Cardio Load
Whole-body aerobic/systemic demand
How Anfold estimates training load and calories burned.
Anfold estimates calories burned from activity-specific MET values (the Compendium of Physical Activities), adjusted for how hard you went and the terrain. Strength sessions count working sets and rest separately. Training load comes out as three numbers: Cardio Load for the whole body, Muscle Load Units (MLU) for each muscle, and Total Load, which blends the two.
These are estimates, not measurements. MLU and the load scores are Anfold’s own units, so they are most useful for comparing your own days and weeks.
The three numbers you’ll see
Muscle Load Units (MLU)
Anfold’s own per-muscle unit, about one hard set’s worth of load on a muscle, so running and lifting add up on the same scale
Total Load
Blended recovery demand that weights systemic and muscular stress (leaning more on cardio for short-term recovery)
These metrics are designed for daily understanding and week-over-week comparisons. They are unitless points derived from duration, intensity and modality.
More
Anfold separates the calculation for each kind of training from the numbers you see. Strength, endurance and mobility each use their own math, then convert into one per-muscle unit (MLU), so a run and a leg day add up on the same scale.
Total Load blends whole-body cardio stress with the average load across your muscles. It leans toward cardio, which drives short-term recovery more.
How the calculation is structured
Example: a 5 km run, for someone who weighs 78 kg. MET value and multipliers as Anfold uses them.
One calculation per kind of training
Separate formulas for strength, endurance and mobility, because each one loads the body differently.
One unit per muscle (MLU)
Each kind of training converts its load into Muscle Load Units with its own conversion factor, so a run and a leg day add up on the same scale.
Targets and interpretation
Weekly guidance per muscle (maintain, build, or a short harder block) plus an overall Total Load to help you plan recovery.
More
Separate scales avoid forcing endurance, strength and mobility into one formula. Each uses a method suited to how it loads the body, then converts to MLU so different kinds of training can be added together.
Targets are deliberately broad. They change with what you’re aiming for (maintain or build) and don’t assume elite training volumes.
What happens when you log a workout
When you log an exercise (text or structured entry), Anfold:
More
AI-generated muscle activations are normalized and validated against expected patterns, and physiologically implausible outputs are flagged.
Fallback activation profiles cover common modalities when AI context is insufficient, ensuring consistent per-muscle allocation across entries.
- Interprets the activity type, duration and intensity (the effort you describe)
- Applies a modality-specific method to compute loads (strength, endurance/sports, or mobility)
- Allocates per-muscle load using the activity context and muscle activation patterns
- Returns Cardio Load, MLU, and Total Load and aggregates them over time
How calories burned are estimated
Anfold estimates the calories each activity burns. It starts from activity-specific MET values and adjusts them for intensity, terrain and exercise type.
Strength training counts working sets and rest separately, with a much lower rate for rest. A typical 45-minute session lands around 150–250 kcal, rather than the 350 kcal or more you get by treating the whole session as continuous work. Compound lifts count higher than isolation exercises, and hard sets a little higher than easy ones.
MET-based estimate
Activity-specific MET values are adjusted by your RPE, terrain (uphill/downhill), and elevation gain for outdoor activities.
Strength blended approach
Active work (time under tension) uses a higher MET; rest periods use a lower MET. Compounds and accessories are weighted differently. Session averages are capped at realistic values to prevent inflated estimates.
Illustration. Work and rest vary with your sets, reps and rest.
How each kind of training is calculated
Strength and calisthenics (INOL-based)
What this captures: heavier sets, more volume, slower tempos, and movements with pronounced lowering phases increase load appropriately; involved muscle groups receive their share.
- Uses an INOL-style approach (Intensity Number of Lifts) to model mechanical demand
- Intensity is estimated from weight x reps or, for bodyweight work, from reps alone
- Handles barbell, machine, and bodyweight movements
- Supports accessories and compounds
- Accounts for hypertrophy-style volume, rep tempo (time under tension), and eccentric emphasis
- Systemic (cardio) component credits active work but not rest
- Per-muscle load is split using muscle activation shares for the exercise
More
Intensity estimation supports both loadxreps and reps-only cases for bodyweight. Rep tempo affects time-under-tension; eccentric emphasis increases mechanical stress. Systemic load considers active work only, excluding rest intervals.
Per-muscle distribution uses activation shares (AI or fallback) before conversion to unified MLU for cross-modality stacking.
Endurance, sports, circuits and HIIT (effort, terrain)
What this captures: longer and/or harder sessions raise systemic demand; different activities and terrains stress muscles differently.
- Sets intensity from your perceived effort (sRPE 1–10), mapped to a sensible intensity range
- Uses a bounded intensity curve that captures how effort compounds at higher intensities without numerical blow-ups
- Terrain- and modality-aware mechanics (e.g., uphill vs. flat, impact/eccentric differences) affect muscle stress
- Longer sessions add diminishing returns within the same workout
- Per-muscle loads use muscle groups to apportion mechanical demand from the session
More
Intensity comes from your session RPE, mapped to an intensity fraction and clamped to a plausible range, with a bounded intensity curve to prevent blow-ups at maximal effort.
Terrain mechanics and impact/eccentric differences (e.g., uphill vs flat, trail vs road) shift per-muscle stress; longer single sessions accrue diminishing returns to reflect fatigue within a workout.
Muscle-specific adjustments capture that, for example, uphill running stresses hip extensors more than flat terrain.
Same run, different ground
Terrain changes both numbers: the calories and where the load lands.
The baseline.
Flat: about 350 kcal burned. Glutes 0.52 MLU, Quads 0.50 MLU, Calf Complex 0.50 MLU, Hamstrings 0.31 MLU.
Example: the same 5 km run in 26 minutes at effort 6, for someone who weighs 78 kg. Multipliers as Anfold uses them; elevation gain adds more.
Mobility, flexibility and balance (active or passive)
What this captures: meaningful but lighter stimulus appropriate for low-intensity movement, with proper distinction between restorative and active mobility work.
- Uses a lighter, stability-focused intensity profile
- Distinguishes passive vs. active work; restorative stretching is credited lightly, active mobility earns more for muscular control
- Scales muscle and systemic load low, to match gentle work
More
Active vs passive work uses a stability-oriented intensity model suited to gentle exercise. Mobility adds small per-muscle and systemic loads and converts to MLU at lower rates, because gentle work is easy to recover from.
What Anfold uses from your entry
Your perceived effort (sRPE) maps to a sensible intensity range. Weight is required only when it materially affects a calculation (e.g., bodyweight strength).
Heart rate isn’t an input
Anfold doesn’t read heart rate from your entries or from devices. Intensity comes from the effort you describe.
More
Your sRPE is mapped to an intensity fraction and clamped to a physiologically reasonable range.
User weight is used for bodyweight strength estimates when it materially changes the result; if missing, those estimates are limited rather than guessed.
From your entry
- Activity type and category (e.g., running, strength training)
- Duration (minutes)
- Perceived effort (sRPE 1–10)
- Terrain (flat, uphill, downhill, trail or mixed) and elevation gain, for outdoor activities
- Strength-specific fields when relevant (sets, reps, load, bodyweight)
- Muscle groups engaged and their approximate contribution to the exercise
How load is split across muscles
- AI provides context-aware muscle activations for each entry; Anfold normalizes and validates them.
- Fallback patterns cover common activities when AI activations are not available.
- For endurance, terrain and modality influence which muscles carry more stress. For strength, activation splits credit across the muscles involved.
- The result is a clear weekly view of which muscles appear to have been trained the most.
More
Endurance distribution accounts for impact and eccentric differences across terrains (e.g., uphill vs downhill) so that local muscle stress reflects real-world mechanics.
Strength distribution uses activation shares to split per-set dose across muscles before unifying into MLU. Single-day caps catch outliers and keep weekly tallies realistic.
Where the load lands, per workout
Ran 5k in 26 min
- Glutes 0.5 MLU
- Quads 0.5 MLU
- Calf Complex 0.5 MLU
Leg day: squats and lunges
- Quads 2.0 MLU
- Glutes 1.6 MLU
- Adductors 0.9 MLU
Push-ups and pull-ups
- Upper Back 1.5 MLU
- Pectorals 1.4 MLU
- Triceps 1.1 MLU
Bouldering, 1 hour
- Forearms 2.2 MLU
- Upper Back 1.6 MLU
- Biceps 1.2 MLU
Example workouts. Values are illustrative; the app lists them per muscle.
Weekly targets and how to read them
- Cardio Load: Daily and weekly guidance reflects time x effort (from sRPE). Targets encourage steady, sustainable progress, not elite-only volumes.
- Muscle Load Units (MLU): Anfold’s weekly guideline per muscle is about 6–8 MLU to maintain and 8–12 MLU to build, roughly that many hard sets. These bands are Anfold’s own guideline, not an established standard. Daily guidance leaves room for recovery between muscle groups.
- Total Load: A weighted blend of cardio and average per-muscle load that leans toward cardio to reflect short-term recovery needs while still accounting for local muscle stress.
These are starting points, not prescriptions. If you’re training for something specific (a race, a hypertrophy block), aim higher or lower.
Tap workouts to add them to the week.
1 workout · About 350 kcal burned · Most load: glutes
Where the load landed, per muscle.
Example week. A run and a leg day add up on the same scale.
Anfold’s own guideline, not an established standard.
Examples
- A moderate run plus a short strength circuit can satisfy a typical daily Total Load without feeling overcooked.
- Two focused strength days for a muscle group, paired with a few quality cardio sessions, generally lands in a progressive “build” week for most people.
- Endurance-focused weeks still add MLU to the main muscles involved, so a running week shows up on your legs too.
More
Weekly guidance is expressed per muscle using broad bands (maintain/build/aggressive). A single-day cap prevents outliers from skewing trends and encourages distribution over “hero” sessions.
Total Load blends systemic and muscular stress to approximate overall recovery demand; it leans toward cardio for near-term fatigue while still accounting for muscle-level stress.
What you’ll see in the app
- For each entry: Cardio Load, MLU (per-muscle), Total Load, and the engaged muscle groups
- For daily/weekly views: totals and trends to help you balance training and manage recovery
- Total Load provides a single number to track overall training stress and recovery needs
Limitations and considerations
- These are practical training load points for personal tracking, not medical measurements
- Load depends on the quality and completeness of your inputs
- Intensity comes from the effort you report (RPE), so it is only as accurate as that estimate
- Bodyweight strength estimates need user weight; without it, strength load is limited
Please note
- Anfold is in beta and for personal tracking. It is not a medical device or a substitute for professional advice.
- If you have a health condition or notice worrying symptoms, talk to a professional before relying on the app.
- Results are only estimates and can vary according to your data and particular circumstances.
- Artificial Intelligence (AI) is used to interpret or fill gaps, which may be inaccurate or incomplete.
- Use your judgment. Don’t use the app for clinical decisions without consulting a professional.
Sources
These are the studies and methods the calculations draw on.
How this page is kept current
When the way Anfold calculates something changes, this page changes with it. Spotted something wrong? Write to [email protected].
Strength load
- Hristov, H. (2004). How to Design Strength Training Programs using Prilepin’s Table? X3MSoftware (online article, introduces INOL). Archived copy
- Epley, B. (1985). Poundage chart. In Boyd Epley Workout (p. 86). Lincoln, NE: Body Enterprises.
- Schoenfeld, B. J., Ogborn, D., & Krieger, J. W. (2017). Dose-response relationship between weekly resistance training volume and increases in muscle mass: a systematic review and meta-analysis. Journal of Sports Sciences, 35(11), 1073–1082. doi:10.1080/02640414.2016.1210197
Training load and intensity
- Banister, E. W. (1991). Modeling elite athletic performance. In J. D. MacDougall, H. A. Wenger, & H. J. Green (Eds.), Physiological Testing of the High-Performance Athlete (2nd ed., pp. 403–424). Champaign, IL: Human Kinetics.
- Foster, C., Florhaug, J. A., Franklin, J., et al. (2001). A new approach to monitoring exercise training. Journal of Strength and Conditioning Research, 15(1), 109–115. doi:10.1519/00124278-200102000-00019
- Impellizzeri, F. M., Rampinini, E., Coutts, A. J., Sassi, A., & Marcora, S. M. (2004). Use of RPE-based training load in soccer. Medicine & Science in Sports & Exercise, 36(6), 1042–1047. doi:10.1249/01.MSS.0000128199.23901.2F
- Chen, M. J., Fan, X., & Moe, S. T. (2002). Criterion-related validity of the Borg ratings of perceived exertion scale in healthy individuals: a meta-analysis. Journal of Sports Sciences, 20(11), 873–899. doi:10.1080/026404102320761787
- Fernández, D., Moya, D., Cadefau, J. A., & Carmona, G. (2021). Integrating external and internal load for monitoring fitness and fatigue status in standard microcycles in elite rink hockey. Frontiers in Physiology, 12, 698463. doi:10.3389/fphys.2021.698463
Terrain and impact
- Gottschall, J. S., & Kram, R. (2005). Ground reaction forces during downhill and uphill running. Journal of Biomechanics, 38(3), 445–452. doi:10.1016/j.jbiomech.2004.04.023
- Mizrahi, J., Verbitsky, O., & Isakov, E. (2000). Shock accelerations and attenuation in downhill and level running. Clinical Biomechanics, 15(1), 15–20. doi:10.1016/S0268-0033(99)00033-9
Calories burned
- Ainsworth, B. E., Haskell, W. L., Herrmann, S. D., et al. (2011). 2011 Compendium of Physical Activities: a second update of codes and MET values. Medicine & Science in Sports & Exercise, 43(8), 1575–1581. doi:10.1249/MSS.0b013e31821ece12
- Minetti, A. E., Moia, C., Roi, G. S., Susta, D., & Ferretti, G. (2002). Energy cost of walking and running at extreme uphill and downhill slopes. Journal of Applied Physiology, 93(3), 1039–1046. doi:10.1152/japplphysiol.01177.2001
- Scott, C. B. (2006). Contribution of blood lactate to the energy expenditure of weight training. Journal of Strength and Conditioning Research, 20(2), 404–411. doi:10.1519/00124278-200605000-00029
- Bloomer, R. J. (2005). Energy cost of moderate-duration resistance and aerobic exercise. Journal of Strength and Conditioning Research, 19(4), 878–882. doi:10.1519/00124278-200511000-00026
- Ratamess, N. A., Falvo, M. J., Mangine, G. T., et al. (2007). The effect of rest interval length on metabolic responses to the bench press exercise. European Journal of Applied Physiology, 100(1), 1–17. doi:10.1007/s00421-007-0394-y
- Boutcher, S. H. (2011). High-intensity intermittent exercise and fat loss. Journal of Obesity, 2011, 868305. doi:10.1155/2011/868305
Note: We avoid brand-specific devices or vendor dependencies; calculations rely on common fields you can enter manually.