Editorial

Endurance Training Insights

Deep dives into training load metrics, AI coaching methodologies, and data-driven strategies verified by AthleteOS performance analysis.

Featured Guide
Featured

CTL, ATL, and TSB: The Mathematical Foundation of Training Load

CTL, ATL, and TSB are the three numbers that tell you whether to push harder or back off. Here's the math behind them and what they mean for your Ironman preparation.

AO
AthleteOS Data Science

Latest Articles

Running

Why Did I Hit the Wall at Mile 20? The Glycogen Math Behind Every Marathon Bonk

Hitting the wall at mile 20 isn't bad luck — your glycogen runs out in 90–120 min at marathon pace. Here's the exact math, and what to do next time.

Read article →
AO
AthleteOS Data Science
Running

What Pace Should My Long Runs Be? The Slower-Than-You-Think Marathon Long-Run Rule

Your long run should be 10–20% slower than marathon goal pace — and for runners targeting 4:30+, the standard 90-sec/mile rule may actually be too slow.

Read article →
AO
AthleteOS Coaching Team
Cycling

How to Take Your First FTP Test: 20-Minute, Ramp, or 8-Minute — Which Is Right for You

Ramp test for most beginners, 20-minute test once you can pace hard efforts. Here's the correction factor behind each protocol and exactly how to execute it.

Read article →
AO
AthleteOS Coaching Team
General Endurance

Why Does My VO2max Decrease as I Age? And What You Can Actually Do About It

VO2max falls ~10%/decade if you stop pushing hard, but masters athletes who keep doing intervals drop only ~5%/decade. Here's the science and the fix.

Read article →
AO
AthleteOS Data Science
General Endurance

How Do I Calculate My Max Heart Rate? Why 220-Age Is Wrong

The 220-age formula was never validated. Its standard error is 12.4 bpm — meaning 1 in 20 people are off by more than 24 bpm. Here's what to use instead.

Read article →
AO
AthleteOS Data Science
Running

Does AI Running Coaching Cause Injuries? The 2026 Data, the Runna Story, and What to Watch For

AI running apps aren't inherently dangerous — but the injury risk is real when load guardrails are absent. Here's the ACWR data, the Runna pattern, and 8 numbers every runner needs.

Read article →
AO
AthleteOS Data Science
General Endurance

How to Combine Strength + Endurance Training Without Killing Either: The Interference Window in 2026

Lift first, wait 3 hours, cap running at 2x/week — these three rules eliminate 90% of the interference effect between strength and endurance training.

Read article →
AO
AthleteOS Coaching Team
General Endurance

Training by Menstrual Cycle Phases: What the 2026 Research Actually Says (and What It Doesn't)

The 'lift heavy in follicular, deload in luteal' rule isn't evidence-based. When McNulty 2020 filtered to high-quality studies only, the effect size was -0.01 — functionally zero.

Read article →
AO
AthleteOS Data Science
Running

Carbon Plate Shoes for Heavy Runners: Is the 4% Real?

The original 4% study tested 64 kg runners at sub-4:20/km pace. If you weigh 85+ kg and race above 5:00/km, your real gain is closer to 1-2% — or nothing statistically significant.

Read article →
AO
AthleteOS Data Science
General Endurance

Garmin VO2max Accuracy vs Lab Testing: The 6.85% MAPE Truth (and Why It Matters)

Garmin VO2max hits 6.85% MAPE vs lab CPET — but that number hides a fitness-level trap: highly trained athletes face 10%+ error and a 6 ml/kg/min underestimate.

Read article →
AO
AthleteOS Data Science
General Endurance

How to Use HRV from Your Whoop or Oura to Actually Adjust Today's Training

Your Whoop or Oura collected HRV last night. Here's the exact rolling-baseline math and decision tree to turn that number into a specific session swap — with peer-reviewed cutoffs.

Read article →
AO
AthleteOS Data Science
Cycling

The Zone 3 Trap: Why Most Amateur Cyclists Lose a Full Season in the Grey Zone

Amateur cyclists spend 25–45% of training time in Zone 3 (76–90% FTP). Elites keep it under 11%. That gap explains a year of stalled fitness.

Read article →
AO
AthleteOS Data Science
Running

AthleteOS vs Pfitzinger Marathon Plan: AI Adaptation vs a Proven Static Block

Pfitzinger's 18/55 wins for structure and cost ($29.95 one-time). AthleteOS wins when you miss a week, when your HRV tanks, or when you belong to the 50% of runners whose physiology doesn't fit a pyramidal plan.

Read article →
AO
AthleteOS Coaching Team
Running

AthleteOS vs Garmin Coach: When Watch-Based Coaching Stops Being Enough

Garmin Coach ignores your HRV Status even when it hits 25. AthleteOS reads that signal and rebuilds your week — here's the exact threshold where free watch coaching fails.

Read article →
AO
AthleteOS Coaching Team
General Endurance

Best AI Coach App for Endurance Athletes in 2026: Which Actually Works

No single AI coach app wins for every athlete. Expert coaches rated AI plans from 2/5 (minimal input) to 4/5 (detailed input). Here's which app fits your training reality.

Read article →
AO
AthleteOS Data Science
Hyrox

Best App for Hyrox Athletes in 2026: 4 Options for Training the Run + Functional Hybrid

AthleteOS wins for unified hybrid load management. TrainingPeaks wins if you're endurance-first. Hevy and Strong only cover half the race.

Read article →
AO
AthleteOS Coaching Team
Triathlon

How to Self-Coach an Ironman in 2026: The Software Stack That Replaces a $300/mo Coach

Hire a triathlon coach and you'll spend $3,600 before race day. Here's the 2026 software stack — $35–$75/month — that covers 80% of what that coach actually does.

Read article →
AO
AthleteOS Coaching Team
General Endurance

How to Track CTL, ATL, and TSB Without Paying for TrainingPeaks Premium

TrainingPeaks Premium costs $19.95/month, but Intervals.icu gives you an identical PMC chart free. Here's every option, what each costs, and where AthleteOS fits.

Read article →
AO
AthleteOS Data Science
General Endurance

AthleteOS vs Intervals.icu: Which One Actually Coaches You?

Intervals.icu is free and full of data. AthleteOS tells you what to do with it. Here's which one you actually need, with pricing, feature tables, and a head-to-head verdict.

Read article →
AO
AthleteOS Data Science
Triathlon

CTL Benchmarks for Ironman: What Fitness Score You Need for Your Finish Time

A sub-10-hour Ironman typically requires a race-day CTL of 120. Sub-9 hours needs 145–150. Here are the benchmarks, the ramp rate rules, and TSB taper targets by athlete tier.

Read article →
AO
AthleteOS Data Science
Cycling

Sweet Spot vs Threshold Training: Which Builds More Watts in Less Time?

Sweet spot (88-94% FTP) and threshold (95-105% FTP) carry identical autonomic recovery costs per Seiler 2007 — but sweet spot lets you accumulate more total minutes before that bill comes due.

Read article →
AO
AthleteOS Coaching Team
Running

The 180 Cadence Myth: Your Stride Rate Isn't Broken

180 spm isn't a target — it's a population mean from elite race-pace data. A 6'2" runner at 162 spm is normal; forcing 180 raises injury risk and wastes energy.

Read article →
AO
AthleteOS Coaching Team
Hyrox

Hyrox Compromised Running: Why Your 5K Pace Is Useless in a Race

82% of athletes need 679m to re-establish normal running mechanics post-exercise — longer than a full Hyrox run. Your 5K pace predicts almost nothing about Run 5.

Read article →
AO
AthleteOS Coaching Team
Hyrox

Hyrox Running Fade: You're Protecting the Wrong Run

Run 5 — not Run 8 — is the slowest split in Hyrox per peer-reviewed data. Here's why, and how to pace all 8 runs to finish 5–10% faster.

Read article →
AO
AthleteOS Coaching Team
General Endurance

Training Past 50: The Physiology Has Changed — Your Plan Should Too

Endurance training alone does not preserve fast-twitch muscle fibers after 50 — Tøien 2023 found masters endurance athletes at 39.3% type II fibers, nearly identical to sedentary older adults at 35.0%.

Read article →
AO
AthleteOS Coaching Team
General Endurance

Is Pain Tolerance Trainable? What the Research Actually Says

Athletes tolerate pain 54% longer than non-athletes — not because they feel less, but because hard training builds a more efficient top-down suppression network. Six weeks of HIIT raises tolerance 41%.

Read article →
AO
AthleteOS Coaching Team
Triathlon

Gut Training to 120 g/hr: The Real Mechanism Isn't Glycogen Sparing

Training your gut to absorb 120 g/hr cuts muscle damage markers by 67% — not because it spares glycogen, but because it prevents breakdown. Here's the 4-week protocol.

Read article →
AO
AthleteOS Coaching Team
General Endurance

Sodium Bicarbonate for Endurance: The 1–3% Performance Hack (and the Sodium Math Nobody Mentions)

Sodium bicarbonate improves performance by 1.7% across 38 studies — but a standard 0.3 g/kg dose delivers ~5,700 mg of sodium, more than twice the US daily reference intake.

Read article →
AO
AthleteOS Coaching Team
Triathlon

Sodium for Heavy Sweaters: How to Calculate Your Real Replacement Need

Generic 300–600 mg/hr sodium guidelines fail the top 20–30% of sweaters. A 10x range in sweat sodium concentration means two athletes in the same race can need 4x different amounts.

Read article →
AO
AthleteOS Coaching Team
Running

Berlin Marathon Negative Split: The Pacing Math Behind Every World Record

Only 0.16% of Berlin finishers run a true negative split, yet 69% of sub-2:30 personal bests are negative splits. Here's the pacing math that separates them.

Read article →
AO
AthleteOS Coaching Team
Running

Return to Run After a Bone Stress Injury: The Evidence-Based 6-Phase Protocol

MRI grade predicts return-to-run timeline with r=0.554 — Grade 1 averages 42 days, Grade 4 averages 99 days. Skipping phases risks recurrence rates up to 29%.

Read article →
AO
AthleteOS Coaching Team
General Endurance

HRV Readiness: Why the Daily Score Is Noise and the 7-Day Trend Is the Signal

Your morning HRV score is dominated by noise — daily RMSSD swings 10–20% from alcohol, sleep position, and measurement error alone. The 7-day rolling trend is what actually predicts readiness.

Read article →
AO
AthleteOS Coaching Team
General Endurance

Concurrent Training Interference: What the Research Actually Shows (and What Hickson Got Wrong)

The interference effect everyone fears impairs strength gains, not VO2max — Wilson 2012 (422 effect sizes) shows concurrent training produces VO2max ES 1.41 vs endurance-only 1.37 (NS).

Read article →
AO
AthleteOS Coaching Team
General Endurance

Chest Strap vs Optical Heart Rate Monitor: The Three-Tier Accuracy Model

Wrist optical HR error reaches 16.5 bpm for dark skin at high intensity. Arm-band optical hits MAE 1.43 bpm. Here's the three-tier framework for choosing the right sensor.

Read article →
AO
AthleteOS Data Science
Running

Sub-3 Marathon: The Physiology, the Mileage, and the Workouts That Actually Work

Sub-3 requires a 63 ml/kg/min VO2max, 55–75 miles per week, and pyramidal intensity (67.5% Zone 2) — not the polarized model most plans prescribe.

Read article →
AO
AthleteOS Coaching Team
Triathlon

Brick Workouts for Triathlon: What the Research Says vs. What Most Athletes Do

Stewart 2022 found mechanical efficiency drops from 48.2% to 42.1% after 40km cycling — the transient 'rubber-leg' phase averages 679m. Here's how to train it out.

Read article →
AO
AthleteOS Coaching Team
Triathlon

Aerobic Decoupling: The Number That Tells You If Your Base Is Actually Working

Aerobic decoupling measures how much your heart rate drifts relative to pace or power. Below 5% means a race-ready aerobic base. Above 10% means your engine is leaking.

Read article →
AO
AthleteOS Data Science
General Endurance

Polarized vs Pyramidal Training: The Research Has a Surprising Answer

Two 2024/2025 meta-analyses found no significant difference between polarized and pyramidal overall (SMD=-0.06). The better question is which model fits your athlete level and event distance.

Read article →
AO
AthleteOS Coaching Team
General Endurance

VO2max Intervals: The 4x4 Rule and What Most Athletes Get Wrong

The Norwegian 4x4 protocol raised VO2max 7.2% in 8 weeks vs 1.8% for easy running. Most athletes do these sessions too often, too short, and at the wrong pace.

Read article →
AO
AthleteOS Coaching Team
Cycling

Zone 2 vs LT1: Why the 60–70% HRmax Definition Is Wrong

The '60–70% HRmax = Zone 2' shortcut is physiologically wrong. LT1 — your real Zone 2 ceiling — spans 69–94% HRmax across trained athletes. Here's how to find yours.

Read article →
AO
AthleteOS Coaching Team
Cycling

FTP in Cycling: What It Is, How to Test It, and Why It Matters

Functional Threshold Power anchors every cycling training zone and training stress calculation. Learn the testing protocols, benchmarks by athlete category, and how to use FTP to train smarter.

Read article →
AO
AthleteOS Data Science
Methodology

TSS vs TRIMP: Which Training Load Metric Should You Use?

TSS uses power data; TRIMP uses heart rate. For multi-sport athletes managing three disciplines, the choice of metric determines whether your training load model is accurate or dangerously misleading.

Read article →
AO
AthleteOS Coaching Team
General Endurance

Zone 2 Training: The Science Behind Slow Riding Making You Faster

Zone 2 is the physiological foundation of every elite endurance program. Here's the mitochondrial biology, the 80/20 data, and precise definitions that separate productive Zone 2 from glorified recovery.

Read article →
AO
AthleteOS Data Science
Cycling

What Is a Good FTP for Cycling? Benchmarks by Athlete Category

Is your 280W FTP impressive or mediocre? It depends entirely on your body weight, training history, and racing category. Here are the definitive W/kg benchmarks every cyclist needs to know.

Read article →
AO
AthleteOS Data Science

Tired of static training plans?

Sign up for AthleteOS. Your AI coach takes these exact metrics, analyzes your Garmin/Strava data, and adapts your schedule every single day.

Generate Your AI Plan
14-day free trial · No credit card required