zones calculator to turn your test results into zones in each system and compare them side by side.

FAQ

How often should I retest?

Every 6–10 weeks, or sooner after a clear change in fitness, illness or a long break. Retest under similar conditions each time: same time of day, similar rest and the same equipment.

Which is better, heart rate or power?

For pacing, power responds instantly and is not affected by heat or fatigue, so it is more precise on the bike. Heart rate is still useful as a ceiling, especially in long or hot sessions, and it is free.

Why is my zone 2 different in every app?

Apps use different anchors (FTP, threshold heart rate, maximum heart rate) and different numbers of zones. A five-zone model and a seven-level model split the same range differently, so the labels do not match.

Can I use the same heart-rate zones for bike and run?

No. Threshold heart rate is usually a few beats lower on the bike than on the run, and lower again in the water. Test each sport separately.

Is the 20-minute test or the ramp test more accurate?

Neither is perfect. The 20-minute test tends to overestimate your true threshold slightly, and the ramp test can overestimate it more if you have a strong sprint. Pick one and stick to it, so changes over time are comparable.

What zones should I race a triathlon in?

Most age-groupers race a 70.3 bike at about 80–85% of FTP and an Ironman bike at about 65–75%. The run is usually done by feel and heart rate, starting easier than your stand-alone race pace.

Sources

  1. Wakayoshi et al. 1992, Critical velocity in swimmers
  2. Wakayoshi et al. 1993, Critical swimming velocity and MLSS (200/400 m)
  3. Coggan, Power training levels (TrainingPeaks)
  4. Borszcz et al. 2019, FTP vs maximal lactate steady state
  5. Inglis et al. 2019, MLSS vs 20-minute FTP test
  6. Friel, Quick guide to setting zones (TrainingPeaks)
  7. Daniels & Gilbert 1979, Oxygen Power (VDOT tables, book)
  8. TrainerRoad, Ramp Test FAQs (75% of best 1-minute power)