In the answer to a question on the validity or otherwise of a world record, the answer quotes the rule book which says:

"The start and finish points of a course, measured along a theoretical straight line between them, shall not be further apart than 50% of the race distance." [i.e., 13.1 miles (21.1 km) for the ~26.2 mile (~42.2 km) marathon distance]

Why should this be so? I can understand stipulations about ascent or descent between the start and end, but why should the endpoints need to be close together? (It means that the original 26.2 mile course, from Windsor to White City in west London, wouldn't be legal.)

  • Wasn't this a problem with the Rio olympic marathon?
    – Bernhard
    May 30, 2012 at 17:23
  • 1
    Just to nitpick, the true actual original marathon wasn't from Windsor to White City, but from Marathon to Athens, hence the name :-) Sep 18, 2012 at 19:14
  • 4
    If you're going to nitpick, be accurate; while the original might have been in Greece, that isn't reasoning behind the 26.2 miles that we run today: en.wikipedia.org/wiki/Marathon#Distance
    – Unsliced
    Sep 18, 2012 at 19:28
  • Responding very late to @Bernhard's comment - the Rio Olympic marathon had both start and finish at the Sambodromo, so it was record-legal, although there may have been earlier, proposed courses which were not. The "classic" Marathon-to-Athens course used in 2004 was not record-legal under this rule.
    – pjmorse
    Jan 17, 2017 at 15:15

2 Answers 2


Several other traditional marathon courses - notably the Polytechnic Marathon (Windsor to Chiswick) which hosted a few WRs in its time - are also not record-valid under this rule, which is relatively recent.

Ken Young of the Association of Road Race Statisticians has done a lot of work on the effects of wind direction on race times, and has published an overview of his work on the topic.

Some 20 years ago, I worked out a simple model that demonstrated runners could receive a significant tail wind benefit on a course whose start and finish lay more than 30% of the race distance apart.

Young asserts, and I believe he's right, that his work is the basis of the USA's current 30% separation rule. The IAAF rule of 50% separation is more lax, and Young believes still allows for what he considers "unacceptable" benefit. (Young also believes records set by women in races which include men should not be accepted either, but that's another story entirely.)

  • While I have no citation for this, for obvious reasons, the rumor is that the IAAF rule allows 50% separation instead of 30% separation because of the lobbying of representatives from several large Japanese marathons which would be record-ineligible under the 30% rule but are OK under the 50% rule. This supposedly also accounts for the IAAF's insistence on women's records coming in separate, not mixed, races.
    – pjmorse
    May 18, 2016 at 12:34

Aside from elevation, prevailing wind direction might dictate close proximity of start and finish to ensure a 'fair' race.

  • 3
    Not that I necessarily doubt you, but do you have a citation for this?
    – corsiKa
    Feb 20, 2012 at 23:09
  • 2
    The basic rules are here: IAAF COMPETITION RULES 2012-2013: rule 260.28 (approximate page 244). I cannot find documents that explain or argue for the rules... Feb 21, 2012 at 12:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.