Riegel vs VDOT: which equivalent to trust
Two well-known methods turn a race result into an equivalent time at another distance: Riegel's simple power law and the VDOT oxygen-cost model. They usually agree between neighbouring distances and drift apart as you extrapolate. Here is how each one works, where they disagree, and which to lean on for a given question.
Open the Race Time PredictorThe Race time predictor runs the Riegel calculation below and lets you adjust the fatigue exponent, so you can see how close it lands to a VDOT-style equivalent for your own result.
Two ways to get an equivalent time
An equivalent time answers a simple question: given what you ran at one distance, what is the comparable performance at another? Two methods dominate, and they start from very different places.
Riegel is a power law. In his 1981 article Athletic Records and Human Endurance, Peter Riegel fitted a curve to race and record data and found time scales with distance by a near-constant exponent:
- T1
- your known time at the reference distance
- D1
- the reference distance
- D2
- the target distance you want an equivalent for
- 1.06
- Riegel's fatigue exponent — a single value fitted across many runners
That is the whole model: one exponent, applied to everyone. It needs no physiology and no tables, which is exactly why it is so widely used.
VDOT takes the opposite route. Built on the Daniels & Gilbert oxygen-cost model and popularised in Daniels' Running Formula, it estimates the rate of oxygen consumption (a VO2 value) that a race performance implies, then reads off the equivalent performances — and matching training paces — that the same physiology would produce at other distances. It was fitted to trained and elite runners, so it encodes how the cost of running changes as pace changes, not just how time grows with distance.
How they differ
The core difference is the shape of the curve. Riegel uses one exponent for everyone and every distance: the relationship between time and distance is assumed to be the same straight line on a log-log plot, whether you are moving from 5K to 10K or from 10K to the marathon. VDOT instead captures the curve of oxygen cost across paces — the fact that the metabolic penalty for going longer is not perfectly constant, and that the drop-off from short to long efforts bends rather than staying linear.
In practice that means the two agree closely between adjacent distances and diverge as the gap widens. Both also share a weakness: they extrapolate poorly at the extremes. Very short efforts (well under a few minutes) and very long ones (beyond a few hours) sit outside the data each model was built on, so predictions there are the least trustworthy.
A side-by-side comparison
Take a runner with a 40:00 10K and project up. The Riegel column uses the formula above with the standard 1.06 exponent. The VDOT column is an approximate, illustrative equivalent in the spirit of the Daniels & Gilbert model — we do not reproduce the full published tables here (see §4.0), so treat these as indicative rather than exact.
| Target distance | Riegel (exponent 1.06) | VDOT-style (approx., illustrative) |
|---|---|---|
| 10K (reference) | 40:00 | 40:00 |
| Half marathon | 1:28:16 | ~1:29 |
| Marathon | 3:04:01 | ~3:07 |
The pattern is typical: the half-marathon figures sit within about a minute of each other, while the marathon estimates drift a few minutes apart. That spread is not noise — it is the honest uncertainty of extrapolating a short race a long way. When two independent models land a few minutes apart on your marathon, the truth is somewhere in that band, and probably slower still if your long-run mileage is thin.
Which should you trust when?
For nearby distances and quick estimates — 5K to 10K, 10K to half — Riegel is hard to beat. It is transparent, needs nothing but arithmetic, and its single exponent is accurate enough over a short extrapolation that a physiology model rarely changes the answer meaningfully.
For a physiologically grounded second opinion, or when you want equivalent performances alongside the training paces that go with them, VDOT is the better tool. It is built from oxygen-cost data rather than a curve-fit, so it reflects how running economy and aerobic demand actually behave — with the caveat that it assumes trained-runner data and can flatter a recreational runner who has not done the work.
The most important caveat applies to both: neither model knows anything about your endurance or training volume. Each converts one performance into another as if your preparation for the target distance matched your preparation for the reference race. Step up in distance — especially toward the marathon — without the mileage, and both Riegel and VDOT will over-predict, handing you a time your legs are not yet trained to hold. The practical move is to run both, treat agreement as confidence and disagreement as a measure of uncertainty, then shade the estimate slower if your long runs do not back it up.
What neither model promises
Both Riegel and VDOT produce equivalent performances from published models, not training plans and not promises of race-day results. Their error grows outside the range each was validated on — roughly efforts shorter than three minutes or longer than about four hours — and grows again whenever you step up in distance without the endurance base the target demands. Read any single number as the centre of a range, cross-check it against the other model, and let your actual training decide how much of it to believe.
Frequently asked questions
Is Riegel or VDOT more accurate?
Neither is universally more accurate. Riegel is a simple power law fitted to race results and is reliable between nearby distances; VDOT is grounded in an oxygen-cost model calibrated to trained runners and captures how oxygen demand curves across paces. Between adjacent distances they usually agree closely; the gap widens as you extrapolate further.
Why do Riegel and VDOT give different marathon times?
Riegel applies one fixed exponent to everyone, while VDOT models a curving oxygen cost that steepens at longer efforts. Extrapolating a short race to the marathon exposes that difference, and both tend to over-predict if you have not built the endurance and mileage the marathon demands.
Can I use these equivalent times as a training plan?
No. Both are performance equivalences from published models, not training prescriptions. They assume you are trained for the target distance. Use them to set expectations and cross-check each other, not as a guarantee of race-day performance.
Related tools and reading
Race time predictor
Get a Riegel equivalent for any pair of distances, with an adjustable fatigue exponent.
ReadHow accurate is the Riegel race predictor?
The 1.06 exponent, where Riegel is reliable, and where it systematically over-predicts.
Sources
- Riegel, P.S. (1981), Athletic Records and Human Endurance, American Scientist, vol. 69, no. 3 — origin of the power-law endurance model and the 1.06 exponent. Checked 24 Aug 2026.
- Daniels, J. & Gilbert, J., Daniels' Running Formula (Human Kinetics) — VDOT equivalence built on the Daniels & Gilbert oxygen-cost model. Full VDOT tables are not reproduced here; consult the book for authoritative values. Checked 24 Aug 2026.