The Science Behind Strength Coaching Analytics

Every number on your coaching dashboard tells a story — but only if you understand the science behind it. Strength coaching analytics translate raw training data into actionable insights: estimated maxes, volume loads, compliance rates, exercise correlations, and trend detection. Each metric is built on published exercise science research, validated formulas, and statistical methods that have been refined over decades.

This guide walks through every analytics metric available in IronCoaching, explains the underlying formula, cites the original research, and shows you how to interpret the numbers for practical coaching decisions. Whether you are reviewing a client's e1RM trend, programming volume for a new block, or interpreting an AI-generated insight, this is the reference that connects the dashboard to the science.

If you want a quick-reference formula sheet, see the full analytics methodology documentation. For a broader overview of how analytics fit into coaching workflow, read the analytics product page.

What Is Estimated One-Rep Max (e1RM)?

Estimated one-rep max (e1RM) is a calculated prediction of the heaviest weight an athlete can lift for a single repetition, derived from a submaximal set. It is the single most important metric in strength coaching analytics because it tracks true strength independent of rep scheme changes.

The concept dates back to the mid-1980s when researchers sought a way to monitor strength progress without requiring athletes to test actual one-rep maxes — a practice that carries injury risk and requires significant recovery time. Two formulas dominate the field.

The Epley Formula

Published by Boyd Epley in 1985, the Epley formula is the most widely used e1RM equation in strength coaching:

e1RM = weight x (1 + reps / 30)

For example, an athlete who squats 120 kg for 5 reps has an estimated max of 120 x (1 + 5/30) = 140 kg. The formula is linear and tends to slightly overestimate at higher rep ranges (above 10). IronCoaching uses the Epley formula as its default calculation because of its simplicity, wide adoption, and strong correlation with actual 1RM tests at moderate rep ranges (Epley, 1985).

The Brzycki Formula

Matt Brzycki proposed an alternative in 1993:

e1RM = weight x (36 / (37 - reps))

The Brzycki formula produces slightly more conservative estimates than Epley, particularly at rep counts above 6. For the same 120 kg x 5 example: 120 x (36 / 32) = 135 kg — about 3.5% lower than Epley's prediction.

Strength progression chart showing estimated one-rep max trending upward

Choosing Between Formulas

Both formulas lose accuracy above 10 reps because muscular endurance increasingly influences the set before maximal strength becomes the limiting factor. For coaching purposes, track e1RM from sets of 1-8 reps to get the most reliable data. When reviewing client analytics, filter for compound lifts with rep counts in this range.

Practical Application

Coaches on the IronCoaching Expert tier see e1RM trends plotted over time for every tracked exercise. A rising e1RM on the squat, bench press, or deadlift — even when the athlete has not tested a true 1RM — confirms that the program is driving strength adaptation. A stalling e1RM signals the need for a programming change: new stimulus, deload, or volume adjustment.

How Is Training Volume Calculated?

Training volume quantifies the total mechanical work performed in a session, week, or training block. It is the primary driver of hypertrophy and a key fatigue management variable in strength programming.

There are two common ways to express volume, and they answer different coaching questions.

Volume Load (Tonnage)

Volume load = sets x reps x weight

A session of 4 sets of 8 reps at 100 kg produces a volume load of 4 x 8 x 100 = 3,200 kg. This metric captures the total mechanical load and is useful for tracking workload trends across mesocycles.

Hard Sets (Effective Volume)

The more modern approach, championed by researchers like Schoenfeld (2017) and Israetel (2019), counts the number of sets taken to within a few reps of failure — often called hard sets or effective sets. This method correlates more strongly with hypertrophy outcomes than raw tonnage because it accounts for the proximity-to-failure stimulus.

A meta-analysis by Krieger (2010) found that performing multiple sets per exercise produced 40% greater hypertrophy than single sets, with a dose-response relationship up to approximately 10 sets per muscle group per week for trained individuals.

IronCoaching's analytics dashboard tracks both tonnage and set counts. The frequency and volume chart breaks volume down by exercise or by muscle group — using the exercise-to-muscle-group mapping from the exercise library — so coaches can verify that each muscle group is receiving adequate weekly stimulus.

Weekly Volume Benchmarks

Research from Schoenfeld (2017) and Israetel (2019) suggests these weekly hard-set ranges for trained individuals:

• Minimum effective volume (MEV): 6-8 sets per muscle group
• Maximum adaptive volume (MAV): 12-20 sets per muscle group
• Maximum recoverable volume (MRV): 20-25+ sets per muscle group

Exceeding MRV leads to accumulated fatigue without additional adaptation. The volume zone coloring in IronCoaching flags sets in the green zone (10-20), yellow zone (below 10 or above 20), and red zone (above 25) to help coaches visually manage volume prescription.

What Does Program Compliance Mean?

Program compliance measures the percentage of prescribed training that the athlete actually completes. It is one of the most undervalued metrics in coaching analytics because even the most scientifically optimal program is worthless if the client only completes 60% of it.

Compliance is calculated as:

Compliance % = (completed sessions / prescribed sessions) x 100

A client prescribed 4 sessions per week who logs 3 has 75% compliance. IronCoaching tracks compliance automatically through the IronLedger integration — when an athlete logs a workout in the IronLedger app, it is matched against the assigned program to calculate adherence.

Why Compliance Matters More Than Programming

Research by Helms (2016) on trained powerlifters found that adherence to any reasonable program produced better outcomes than inconsistent adherence to a theoretically optimal one. The practical implication: a simpler 3-day full body program completed at 95% compliance will outperform a complex 6-day PPL split completed at 60%.

Compliance Thresholds

When compliance drops below 75%, the analytics signal is clear: the program needs to adapt to the athlete's real life, not the other way around. Check the client management dashboard for compliance trends before adjusting volume or intensity.

How Does Exercise Correlation Work?

Exercise correlation measures how strongly the performance of two exercises moves together over time. If a client's squat e1RM rises and their front squat e1RM rises at a similar rate, those exercises have a high positive correlation.

IronCoaching uses the Pearson correlation coefficient (r), introduced by Karl Pearson in 1895, to quantify this relationship:

r = covariance(X, Y) / (std_dev(X) x std_dev(Y))

The result ranges from -1.0 to +1.0:

• r = +0.8 to +1.0: Strong positive correlation — exercises improve together
• r = +0.5 to +0.79: Moderate positive — some shared strength qualities
• r = -0.5 to -1.0: Negative correlation — one improves as the other declines (rare, may indicate fatigue interference)
• r near 0: No meaningful relationship

Practical Use Cases

Coaches on the Expert tier can compare up to three exercises simultaneously in the exercise comparison chart. Common coaching questions this answers:

1. Transfer effect: Does improving a client's paused squat predict improved competition squat? A high correlation (r > 0.7) confirms the accessory is doing its job. 2. Fatigue interference: If deadlift e1RM drops while squat volume increases, a negative correlation suggests the squat volume is interfering with deadlift recovery. 3. Exercise redundancy: Two exercises with r > 0.9 may be redundant. The coach can drop one and allocate volume elsewhere.

The statistical significance of the correlation depends on sample size. IronCoaching requires a minimum of 8 data points per exercise before displaying correlation insights, following standard statistical practice (Cohen, 1988).

What Is RPE Accuracy and Why Does It Matter?

RPE (Rate of Perceived Exertion) is a subjective intensity scale where athletes rate how hard a set felt. The modified Borg RPE scale for resistance training (Borg, 1982; Zourdos, 2016) ranges from 1 (very easy) to 10 (maximal effort). RPE accuracy measures how well an athlete's subjective RPE ratings align with their actual performance data.

For example, if an athlete reports RPE 8 on a set of 5 (meaning 2 reps in reserve), but their e1RM calculation suggests they had 4 reps in reserve, their RPE accuracy is low — they are overestimating effort. Conversely, athletes who consistently under-report RPE may be training harder than prescribed.

Why It Matters for Coaches

Zourdos (2016) validated the modified RPE scale for powerlifting and found that trained athletes are generally accurate within 1 RPE point, but beginners can be off by 2-3 points. Tracking RPE accuracy over time serves two purposes:

1. Calibration: Athletes get better at rating effort when they receive feedback on accuracy. The IronCoaching analytics panel shows RPE accuracy trends that coaches can review with clients. 2. Autoregulation confidence: If an athlete's RPE accuracy is high (within 1 point), the coach can confidently prescribe RPE-based programs. If accuracy is low, percentage-based programming is safer. Learn more in the RPE vs RIR guide.

Calculating RPE Accuracy

IronCoaching estimates actual RPE from the relationship between performed reps, weight, and e1RM:

Estimated RPE = 10 - (estimated reps remaining)

Where estimated reps remaining is derived from the Epley formula run in reverse. The difference between reported RPE and estimated RPE is the accuracy gap. A rolling average of this gap over the last 20 sets gives the athlete's RPE accuracy score.

How Are Strength Trends Detected?

Trend detection identifies whether an athlete's strength on a given exercise is improving, plateauing, or declining. Rather than looking at individual sessions — which are noisy due to daily readiness variation — IronCoaching applies statistical smoothing to identify the underlying direction of change.

Moving Average

The simplest approach: a rolling average of e1RM over the last N sessions (typically 4-6). This smooths out day-to-day variation and reveals the general trajectory.

Linear Regression

A more robust method fits a line to the e1RM data points over a defined window (typically 4-8 weeks). The slope of the regression line quantifies the rate of change:

• Positive slope: Strength is increasing
• Zero slope: Plateau
• Negative slope: Strength is declining

The R-squared (R²) value indicates how well the trend line fits the data. An R² above 0.7 means the trend is reliable; below 0.3 suggests too much noise for a clear conclusion.

Practical Application

AI insights on the Expert tier use trend detection to generate coaching recommendations automatically. A declining trend on a main lift triggers a suggestion to review volume, recovery, or exercise selection. A plateau lasting more than 3 weeks prompts a periodization adjustment recommendation. Coaches can review these automated insights alongside the raw data in the client detail view.

Training volume distribution heatmap showing muscle groups and workout frequency

What Are Optimal Training Volume Zones?

Volume zones are color-coded ranges that indicate whether a muscle group is receiving too little, enough, or too much weekly training volume. They translate the research on dose-response volume relationships into a visual coaching tool.

The concept builds on the volume landmark framework popularized by Israetel (2019) and supported by Schoenfeld's (2017) meta-analytic findings on weekly set volume and muscle growth.

Zone Definitions

These zones are general guidelines derived from meta-analyses of trained populations. Individual variation is significant: a well-recovered athlete with years of training history may thrive at 22 sets per week for quads, while a newer lifter overreaches at 14 sets.

How IronCoaching Implements Volume Zones

The frequency and volume chart calculates weekly hard sets per muscle group by:

1. Mapping each logged exercise to its target muscle groups using the exercise library 2. Counting sets where reported RPE is 7 or higher (or estimated RPE if not reported) 3. Coloring the bar chart according to the zone thresholds above

Coaches can toggle between per-exercise view and per-muscle-group view. The muscle-group view is particularly useful for identifying imbalances — for example, a client with 18 sets of pushing volume but only 8 sets of pulling volume per week.

Adjusting Zones for Individual Athletes

The default thresholds work for most intermediate athletes, but coaches should adjust based on:

• Training age: Beginners respond to fewer sets; advanced athletes need more
• Recovery capacity: Sleep, nutrition, stress, and age affect MRV
• Training phase: Accumulation blocks may intentionally push into yellow/red before a deload
• Drug-free status: Natural athletes generally have lower MRV than enhanced athletes

How Do Prediction Accuracy and Retention Metrics Work?

Prediction accuracy tracks how well AI-generated predictions match actual outcomes. When IronCoaching's AI predicts that a client will hit a 150 kg squat within 4 weeks, and the client actually hits 147.5 kg, the prediction accuracy is calculated as:

Self-Improving Predictions

The AI system uses a feedback loop inspired by few-shot learning principles. When a coach rates an AI insight as helpful (thumbs up), that insight becomes a few-shot example for future prompts — the system literally learns from the coach's feedback (Phase 3A). Prediction accuracy history is also injected into prompts so the AI can calibrate its confidence levels. Over time, this produces increasingly coach-specific recommendations.

Retention Metrics

Client retention is tracked on the business dashboard (Pro tier and above):

• Churn rate: Percentage of clients who leave per month
• Average tenure: Mean duration of coach-client relationships
• Activity rate: Percentage of clients who logged at least one workout in the last 14 days

The retention trend chart shows a 6-month rolling view. Combined with the analytics data, coaches can correlate retention with programming quality — for instance, clients with higher compliance rates and visible strength trends tend to stay longer. Exploring the marketplace can also help coaches understand competitive positioning and retention benchmarks.

Statistical Confidence

Prediction accuracy requires a minimum sample size to be meaningful. IronCoaching follows Cohen's (1988) guidelines for effect size interpretation:

This prevents coaches from over-interpreting accuracy metrics based on a handful of predictions. The system becomes more reliable as data accumulates over months of coaching.

References

• Borg, G. (1982). Psychophysical bases of perceived exertion. *Medicine and Science in Sports and Exercise*, 14(5), 377-381.
• Brzycki, M. (1993). Strength testing: predicting a one-rep max from reps-to-fatigue. *Journal of Physical Education, Recreation & Dance*, 64(1), 88-90.
• Cohen, J. (1988). *Statistical Power Analysis for the Behavioral Sciences* (2nd ed.). Lawrence Erlbaum Associates.
• Epley, B. (1985). Poundage chart. In *Boyd Epley Workout*. Body Enterprises.
• Helms, E. R., et al. (2016). Application of the repetitions in reserve-based rating of perceived exertion scale for resistance training. *Strength and Conditioning Journal*, 38(4), 42-49.
• Israetel, M., Hoffmann, J., & Smith, C. W. (2019). *Scientific Principles of Hypertrophy Training*. Renaissance Periodization.
• Krieger, J. W. (2010). Single vs. multiple sets of resistance exercise for muscle hypertrophy: a meta-analysis. *Journal of Strength and Conditioning Research*, 24(4), 1150-1159.
• Pearson, K. (1895). Notes on regression and inheritance in the case of two parents. *Proceedings of the Royal Society of London*, 58, 240-242.
• Schoenfeld, B. J., et al. (2017). Dose-response relationship between weekly resistance training volume and increases in muscle mass. *Medicine and Science in Sports and Exercise*, 49(3), 661-671.
• Zourdos, M. C., et al. (2016). Novel resistance training-specific rating of perceived exertion scale measuring repetitions in reserve. *Journal of Strength and Conditioning Research*, 30(1), 267-275.