Draft site, not public. Numbers marked verify are still being checked against primary sources.
Tri Toolkit
Start here › Guide

Training Zones Explained

Updated 2026-11-20 · draft, awaiting author review

Calculators for this guide

Training zones are ranges of effort built from one test result per sport: Critical Swim Speed (CSS) for swimming, Functional Threshold Power (FTP) for cycling, and threshold heart rate or a VDOT score for running. All of them estimate roughly the same thing, the hardest effort you can sustain for a long time, but each system uses a different test and divides effort into different bands. That is why your “zone 2” in one app will not match another, and why the numbers are a starting point rather than 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 (m/s) = 200 ÷ (T400 − T200)

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

Example: 400 m in 7:20 (440 s), 200 m in 3:30 (210 s). CSS = 200 ÷ 230 = 0.87 m/s, 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 mmol/L 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:

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-min FTP, meaning the formula overestimated it Moderate evidence. The ramp test factor is a platform convention with little published validation 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.

Run: Friel heart-rate zones

Joe Friel’s system bases run zones on lactate threshold heart rate (LTHR). A common field test: 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.

Author note (to write): which run system you use off the bike, and whether VDOT paces from a stand-alone race hold up in a triathlon.

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. 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.

Author note (to write): how you use zones for a 70.3 or full race plan, for example target power as % FTP and run pace off the bike.

Open checks

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-min FTP test
  6. Friel, Quick guide to setting zones (TrainingPeaks)
  7. Daniels & Gilbert 1979, Oxygen Power (VDOT tables)