Three factors set your race speed
High confidence
Three things explain most of why one runner is faster than another, from 1500m up to the marathon. One is how much oxygen the body can use (VO2max). The next is the share of that you can hold at race pace (lactate threshold). The last is running economy, or how little oxygen each mile costs. Improve any one of them and your race times usually drop.
Why it works
VO2max sets the ceiling on how much aerobic energy the body can make. Lactate threshold decides how much of that ceiling you can hold for the whole race. Running economy sets how much oxygen it costs to cover each unit of distance.
What it means in practice
This framework maps each workout to what it targets. Easy-aerobic and long runs build the cardiovascular base that supports VO2max and threshold. Interval work targets VO2max directly. Threshold and tempo runs raise the share you can hold at race pace. Strength and plyometric work (short, springy jumps) improve running economy.
The evidence
-
Joyner modeled marathon performance as the product of three physiological factors: VO2max sets the ceiling on aerobic energy, lactate threshold fixes the fraction of that ceiling sustainable over the race, and running economy sets the oxygen cost of a given speed. Feeding physiologically plausible upper bounds for each factor into the model yielded a theoretical optimum of roughly 1:57:58, about 8 minutes under the world record at publication (near 2:06). This is theory rather than data. It extrapolates from elite male values, pushes running economy to the extreme low end of observed values, and does not account for pacing, environment, or fatigue durability. The three-factor framing still anchors endurance physiology and later informed the Nike Breaking2 and INEOS 1:59 projects.
-
Joyner and Lundby argue that fixed individual differences in trainability have been overstated. Under more intense or progressive training, such as interval work and longer programs, VO2max rises in essentially all participants. That undercuts earlier reports of a substantial 'non-responder' fraction seen at modest training doses, where under-dosed protocols produce small mean responses with high apparent variability. The authors read this as a stimulus problem more than a genetic one, and call for new twin-study designs that separate true genetic limits from inadequate training.
-
Several intrinsic factors track with better running economy: preferred stride length or up to 3% shorter (but not longer); lower vertical oscillation; greater leg stiffness; lower lower-limb moment of inertia; less leg extension at toe-off; larger stride angles (the angle of the parabolic tangent of the center of mass at toe-off); ground reaction force aligned with the leg axis during propulsion; maintained arm swing; low thigh antagonist-agonist coactivation; and low lower-limb muscle activation through the propulsive phase. Moore cautions against prescribing one 'ideal' form, because individual variation is large and many of these features emerge naturally with training rather than from explicit cueing. Beginners who improved their economy over 10 weeks did so through kinematic changes (knee and ankle angle at toe-off), not a dramatic form overhaul.
-
Lactate-threshold gains stayed specific to the trained mode, while VO2peak gains transferred across modes; the pattern suggests peripheral, muscle-level threshold adaptation against central VO2peak gains that carry over. Run training raised VO2 at LT more on the treadmill (+58.5%) than the cycle ergometer (+30.3%), a significant mode difference. Cycle training raised cycle-ergometer VO2 at LT by 38.7% but produced no significant treadmill change (23.6 to 24.0 ml/kg/min). VO2peak rose 11.9% to 20.7% in both training groups regardless of test mode, and controls held flat; all of this comes from 16 untrained men over 10 weeks in 1990.
n=16
Why we call confidence high
Joyner's 1991 theoretical model has anchored endurance physiology for over three decades. Later research refined it, notably by adding resilience as a fourth factor, but it has not displaced the core three-factor structure.
Where it applies
Adult distance runners at every training level. The evidence is strongest in trained-to-elite male runners, but the underlying physiology applies to all runners.
Does not apply to: sprint athletes; ultra-endurance events where pacing, fueling, and resilience dominate.
Last reviewed Apr 30, 2026. See how we score.