Athlete summary
FTP (Functional Threshold Power) is the highest power you can hold in a quasi-steady state for about an hour. A full 60-minute all-out test is hard to pace and recover from, so coaching practice uses shorter field tests and multiplies the result by a fixed factor.
Koyaro locks these factors (no silent formula changes):
| Protocol | Engine input | Factor | Derived FTP |
|---|---|---|---|
| 20-minute | best 20-min mean power | × 0.95 | ftp_from_20min |
| 8-minute (×2) | best 8-min mean (of the pair) | × 0.90 | ftp_from_8min |
| Ramp | final 1-min / MAP peak | × 0.75 | ftp_from_ramp |
Setting FTP from a test recomputes Coggan power zones. eFTP from a ride’s mean-maximal curve uses the same 20-min / 8-min factors when the effort looks maximal.
Prefer Critical Power when you have enough maximals
A single bout × a population factor is a surrogate. When you have several well-spaced maximal efforts (~2–20 minutes), a Critical Power / W′ fit with quality gates (accepted) is preferred: it models the whole power–duration curve instead of assuming you are exactly 5%, 10%, or 25% “above threshold” on one effort.
See Critical Power, W′, and fit quality. Use dedicated 20-min / 8-min / ramp sessions when you want a simple FTP refresh or when you do not yet have a usable CP set.
20-minute test (× 0.95)
Source: Allen & Coggan, Training and Racing with a Power Meter — classic field protocol: hard warm-up (often including a short blow-out), then 20 minutes all-out; FTP = 95% of mean power.
The 5% discount accounts for anaerobic contribution that still inflates a 20-minute effort above true ~60-min threshold power. It is a population average; athletes with unusually large or small anaerobic capacity may sit slightly off the line.
Citation key: allen-coggan-ftp-20min
8-minute test (× 0.90)
Source: Carmichael Training Systems (CTS) 2×8 protocol — two maximal 8-minute efforts with recovery between; coaching practice takes the higher (mean) 8-min power and multiplies by 0.90.
Eight minutes draws more from anaerobic reserves than twenty, so the correction is larger (10%). Useful when a full 20-minute effort is psychologically or logistically hard; expect more day-to-day noise than the 20-min test.
Citation key: cts-ftp-8min
Ramp test (× 0.75)
Source: TrainerRoad-style (similar) ramp conventions — progressive steps to failure; take the highest 1-minute mean power (often treated as MAP) and estimate FTP as 75% of that peak.
In trained endurance athletes, FTP commonly falls around 72–82% of MAP; 75% is the midpoint convention used for software defaults. Athletes with high anaerobic contribution or short ramp steps can bias the estimate; treat ramp FTP as convenient, not gold-standard.
Citation key: trainerroad-ramp-ftp
Limits (all single-effort protocols)
- Factors are population coaching conventions, not athlete-specific physiology.
- Pacing, warm-up, heat, carbohydrate, and motivation move the raw power.
- Ramp and short TTs are more sensitive to anaerobic capacity than a long TT.
- Do not change these multipliers in the engine without updating golden tests and this article together.
Literature
- Allen H, Coggan AR (later editions with McGregor) — Training and Racing with a Power Meter; FTP definition and 20-min × 0.95 field protocol.
- Carmichael Training Systems — 2×8-minute FTP test; 90% conversion used widely in time-crunched coaching.
- TrainerRoad (related indoor-training practice) — ramp to failure; FTP ≈ 75% of maximal 1-min / MAP as the default software conversion.
- Validation / critique literature (e.g. Borszcz and others on FTP20 vs MLSS) — treat field FTP as an estimate correlated with threshold markers, not identical to lab MLSS for every athlete.
- Jones & Vanhatalo — Critical Power / W′ as the preferred multi-effort framework when enough maximal efforts exist (see CP article).