Data & method

Distances

The half marathon is 21.0975 km and the marathon is 42.195 km. This site uses those figures everywhere, not 21.1 and 42.2. The difference sounds pedantic and is not: at four-hour marathon pace the rounding error is about fifteen seconds, which is precisely the margin somebody chasing a round number cares about.

Pace and time

Pace is total time divided by distance, and everything else is a unit conversion. Seconds per kilometre × 1.609344 gives seconds per mile; 3,600 ÷ seconds per kilometre gives km/h. There is no model and nothing to calibrate. All displayed times are rounded to the nearest second, and split tables are scaled so the splits sum exactly to the target — no accumulated rounding drift.

Race prediction — Riegel, 1981

T₂ = T₁ × (D₂ ÷ D₁)^1.06

Peter Riegel published this in American Scientist in 1981, fitting the exponent to a large body of race results. It works well between neighbouring distances for a runner who has trained for the longer one, and it flatters badly when used to predict a marathon from a 5K. The predictor page says so on the page rather than in a footnote, and shows a warning when the input distance is 10 km or shorter.

VO2max — Daniels & Gilbert, 1979

VO₂ = −4.60 + 0.182258·v + 0.000104·v²   (v in metres per minute)
%VO₂max = 0.8 + 0.1894393·e^{−0.012778·t} + 0.2989558·e^{−0.1932605·t}   (t in minutes)
VDOT = VO₂ ÷ %VO₂max

From Jack Daniels and Jimmy Gilbert, Oxygen Power (1979) — the equations underlying the VDOT tables in Daniels' Running Formula. What they estimate is running performance, which bundles true VO2max with running economy. That is more useful to a runner than either alone, but it is not a laboratory measurement and this site does not present it as one.

The Cooper test — Cooper, 1968

VO₂max ≈ (metres run in 12 minutes − 504.9) ÷ 44.73

Kenneth Cooper published this field test in JAMA in 1968. It is crude and it is free, and it needs nothing but a 400 m track and a watch.

Treadmill conversion

Speed to pace is arithmetic. The 1% incline convention comes from Jones and Doust, Journal of Sports Sciences (1996), who found that a 1% gradient equalised the oxygen cost of treadmill and outdoor running — at speeds of roughly 10.5 to 18 km/h. That speed qualifier is almost always dropped when the advice is repeated; this site keeps it, and the calculator tells you when it does not apply to you.

Fuelling and hydration

Carbohydrate targets (30–60 g/h for one to two and a half hours, up to 90 g/h beyond that with a glucose–fructose mix) and fluid ranges follow the published sports-nutrition consensus. They are given as ranges, and the site pushes you towards measuring your own sweat rate instead, because between-runner variation dwarfs any published average — especially in Indian heat. Sodium figures are per litre of fluid, not absolute daily amounts.

Training plans

The twelve plans are constructed from mainstream distance-running practice rather than any single published programme: roughly 80% easy running, one quality session a week for beginners and two for the faster goals, a cutback week every third or fourth week, weekly volume growing by no more than about 10%, and a real taper. They are illustrative schedules, not individual coaching, and the entry requirement is stated at the top of every one.

Race pages — what is deliberately absent

No dates, no entry fees and no invented organiser URLs. Race dates move every year and publishing a wrong one is not a small error — it is the kind that makes somebody miss a race they trained six months for. Each page names the month the event has traditionally fallen in, the organiser, and the month this site last checked the page against them, and then sends the reader to the organiser. No Event structured data is emitted either: startDate is a required property of Event, this site will not publish a date it cannot guarantee, and an event entity missing its date is an invitation for a search engine to fill the gap itself. The race pages declare what they actually are — authored course guides, marked up as Article. Only events with a long, continuous history are covered, which is why the list runs to about a dozen rather than the 150-plus a scraped calendar would claim.

City route pages — what is deliberately absent

Cities we could not describe confidently. Every route listed is a well-known public place to run, described from widely published knowledge of the city; distances are approximate, labelled as such, and unverified by survey — a park perimeter varies by a few hundred metres depending on which path you take. Ticketed entries, opening hours and campus access rules are noted where they apply. It would have been easy to generate forty more cities with plausible-sounding parks in them. It would also have sent somebody out at 05:30 to look for a path that does not exist.

Air quality and heat

The six air-quality bands, their ranges and their health statements are India's National Air Quality Index as published by the Central Pollution Control Board. The exercise thresholds attached to them on this site — reduce intensity above 200, stop training outdoors above 300 — are this site's own interpretation and are labelled as such, not official guidance. The heat table is a planning approximation; the underlying relationship between rising heat stress and slower marathon times is well established, but the per-band percentages are a rule of thumb, not a measured constant.

Corrections

Data corrections take priority over everything else here. If a formula is misapplied, a route has closed, or a race no longer runs, send the page URL and it gets fixed or removed on the next build. Every named source, with the year it was published and the month it was last checked, is listed on the sources page.