VDOT Explained: The Number Behind Jack Daniels' Training Paces

VDOT is a fitness score derived from race performance, created by exercise physiologist Jack Daniels, PhD, and engineer Jimmy Gilbert in their 1979 work 'Oxygen Power: Performance Tables for Distance Runners.' The name comes from scientific notation: VO2 with a dot over the V (V̇O2) denotes the rate of oxygen consumption per minute, and Daniels' students shortened 'V-dot-O2' to simply 'VDOT.' Despite sharing units with VO2 max (ml/kg/min), VDOT is not a measurement of your oxygen uptake — it is the VO2 max you would need to run your race time if you had average running economy.

That distinction is the entire point of the system. Daniels and Gilbert analyzed elite runners and built two curves: one describing the oxygen cost of running at any speed (economy), and one describing what fraction of VO2 max a runner can sustain for any race duration (fractional utilization). Feeding a race time through both curves produces a single number that ranks performances across all distances on one scale. A 19:57 5K, a 41:21 10K, and a 3:10:49 marathon are all VDOT 50 — by definition, equivalent performances.

Because VDOT is anchored to what you actually ran, it absorbs every performance-relevant trait at once: aerobic capacity, running economy, lactate threshold, mental toughness on race day, even pacing skill. Daniels called it a measure of 'current running ability' rather than aerobic power. This is why VDOT became the backbone of his book Daniels' Running Formula — one of the most widely used training systems in the world — and why nearly every modern running calculator, including race predictors and training pace tools, is built on VDOT or a close descendant of it.

A laboratory VO2 max test measures the maximum rate at which your body consumes oxygen during a graded treadmill test to exhaustion, typically 30–85 ml/kg/min in trained runners. VDOT back-calculates a 'gross effective' value from race results. The two numbers rarely match, and the gap is informative. Daniels' own data on elites showed runners with identical 4:15-mile ability whose lab VO2 max values ranged from the low 60s to the high 70s — economical runners need less oxygen to hold the same speed, so they race faster than their lab number suggests.

Running economy — the oxygen cost of holding a given pace — varies by 20–30% among runners with similar VO2 max (Conley & Krahenbuhl, 1980). Fractional utilization, the percentage of VO2 max you can sustain for an hour, also varies with training. VDOT compresses all three variables into one practical number. If your lab VO2 max is 60 but your VDOT is 52, the difference is not an error: it quantifies how much performance you are leaving on the table through economy, utilization, or race execution — and it still prescribes the correct training paces for the runner you are today.

For training prescription, VDOT is the more useful number. Lab VO2 max tells you the size of your aerobic engine; VDOT tells you what the whole car does on the track. Two practical consequences follow. First, never plug a lab-tested VO2 max into a VDOT table — you will get training paces that are too fast for most runners, since lab values typically exceed VDOT by 5–15%. Second, improving VDOT does not require improving VO2 max: many masters runners hold or raise VDOT for years on a declining ceiling by improving economy and durability.

Daniels and Gilbert published the exact equations, and they are still used by most calculators today. The first equation estimates the oxygen cost of running at a given velocity v (in meters per minute) — the 'economy curve' fitted to data from trained runners:

v = velocity in meters/minute. Example: 268 m/min (4:30/km marathon pace) costs about 51.6 ml/kg/min.

The second equation estimates the fraction of VO2 max a runner can sustain for a race lasting t minutes — the 'drop-off curve.' It says a runner can hold roughly 100% of VO2 max for about 11 minutes, ~93% for 30 minutes, ~86% for 60 minutes, and ~80% for over 2 hours:

t = race duration in minutes. F is the sustainable fraction of VO2 max for that duration.

VDOT is the value that makes both curves agree with your race: take your average race velocity, compute its oxygen cost with the first equation, then divide by the sustainable fraction for your race duration from the second equation. VDOT = VO2(v) / F(t). Because the equations are smooth functions, the same VDOT can then be run backward to generate an equivalent time at any other distance and a pace for any training intensity. There is no lookup magic in the tables — they are simply these two equations evaluated and printed.

Worked example: a 41:21 10K is 241.9 m/min. The economy curve prices that velocity at about 45.5 ml/kg/min. The drop-off curve says 41.4 minutes is sustainable at about 91% of VO2 max. Dividing 45.5 by 0.91 gives a VDOT of 50. Every number in the famous tables traces back to this same two-step calculation.

The table below shows equivalent race performances across the most common distances. Find your most recent race time in its column; your row is your current VDOT, and every other time in that row is the performance the model considers equivalent. Recreational runners typically fall between VDOT 30 and 45, committed club runners between 45 and 60, and sub-elite/elite runners from 60 up to the mid-80s (a 2:05 marathoner is roughly VDOT 84).

Equivalent race times by VDOT (Daniels & Gilbert tables)

Two reading rules keep the table honest. First, enter it with your best recent performance — a race from the last 6–8 weeks, or a genuinely all-out time trial. A 5K you jogged with a friend produces a VDOT that will make every training pace too slow; a PR from two years ago produces paces too fast. Second, when races at different distances give different VDOT values (very common), Daniels' advice is to train from the higher value only if it is recent, and to read the gap as diagnostic: a 10K VDOT well below your 5K VDOT signals endurance, not speed, is the limiter.

The reason VDOT survived four decades is not the prediction table — it is the pace prescription. Daniels mapped five training intensities to physiological purposes, each anchored to a percentage of VDOT. Easy (E) pace, at 59–74% of VDOT (65–79% of max heart rate), builds aerobic base, capillarization, and mitochondrial density. Marathon (M) pace sits at 75–84%. Threshold (T) pace, at 83–88%, is 'comfortably hard' running designed to raise lactate clearance — roughly the pace you could race for an hour. Interval (I) pace, at 95–100%, targets VO2 max itself in repeats of 3–5 minutes. Repetition (R) pace is faster than VO2 max work and trains speed and economy with full recoveries.

Daniels' five training zones

The percentages matter more than the labels. Because every zone is indexed to the same VDOT, the system self-corrects as you improve: race a new PR, update your VDOT, and all five paces shift together. This is also why borrowing paces from a faster training partner fails — their T pace may be your I pace, which turns a controlled lactate-clearance session into a VO2 max workout with ten times the fatigue cost.

One subtlety Daniels emphasized and most runners miss: E pace is a range, not a floor to break through. Running easy days at the fast edge (or beyond) does not accelerate aerobic adaptation — the mitochondrial and capillary signaling from a 5:10/km easy run and a 5:45/km easy run is nearly identical for a VDOT 50 runner, but the faster version costs measurably more recovery. The 80/20 intensity-distribution literature reaches the same conclusion from a different direction: most elite training volume sits firmly inside E pace.

VDOT's equivalence rows double as a race predictor: run 19:57 for 5K, read across, and the table offers 41:21 for 10K and 3:10:49 for the marathon. For the 5K-to-10K and 10K-to-half jumps, studies and large-scale Strava data analyses find VDOT predictions track real outcomes well for trained runners — typically within 1–2%. The model is genuinely good at what it was built for: ranking and converting performances in the 12-minutes-to-2-hours range where its source data lived.

The marathon is the known exception. The equivalence assumes you are 'equally trained' for both distances, and almost nobody who just ran a sharp 5K is marathon-trained to the same degree. Analyses of recreational marathoners consistently show median results 10–20 minutes slower than their short-race VDOT equivalent, with low-mileage runners missing by the most. Alternative models make the same point differently: Riegel's power-law formula (T2 = T1 × (D2/D1)^1.06) and critical-speed-based predictions both flag that marathon performance depends on durability and fuel economy that short races never test.

The practical reading: treat cross-distance equivalents as a ceiling, not a forecast. A 24:08 5K makes 3:49:45 the marathon you could run with full marathon preparation — 60+ km weeks, regular long runs, fueling practice — not the marathon you will run off 5K training. Conversely, the equivalence table is excellent for setting honest goal paces: if your goal marathon requires VDOT 55 and your current 5K says VDOT 48, the gap is measured, not motivational.

VDOT is a model of a mid-pack 1970s-to-1990s trained male runner's physiology, generalized remarkably well — but it has edges. It knows nothing about conditions: the tables assume sea level, mild temperatures, flat firm ground, and even pacing. Heat above ~15°C measurably slows endurance performance; altitude above ~900 m reduces VO2 max roughly 6% per additional 1,000 m; trail terrain breaks the economy curve entirely. Race in those conditions and your computed VDOT underestimates your fitness; prescribe paces from a PR set in perfect conditions and your workouts run too hot on bad days.

It also compresses individual physiology into population curves. Runners with strong anaerobic capacity over-perform the tables at short distances and under-perform at long ones; high-durability runners do the opposite. The fractional-utilization curve is fixed, but real fatigue resistance — how little you slow late in long races — varies hugely and is trainable. This is exactly the gap newer two-parameter models like Critical Speed/D′ try to address by fitting your personal power-duration curve instead of a universal one. VDOT also has no concept of durability: two VDOT-55 runners can differ enormously in how well they hold pace after 90 minutes of work.

Finally, VDOT ages with the performance that produced it. Fitness decays within weeks of detraining, and a VDOT from last season is a memory, not a prescription. None of these limits make the system bad — they define its scope. VDOT remains the best single-number framework for converting race results into training paces; it just should not be mistaken for a physiology lab or a weather service.

The workflow is simple. Step one: race, or run an honest solo time trial — 5K is the most accessible distance and recent enough data matters more than distance choice. Step two: compute your VDOT and write down all five training paces. Step three: build your week so that easy runs sit inside the E range, quality sessions use T, I, or R paces as the plan demands, and nothing habitually drifts faster. Step four: re-test every 6–8 weeks or after a goal race, and move paces only when a performance proves the fitness.

Daniels' own conservative rule deserves repeating because it cuts against instinct: increase VDOT by at most 1 point per 4–6 weeks, even if workouts feel easy, and never raise it because a workout went well — workouts are supposed to go well. The single most common failure mode in self-coached VDOT training is 'pace creep': treating training paces as a test to beat rather than a dose to absorb. Hitting T pace exactly is the workout; beating T pace by 10 seconds per km is a different, harder workout that the plan did not call for.

Used this way, VDOT becomes a feedback loop rather than a fortune teller: race results update the number, the number disciplines training intensity, disciplined training produces the next race result. Our VDOT calculator implements the full Daniels-Gilbert equations — enter any recent race and it returns your VDOT, all five training paces, and equivalent times across distances, so the only judgment left to you is the honest input.