Runima

Runima Team

Sub-4 Marathon: It's Volume, Not Talent

What research on 150,000+ marathons says it takes to break 4 hours: why mileage beats genetics, a timeline by starting point, and how to build the cycle.

Sub-4 Marathon: It's Volume, Not Talent
Here's the split that makes the sub-4 marathon strange: at the Berlin Marathon, across 873,334 finishers from 1999 to 2025, the men's median finish time is 3:57:46 — meaning more than half the male field already runs sub-4 without it being anyone's stretch goal. The women's median is 4:26:02. For men, four hours is basically the middle of the pack. For women, it's a genuinely fast, top-quartile time. Same barrier, two very different climbs — and neither one is really about talent.

That asymmetry is the tell. Sub-3 is a sub-elite physiological standard — VO2max, lactate threshold, and running economy all have to clear a real bar on the same day. Sub-4 is a different kind of problem. The research keeps landing on the same answer: the single best predictor of whether you break four hours isn't your engine size, your threshold, or your economy — it's how many easy kilometers you've banked in the months before race day. This is a research-backed answer for what that actually looks like, with every claim rated for how solid the evidence is.

Why volume beats talent at this pace

The case for volume-as-destiny doesn't rest on one study — three independent lines of evidence converge on it.

The training-only model

Giovanni Tanda's model predicts marathon finish time from 8 weeks of training data alone — no race required: Pace = 17.1 + 140.0·exp(−0.0053·weekly km) + 0.55·average training pace. Built from 22 runners across 46 marathons, weekly mileage and training pace together explained ~77% of the variance in finish time, independent of body composition. It's held up well enough that a companion analysis later confirmed body-fat percentage only matters as a secondary factor once volume and pace are accounted for.

The 119,452-runner reality check

Muniz-Pumares et al. (2025) analyzed 16 weeks of Strava training data preceding 151,813 marathons. Average volume across the whole population was 45.1 ± 26.4 km/week — but runners finishing 2:00-2:30 averaged ~107 km/week, while runners finishing over 4:00 averaged just ~35 km/week. The gap wasn't harder workouts. It was mostly easy-paced mileage, and the fastest runners built it primarily by adding more Zone 1 running, not more intensity.

The Dutch cohort's hard numbers

Fokkema et al. (2020) tracked 441 marathon and 556 half-marathon runners and found volume under 40 km/week predicted a slower finish (β 6.33), over 65 km/week predicted a faster one (β −14.09), and a longest run under 25 km predicted a slower finish (β 13.44) — all independent of each other. Just as notably: higher volume was not associated with more injuries in this cohort.

The volume story holds at every scale researchers have checked it — a 22-runner physiological model, a 997-runner Dutch cohort, and a 119,452-runner Strava dataset all point the same direction.

What does a recent race actually tell you?

DistanceApprox. equivalent for sub-4 (3:59:xx)
Marathon pace5:41/km (9:09/mile)
Half marathon~1:52-1:55
10K~50-52 min
VDOT~40-42

Those numbers are a floor, not a target. Riegel's classic exponent (1.06) and VDOT tables are accurate to within 2-5% between adjacent distances, but they systematically over-predict marathon capability, because they don't model glycogen depletion, pacing inexperience, or the "wall." Vickers & Vertosick's analysis of roughly 2.5 million NYC Marathon finishes found the real population-average half-to-marathon multiplier is closer to 2.14, not Riegel's 2.11 — and traditional formulas are only about 80% accurate, meaning roughly 1 in 5 runners miss their predicted time by a meaningful margin. Build in a 5-10% buffer if you're a first-timer, more if your longest run sits under 25-30 km.

Plug a recent result into the Race Time Predictor for your own equivalents, and use the Race Time Prediction cheatsheet for the one-screen version of how these formulas actually work.

The VDOT-40 trap

Here's the detail that explains why so many fit runners still miss sub-4: at VDOT 40, your lactate-threshold pace is already ~5:41/km — the exact pace a sub-4 marathon demands, for 42.2 km, not a 20-minute tempo. A raw VDOT-40 runner is essentially being asked to race a marathon at threshold effort, which isn't sustainable. That's why the safer readiness band is VDOT 42-44 (10K ~48-50 min, half ~1:48-1:52), where marathon pace sits comfortably below threshold rather than right on top of it. Fitness on paper and fitness that survives 42 km aren't the same thing — which is exactly why the big-three physiology model (VO2max × threshold utilization × running economy — 72.1% of marathon variance) now has a fourth term. Andrew Jones's 2024 "fourth dimension" paper and a 2025 study running well-trained marathoners for 90 and 120 minutes both found VO2max, threshold utilization, and running economy all deteriorate as the race goes long — "durability" now sits alongside the classic three, and it's exactly why short-race predictors flatter you.

The training that gets you there

No single branded plan — Daniels, Pfitzinger, Hansons, Higdon — has been crowned by a head-to-head trial. A quantitative analysis of 92 sub-elite 12-week plans found peak weekly volumes cluster at 108 km (high-volume tier), 59 km (middle tier), and 43 km (low tier) — sub-4 runners belong in that middle band, building toward 55-70 km/week at peak.

On intensity distribution, the field mostly agrees on ~80% easy, ~20% at or above threshold (Stöggl & Sperlich, 2014). But it's worth knowing the honest counter-evidence: Festa et al. (2020) put 38 recreational runners through 8 weeks of either an 77/3/20 polarized split or a 40/50/10 "focused endurance" split matched for total load, and found statistically similar gains — with the focused-endurance group saving 17% of training time. For a time-limited recreational runner, the takeaway isn't "hit 80/20 exactly." It's simpler: run easy often, run enough of it, and don't stress the precise ratio.

Sample sessions, anchored to VDOT ~42 paces:

Session typePaceExample
Easy~6:45-7:15/kmThe bulk of weekly volume, conversational
Marathon-pace long run5:41/kmProgress from 24 km toward 32-35 km; every 2-3 weeks, finish 12-16 km at goal pace
Threshold~5:20-5:35/km2×20 min with 3 min jog, or 5×2 km cruise intervals
Marathon-specific5:41/km16-20 km within a medium-long run, or 3×5 km at MP with 1 km float
VO2max (sharpening)~4:50-5:05/km5-6×1000 m with equal jog recovery

A peak week (~60 km) runs 5-6 days: one threshold session, one marathon-pace or medium-long run, easy days between, long run on the weekend.

On progression: the popular 10% rule has no RCT support. The GRONORUN trial put 532 novice runners through either an 8-week standard program or a 13-week graded program built explicitly on the 10% rule, and found no reduction in injuries (HR 0.8, 95% CI 0.6-1.3). Treat it as a conservative heuristic for avoiding sudden spikes, not an evidence-based safety net — build volume gradually with a deload week roughly every 3-4 weeks (cut ~20-30%).

Taper, strength, and fuel

The taper

A 27-study meta-analysis found the optimal taper is 2 weeks, cutting volume 41-60% exponentially while holding intensity and frequency steady — worth roughly a 2-3% performance gain. A follow-up meta-analysis of 14 studies confirmed it, with the largest effects at 8-14 days. A 5-day taper is not enough; don't cut intensity, only volume.

Strength training

Balsalobre-Fernández's meta-analysis and Blagrove's systematic review both confirm real economy gains from heavy lower-body work; a 2024 update found combined heavy-load-plus-plyometric methods work best, with plyometrics especially useful below 12 km/h — squarely in sub-4 territory. A 2025 RCT found 10 weeks of strength work also improved economy durability across 90 minutes of running — directly relevant to how you feel at kilometer 35. Aim for 2×/week, ~40-70% 1RM plus plyometrics.

Fueling

Consensus is 60-90 g/hour of carbohydrate for events over 2.5 hours. A 2025 study of 160 Seville Marathon finishers found runners who hit that range were significantly more likely to finish under 3 hours (p=0.035) — yet actual mean intake was just 35 ± 17 g/h. Most runners badly under-fuel; target 60-80 g/h for a sub-4 pace. Carb-load at 8-12 g/kg/day for 36-48 hours beforehand — a classic protocol study found a single day at ~10 g/kg plus rest raised muscle glycogen from 95 to 180 mmol/kg wet weight and held it there. No depletion phase needed.

The injury trap: watch the signal, not the slogan

A systematic review of 23,047 runners across 36 studies puts overall running-injury incidence at 26.2% — 14.9% in novices, 26.1% in recreational runners, a striking 62.6% in competitive runners. Prospective studies flag weekly distances above roughly 30 km, 64 km, or 60-70 km as raising risk, which is an uncomfortable fact: the same volume that gets you to sub-4 also raises your exposure, so it has to be built patiently, not accumulated in a rush.

Two popular interventions that failed in RCTs: the 10% rule (no injury reduction), and a multifactorial online injury-prevention program tested on 2,378 runners, which showed identical injury rates in the intervention and control groups (37.5% vs 36.7%, OR 1.08). What did work: strength training, reducing acute injuries by 35% and overuse injuries by nearly half (RR 0.527, 95% CI 0.373-0.746) across 26,610 people. Smaller but real: increasing cadence 5-10% cut knee-joint energy absorption 20-34% in a controlled biomechanics study.

Practical translation: manage total load, prioritize consistency over heroic long-run weekends, address old injury sites specifically, lift twice a week, and respect pain rather than run through it.

How long will this actually take you?

Large single-race datasets put the overall average marathon finish somewhere between 4:20 and 4:45. Sub-4 sits faster than that average for both sexes and roughly in the top third of a big-city field — even while, per the Berlin numbers above, it's below the male median.

Starting pointRealistic timelineWhy
Complete beginner (couch)~18-24 monthsFinish one marathon first (typical first-timer: 4:30-5:30), build a durable base, then target sub-4 in a later cycle
Base fitness, no marathon yet (half ~1:50-1:55)16-20 weeks, sometimes two cyclesOften sub-4-ready off a single well-structured build
Already finishing 4:05-4:25Single 16-20-week cycleThe highest-probability group — add volume to ~55-65 km/week plus marathon-pace work
Masters (50+)Achievable, longer build-upRecreational performance declines meaningfully from ~50; needs more recovery, not less ambition
Age is a real headwind but not a wall. Zavorsky, Tomko & Smoliga (2017) analyzed NYC, Boston, and Chicago age-group winners from 2001-2016 and found peak performance at 28.3 years (men) and 30.8 years (women) — but also that men's age-group decline runs a gentler ~2:06/year after 35, versus ~2:33/year for women. The same 873,334-runner Berlin dataset that gave us the median split above also found men are about twice as likely to "hit the wall" — a ≥20% second-half slowdown — which matters more to a sub-4 attempt than most people assume.

Tracking readiness over the cycle

Field tests. Retest a 5K/10K or threshold benchmark every 4-8 weeks and recompute VDOT or the Tanda prediction — it should climb 1-3 points per cycle if training is working. Treat any short-race prediction as an upper bound, not a guarantee: a controlled 16-20 km run at 5:41/km is a better readiness signal than any 10K calculator.

Training-load models. CTL/ATL/TSB — rooted in Banister's fitness-fatigue model — are useful trend tools: build CTL ~3-5 points/week, and aim to arrive at the start line with TSB around +15 to +25. Treat the exact thresholds as platform convention, not physiology; they ignore sleep, nutrition, and life stress.

HRV-guided training has real evidence behind it. Vesterinen et al. (2016) found an HRV-guided group improved 3000 m speed significantly (+2.1%, p=0.004) while a matched predefined-training group's gain (+1.1%) fell short of significance — and the HRV group did it with fewer hard sessions. Cyclist trials by Javaloyes et al. (and a follow-up) found similar advantages in peak power and time-trial performance. Meta-analyses are more measured: Granero-Gallegos et al. found a small but real VO2max edge, while Manresa-Rocamora et al. and Düking et al. found the performance advantage often falls short of significance — but HRV-guided training reliably produces fewer negative responders. Net read: at least equivalent, probably more efficient.

Learn to increase your HRV and set your zones with the Heart Rate Zone Calculator, then let the Runima app surface the trend that matters — pace at a given effort, improving month over month.

Overtraining warning signs, per the ECSS/ACSM consensus statement: resting heart rate elevated more than 5 bpm above baseline, a suppressed or declining HRV trend, loss of motivation (one of the most reliable signals), unusually prolonged soreness beyond 72 hours, and performance stagnating at normal effort. Counterintuitively, functional overreaching has been linked to faster heart-rate recovery, not slower — a reminder that these signals need to be read as a pattern, not a single number.

The build, stage by stage

Stage 0 — Confirm readiness. A recent half marathon at 1:52-1:55 or better, or a Tanda-model prediction under 4:00 from current training. You should be able to complete a 28+ km long run and hold 5:41/km for 2×20 minutes without it feeling desperate. If you can't yet, spend a block building base first rather than forcing the cycle.

Stage 1 — Base (6-8 weeks). Build easy volume toward 45-55 km/week at ~6:45-7:15/km, roughly 80/20 distribution, long run progressing to 26-28 km, strength training twice a week.

Stage 2 — Build (6-8 weeks). Peak at 55-70 km/week. One threshold session weekly, one marathon-pace session progressing to 16-20 km at 5:41/km, long run to 32-35 km. Start practicing race fueling (60-80 g carbs/h) on the long runs — don't save it for race day.

Stage 3 — Peak and taper (final 2-3 weeks). A short burst of sharpening (5×1000 m at ~5:00/km), then the 2-week, 41-60% exponential taper. Carb-load 8-12 g/kg/day for the final 36-48 hours. Target TSB around +15 to +25.

Start at or just slightly slower than 5:41/km — even or slightly negative splits, not positive ones. Men especially should guard against that ~20% wall slowdown the Berlin data flags. Fuel early, before you feel like you need it; by the time hunger or fatigue shows up, you're already behind on glycogen. Rehearse the split plan with the Race Strategy Calculator before race week, not during it.

How solid is each claim, really?

TopicEvidence qualityNotes
Volume → performanceStrongConverges across a physiological model, a 997-runner cohort, and a 119,452-runner dataset
Taper protocolStrong (two meta-analyses)2 weeks, 41-60% volume cut, hold intensity
Strength training → economy & injuryStrong-moderateEconomy gains are consistent; injury reduction is one of the best-evidenced interventions available
Carbohydrate strategyStrongWell-established sports-nutrition science
Race-time prediction (Riegel/VDOT)Strong for method, moderate for marathon accuracySystematically over-predicts marathon time; ~1 in 5 miss significantly
Big-three physiology + durabilityStrong, durability construct newer2023-2025 literature on fatigue-related deterioration is still developing
10% rule / online injury programsDebunked in RCTsPopular but not supported; strength training outperforms both
Intensity distribution optimumModerate, contested"Mostly easy" is solid; the exact 80/20 split is not load-bearing for recreational runners
Timeline by starting pointWeak / indirectNo direct cohort tracks time-to-sub-4; synthesized from adaptation science and coaching consensus
HRV-guided trainingModerateReliable for trends and reducing negative responders; performance edge is small and inconsistent
Age & sex differencesStrong (large datasets)Berlin and NYC/Boston/Chicago cohorts are about as close to population-level truth as this field gets

The takeaway

Sub-3 asks three separate physiological systems to peak on the same day. Sub-4 asks something more mundane and, in a way, more democratic: show up consistently enough, for long enough, at mostly-easy paces, and the fitness accumulates on schedule. The data backs this up almost embarrassingly well — a formula built on nothing but your training log predicts your finish time about as well as a lab test would. That's not a knock on the goal. It means sub-4 is one of the most controllable big outcomes in distance running, and the runners who miss it are usually the ones who tried to out-clever the mileage rather than just log it.

Once four hours is behind you, sub-3 is where the physiology trilogy — VO2max, lactate threshold, and running economy — starts to matter far more than volume alone. For now: build the base, nail your training paces, and track the trend that tells you whether this cycle is actually working.

References

  1. Joyner MJ (1991). Modeling: optimal marathon performance on the basis of physiological factors. J Appl Physiol. 70(2):683-687.
  2. di Prampero PE et al. (1986). The energetics of endurance running. Eur J Appl Physiol. 55(3):259-266.
  3. Joyner MJ, Coyle EF (2008). Endurance exercise performance: the physiology of champions. J Physiol. 586(1):35-44.
  4. Coyle EF (2007). Physiological regulation of marathon performance. Sports Med. 37(4-5):306-311.
  5. Jones AM (2024). The fourth dimension: physiological resilience as an independent determinant of endurance exercise performance. J Physiol. 602(17):4113-4128.
  6. Zanini M, Folland JP, Blagrove RC (2025). The Effect of 90 and 120 Min of Running on the Determinants of Endurance Performance in Well-Trained Male Marathon Runners. Scand J Med Sci Sports. 35:e70076.
  7. Riegel PS (1981). Athletic records and human endurance. American Scientist. 69(3):285-290.
  8. Vickers AJ, Vertosick EA (2016). An empirical study of race times in recreational endurance runners. BMC Sports Sci Med Rehabil. 8:26.
  9. Tanda G (2011). Prediction of marathon performance time on the basis of training indices. J Hum Sport Exerc. 6(3):511-520.
  10. Tanda G, Knechtle B (2013). Marathon performance in relation to body fat percentage and training indices in recreational male runners. Open Access J Sports Med. 4:141-149.
  11. Muniz-Pumares D, Hunter B, Meyler S, Maunder ED (2025). The Training Intensity Distribution of Marathon Runners Across Performance Levels. Sports Med. 55(4):1023-1035.
  12. Fokkema T et al. (2020). Training for a (half-)marathon: Training volume and longest endurance run related to performance and running injuries. Scand J Med Sci Sports. 30(9):1692-1704.
  13. DeJong Lempke AF et al. (2025). Training Volume and Training Frequency Changes Associated with Boston Marathon Race Performance. Sports Med.
  14. Knopp M et al. (2024). Quantitative Analysis of 92 12-Week Sub-elite Marathon Training Plans. Sports Med Open. 10:50.
  15. Stöggl T, Sperlich B (2014). Polarized training has greater impact on key endurance variables than threshold, high intensity, or high volume training. Front Physiol. 5:33.
  16. Festa L et al. (2020). Effects of Different Training Intensity Distribution in Recreational Runners. Front Sports Act Living. 2:70.
  17. Bosquet L et al. (2007). Effects of tapering on performance: a meta-analysis. Med Sci Sports Exerc. 39(8):1358-1365.
  18. Wang Z et al. (2023). Effects of tapering on performance in endurance athletes: a systematic review and meta-analysis. PLoS ONE. 18(5):e0282838.
  19. Balsalobre-Fernández C, Santos-Concejero J, Grivas GV (2016). Effects of strength training on running economy in highly trained runners: a meta-analysis. J Strength Cond Res. 30(8):2361-2368.
  20. Blagrove RC, Howatson G, Hayes PR (2018). Effects of strength training on the physiological determinants of middle- and long-distance running performance: a systematic review. Sports Med. 48(5):1117-1149.
  21. Llanos-Lagos C et al. (2024). The effect of strength training methods on middle- and long-distance running economy: a systematic review with meta-analysis. Sports Med. 54(4):895-932.
  22. Zanini M, Folland JP, Wu C, Blagrove RC (2025). Strength Training Improves Running Economy Durability and Fatigued High-Intensity Performance in Well-Trained Male Runners: A Randomized Control Trial. Med Sci Sports Exerc.
  23. Jiménez-Alfageme R et al. (2025). Nutritional Intake and Timing of Marathon Runners: Influence of Athlete's Characteristics and Fueling Practices on Finishing Time. Sports Med Open. 11:26.
  24. Bussau VA, Fairchild TJ, Rao A, Steele P, Fournier PA (2002). Carbohydrate loading in human muscle: an improved 1 day protocol. Eur J Appl Physiol. 87(3):290-295.
  25. Fredette A, Roy JS, Perreault K, Dupuis F, Napier C, Esculier JF (2022). The Association Between Running Injuries and Training Parameters: A Systematic Review. J Athl Train.
  26. Buist I et al. (2008). No effect of a graded training program on the number of running-related injuries in novice runners: a randomized controlled trial. Am J Sports Med. 36(1):33-39.
  27. Fokkema T et al. (2019). Online multifactorial prevention programme has no effect on the number of running-related injuries: a randomised controlled trial. Br J Sports Med. 53(23):1479.
  28. Lauersen JB, Bertelsen DM, Andersen LB (2014). The effectiveness of exercise interventions to prevent sports injuries: a systematic review and meta-analysis of randomised controlled trials. Br J Sports Med. 48(11):871-877.
  29. Heiderscheit BC et al. (2011). Effects of step rate manipulation on joint mechanics during running. Med Sci Sports Exerc. 43(2):296-302.
  30. Seffrin A, Villiger E, Andrade MS et al. (2026). Sex differences in marathon pacing: analysis of 873,000 Berlin marathon runners reveals men are twice as likely to "hit the wall." Sci Rep. 16:19529.
  31. Zavorsky GS, Tomko KA, Smoliga JM (2017). Declines in marathon performance: sex differences in elite and recreational athletes. PLoS ONE. 12(2):e0172121.
  32. Vesterinen V et al. (2016). Individual Endurance Training Prescription with Heart Rate Variability. Med Sci Sports Exerc. 48(7):1347-1354.
  33. Javaloyes A et al. (2019). Training Prescription Guided by Heart-Rate Variability in Cycling. Int J Sports Physiol Perform. 14(1):23-32.
  34. Javaloyes A, Sarabia JM, Lamberts RP, Plews D, Moya-Ramon M (2020). Training Prescription Guided by Heart Rate Variability Vs. Block Periodization in Well-Trained Cyclists. J Strength Cond Res. 34(6):1511-1518.
  35. Granero-Gallegos A et al. (2020). HRV-based training for improving VO2max in endurance athletes: a systematic review with meta-analysis. Int J Environ Res Public Health. 17(21):7999.
  36. Manresa-Rocamora A, Sarabia JM, Javaloyes A, Flatt AA, Moya-Ramon M (2021). Heart Rate Variability-Guided Training for Enhancing Cardiac-Vagal Modulation, Aerobic Fitness, and Endurance Performance: A Methodological Systematic Review with Meta-Analysis. Int J Environ Res Public Health. 18(19):10299.
  37. Düking P et al. (2021). Monitoring and adapting endurance training on the basis of heart rate variability monitored by wearable technologies: a systematic review with meta-analysis. J Sci Med Sport. 24(11):1180-1192.
  38. Meeusen R et al. (2013). Prevention, diagnosis and treatment of the overtraining syndrome: Joint consensus statement of the ECSS and ACSM. Eur J Sport Sci. 13(1):1-24.
  39. Le Meur Y et al. (2013). A multidisciplinary approach to overreaching detection in endurance trained athletes. J Appl Physiol. 114(3):411-420.

This article is for general education and isn't medical advice. If you're new to distance running, returning from injury, or managing a health condition, clear a marathon build-up with your clinician before you increase volume or intensity.

Keep reading

Sub-3 Marathon: The 4% Club, and How to Join It

Sub-3 Marathon: The 4% Club, and How to Join It

Your First Marathon: It's Tendons, Not Fitness

Your First Marathon: It's Tendons, Not Fitness

Your Heart Rate Zones Are Probably Wrong

Your Heart Rate Zones Are Probably Wrong