Peak HybridEvery number, explained.
The formulas behind your training load, fitness, zones, estimated FTP, and VO2max, written out the way the app runs them, with the research each one rests on. Try the load math on your own sessions below.
1 scale for the gym and the road.
Every session gets a load score on the same scale: 100 is 1 hour at your threshold. A heavy leg day and a long ride land in the same week, added together.
load = minutes × (RPE ÷ 10) × 0.8- RPE is the average of the RPE you log on your sets. It counts as at least 6 and at most 10.
- Density can raise it. Each working rep counts as 3.5 seconds of work and timed sets count their logged time. When work fills more than 15% of the session, effective RPE climbs 1.33 for every 10 points of density above that, up to 10.
- Your logged RPE wins when it is higher. Density only ever raises the score.
- Most of a lifting session is rest, so an hour at RPE 10 scores 48.
- A lift recorded outside the app, in Strava or Apple Health, counts at RPE 6: minutes × 0.6 × 0.8.
load = hours × intensity² × 100- Rides with a power meter: intensity is normalized power ÷ FTP, over moving time.
- Runs, hikes, swims, and rides without power: intensity is average heart rate ÷ LTHR. Running power is left out, because your FTP is a cycling number.
- Sessions you log by hand: intensity comes from your effort rating, in the table below. A session saved without a rating counts as 6.
- A walk counts as 0. Add a pack weight and it counts as a ruck, scored by heart rate.
Effort rating to intensity · load per hour
Anchored on common intensity conventions at the even ratings: 4 is endurance, 8 is threshold. Ratings above 8 climb faster, since effort above threshold costs more than the scale suggests.
Run the numbers on your day.
The same math the app runs, filled in as you type. The zone picker uses the mid-zone intensity the app uses to estimate a planned workout: Z1 0.475, Z2 0.65, Z3 0.83, Z4 0.98, Z5 1.13.
Starts from 0 with the same day on 4 days of every week, and rest days at 0. Your own chart starts from your real history, so read this for the shape of the curves: fatigue answers in days, fitness in weeks.
What your training is building.
Fitness is your load averaged over about 42 days. Fatigue is the same over about 7 days. Form is the gap between them. All 3 are free in the app.
fitness = yesterday's fitness × e^(−1/42) + today's load × (1 − e^(−1/42))
fatigue = yesterday's fatigue × e^(−1/7) + today's load × (1 − e^(−1/7))
form = fitness − fatigue- Each day keeps about 97.6% of yesterday's fitness and about 86.7% of yesterday's fatigue, so recent days weigh most and older days fade smoothly.
- Rest days count as a load of 0, so form rises on the days you rest.
- Each session counts on your local calendar day, so an evening ride stays on today.
The bands are a coaching read of training state, set from practice rather than a study cutoff.
Set from your own sessions.
A value you type in Profile always wins. Peak Hybrid fills only the fields you leave empty, and marks each one as estimated.
- Max heart rate: the higher of 208 − 0.7 × age (Tanaka 2001) and the highest heart rate recorded in your synced activities.
- LTHR: 92% of max heart rate.
- FTP: your estimated FTP from power-meter rides, below. When your typed FTP and your rides differ by more than 3%, the app offers to use the ride number or keep yours.
- Run pace: anchored on 5K pace. A typed 5K time is used as is; a 10K, half, marathon, or mile time converts with Riegel's formula, time₂ = time₁ × (distance₂ ÷ distance₁)^1.06.
- With no race times, the anchor comes from your runs: the faster of your speed at your zone 2 heart-rate cap × 1.17 (marathon pace to 5K pace, after Daniels) and your best 12 minutes, converted to 5K.
| Power zone | % of FTP |
|---|---|
| Active recovery | under 55% |
| Endurance | 55 to 75% |
| Tempo | 75 to 90% |
| Threshold | 90 to 105% |
| VO2max | 105 to 120% |
| Anaerobic | 120 to 150% |
| Neuromuscular | over 150% |
| HR zone | % LTHR | % max |
|---|---|---|
| Z1 Recovery | under 75% | 50 to 60% |
| Z2 Endurance | 75 to 89% | 60 to 70% |
| Z3 Tempo | 89 to 94% | 70 to 80% |
| Z4 Threshold | 94 to 100% | 80 to 87% |
| Z5 VO2max | 100 to 106% | 87 to 100% |
| Run pace | 5K pace |
|---|---|
| Easy | × 1.30 |
| Marathon | × 1.17 |
| Tempo | × 1.07 |
| Threshold | × 1.03 |
| Interval | × 1.00 |
| Repetition | × 0.95 |
Power zones follow the Coggan levels. Heart-rate zones use % of LTHR once you set an LTHR, and % of max heart rate until then. The zone 1 top sits at 75% of LTHR, where most coaches place the aerobic threshold; the upper boundaries follow Friel.
Your FTP from rides you already did.
Critical-power fit on your best 2 to 15 minute efforts from the last 365 days. Power-meter rides only; e-bike rides stay out.
power(t) = CP + W′ ÷ t fitted on your best 2 to 15 min efforts
20 min = CP + W′ ÷ 1200
eFTP = factor × 20 min factor 0.90 to 0.95How it is built
- The fit reads 2 to 15 minutes. Shorter efforts lean on anaerobic power and inflate the fit; longer ones are often below your best.
- 60 minute power runs about 90% of a maximal 20 minutes (MacInnis 2019), so the factor runs 0.90 to 0.95, 0.92 by default. A curve heavy on 1 minute power sits near 0.90, and heart rate near LTHR on 8 to 15 minute efforts is what lets it reach 0.95.
- A 20 minute effort you actually held sets a floor: the estimate re-anchors on it.
- A typed FTP always stays in force. The estimate fills the field only when you leave it empty.
How it ages
- An effort counts in full for 8 weeks.
- After that it fades 1% a week while you keep riding, and 2.5% a week through weeks off the bike. Past 365 days it drops out.
- The rates sit inside the detraining research (Coyle 1984; Mujika & Padilla 2000): slow while you ride, faster through a break.
Heart rate as a 2nd read
- Heart rate can raise eFTP by up to 8%. It never lowers it.
- It needs both a power-to-heart-rate line from 10 or more steady segments across 2 or more rides, and a steady effort of 25 minutes or more at 80 to 97% of LTHR, both in the last 6 weeks.
- It stays off while your LTHR is an estimate. When heart rate drives the number, confidence tops out at medium.
Read eFTP as a band of about ±5%, not a point. Individual 20 minute power varies about 5% with the warm-up alone (Sitko 2022), and the validation studies were small groups of trained men (8 to 23 riders each). The fit window and factor were checked against MacInnis 2019, Mattioni Maturana 2018, Borszcz 2018, Karsten 2021, Nimmerichter / Prinz 2020, and Spragg 2023.
From rides with power and runs with heart rate.
Each sport only shows as much of your ceiling as your skill in it allows, so the headline is the higher of your ride and run estimates. Both are on the Body tab, free.
VO2max = 3.5 + (VO2 cost of pace − 3.5) ÷ %HRR
%HRR = (heart rate − resting) ÷ (max − resting)- The share of heart-rate reserve you use matches the share of oxygen reserve (Swain). The oxygen cost of a pace comes from Daniels' equation.
- Each run is read on its best 12 minute window, between 40% and 95% of heart-rate reserve.
- The estimate is your 2nd best run of the last 90 days, so 2 efforts have to agree before it moves. It needs 5 runs with heart-rate and speed streams; with fewer, it reads whole-run averages (the 85th percentile) from 3 runs or more.
- Needs a resting heart rate, from Apple Health.
VO2max = 10.8 × (5 min power ÷ body mass in kg) + 7- The 5 minute power is your best genuine maximal effort of the last 12 months: 1.1 to 1.6 × the eFTP you had then. When heart rate was recorded, 2 of 3 checks must pass: heart rate at 86% of max or more, heart rate within 5 bpm of LTHR or above, power at 1.15 × eFTP or more.
- 2 more reads join it: a power-to-heart-rate line read at your max heart rate, and a threshold read, (12.35 × FTP + 300) ÷ kg ÷ 0.80 (Joyner & Coyle 2008). Each is weighted by how tight it is, and the threshold read is capped at 25% of the blend.
- The headline moves at most 1.0 a week, except straight up to a new maximal effort. An older max eases down only on evidence from your power at matched heart rate, capped at 15% (Coyle 1984).
- Divided by today's body mass. Power-meter rides only.
Percentile
- Compared with the FRIEND registry norms by age decade and sex (Kaminsky 2015): lab treadmill and bike tests, people of every fitness level.
- Each band is modeled as a normal curve with a spread of 8 for men and 7 for women. Since VO2max is an estimate, the percentile is shown as a range.
Every plan rule says what stands behind it.
Each rule in a Peak Hybrid plan spec carries 1 of 4 labels. Where coaching practice fills a gap the trials leave, the rule says so, and its number is a starting value to tune.
High
Several trials or a meta-analysis in a comparable population.
Moderate
1 to 3 trials or 1 large cohort, or good trials in a different population (elite, young, novice).
Low
Small or indirect studies, or reviews that say the data are thin.
Practice-based
Coaching convention with no direct trial. The constant is a starting value to tune with testers.
| Plan | Rule | Source | Level |
|---|---|---|---|
| Strength for Cyclists | 2 heavy leg sessions a week, main lift 3 sets of 4 to 6 | Rønnestad 2011; Vikmoen 2016; Aagaard 2011; Rønnestad 2015; Sunde 2010; Rønnestad & Mujika 2014 | Moderate |
| Strength for Cyclists | Trap bar deadlift as the main lift of Day B, at every age | Swinton 2011 | Moderate (mechanics only) |
| Strength for Cyclists | Race season keeps 1 heavy, low-volume leg dose a week | Rønnestad 2010; Spiering 2021; Mujika 2016 | Moderate to low |
| Strength for Cyclists | Heavy leg days placed away from the day before a key ride or run | Doma 2017 | Moderate |
| 5K / 10K | No run longer than 110% of your longest run in the last 30 days | Frandsen 2025 | Moderate |
| 5K / 10K | 1 key run a week beside an easy run and a long run; no plan is threshold-only | Seiler 2010; Stöggl & Sperlich 2014; Neal 2013; Rosenblat 2019 | High for mostly easy over mostly threshold |
| 5K / 10K | 1 hard endurance session a week at 4 sessions or fewer, because heavy legs add a 3rd hard stress | Seiler 2010; Seiler 2013; Helgerud 2007 (trained athletes who did not lift) | Practice-based for hybrid athletes |
| 5K / 10K | Taper: cut volume, keep intensity and the number of sessions | Bosquet 2007; Wang 2023; Mujika & Padilla 2003 | High for the principle; the 7-day length is practice-based |
| 5K / 10K | Week 1 holds at 130% of your recent weekly run minutes or less | Nielsen 2014 | Practice-based (low evidence) |
Every citation on this page.
Listed as the app's research notes record them. Methods named by author alone: Coggan power levels, Friel heart-rate zones, Swain's heart-rate reserve method, Daniels' oxygen cost and pace ratios, and Riegel's race-time formula.
Estimated FTP
- MacInnis, Thomas & Phillips 2019, IJSPP 14(1):38-45, PMID 29809063
- Mattioni Maturana et al. 2018, J Sci Med Sport
- Borszcz et al. 2018, PMID 29801189
- Karsten et al. 2021
- Nimmerichter / Prinz 2020, Frontiers in Physiology
- Spragg et al. 2023, J Sports Sci 41(10), PMID 37660315
- Sitko et al. 2022, Int J Sports Med, PMID 34749416
- Coyle et al. 1984, J Appl Physiol
- Mujika I, Padilla S. Detraining: loss of training-induced physiological and performance adaptations. Part I. Sports Med 2000;30(2):79-87. PMID 10966148
VO2max and heart rate
- Kaminsky et al., Mayo Clin Proc 2015 (FRIEND registry)
- Joyner & Coyle 2008, J Physiol, "Endurance exercise performance: the physiology of champions"
- Tanaka et al. 2001 (age-predicted max heart rate)
Zones
- Allen H, Coggan AR, McGregor S. Training and Racing with a Power Meter, 3rd ed. VeloPress 2019
- Friel J, LTHR from a 30 min test (TrainingPeaks, "Joe Friel's Quick Guide to Setting Zones")
Strength plans
- Rønnestad BR, Hansen EA, Raastad T. Scand J Med Sci Sports. 2011;21(2):250-259
- Vikmoen O, et al. Scand J Med Sci Sports. 2016;26(4):384-396
- Aagaard P, et al. Scand J Med Sci Sports. 2011;21(6):e298-e307
- Rønnestad BR, et al. Scand J Med Sci Sports. 2015;25(1):e89-e98
- Sunde A, et al. Maximal strength training improves cycling economy in competitive cyclists. J Strength Cond Res. 2010;24(8):2157-2165
- Rønnestad BR, Mujika I. Scand J Med Sci Sports. 2014;24(4):603-612
- Swinton PA, et al. J Strength Cond Res. 2011;25(7):2000-2009
- Rønnestad BR, Hansen EA, Raastad T. Eur J Appl Physiol. 2010;110(6):1269-1282
- Spiering BA, et al. 2021
- Mujika I, Rønnestad BR, Martin DT. Int J Sports Physiol Perform. 2016;11(3):283-289
- Doma K, Deakin GB, Bentley DJ. 2017
Run plans
- Frandsen JSB, et al. How much running is too much? Identifying high-risk running sessions in a 5200-person cohort study. Br J Sports Med. 2025;59(17)
- Seiler S. What is best practice for training intensity and duration distribution in endurance athletes? Int J Sports Physiol Perform. 2010;5:276-291
- Stöggl T, Sperlich B. Polarized training has greater impact on key endurance variables than threshold, high intensity, or high volume training. Front Physiol. 2014;5:33
- Neal CM, et al. Six weeks of a polarized training-intensity distribution leads to greater physiological and performance adaptations than a threshold model in trained cyclists. J Appl Physiol. 2013;114(4):461-471
- Rosenblat MA, Perrotta AS, Vicenzino B. Polarized vs. threshold training intensity distribution on endurance sport performance: a systematic review and meta-analysis of RCTs. J Strength Cond Res. 2019;33(12):3491-3500
- Seiler S, et al. Adaptations to aerobic interval training: interactive effects of exercise intensity and total work duration. Scand J Med Sci Sports. 2013;23:74-83
- Helgerud J, et al. Aerobic high-intensity intervals improve VO2max more than moderate training. Med Sci Sports Exerc. 2007;39(4):665-671
- Bosquet L, et al. Effects of tapering on performance: a meta-analysis. Med Sci Sports Exerc. 2007;39(8)
- Wang Z, et al. Effects of tapering on performance in endurance athletes: a systematic review and meta-analysis. PLoS One. 2023;18(5):e0282838
- Mujika I, Padilla S. Scientific bases for precompetition tapering strategies. Med Sci Sports Exerc. 2003;35(7):1182-1187
- Nielsen RØ, et al. Excessive progression in weekly running distance and risk of running-related injuries: an association which varies according to type of injury. J Orthop Sports Phys Ther. 2014;44:739-747