PROOF - OPEN YOUR EYES READ HERE (this only took a couple of minutes of searching maybe learn to use google) ...
http://www.triathlon.org/docs/competition-rules-20080601-vf.pdf
on page 41 (for triathlons and DUATHLONS):
"Demanding and technical courses should always be preferred. A 5% error margin will be accepted for the cycle and run course provided the margin has been approved by the Technical Delegate"
Does this mean 5% error for total distance of duathlon? because if it did the bike leg could be made correct or long and then make the run leg say 8-10% short and total duathlon distance still within 5% error. Even if its 5% less per individual leg that is just RIDICULOUSLY INACCURATE. And the demanding and technical courses part is crap the triathlon world champs on the gold coast were pancake flat ...
IAAF rules for road races (to access click competitions>technical area>IAAF competition rules 2009):
http://www.iaaf.org/mm/Document/Competitions/TechnicalArea/04/95/59/20090303014358_httppostedfile_CompetitionRules2009_printed_8986.pdf
on page 214:
"The length of the course shall not be less than the official distance for the event. In competitions under Rules 1.1(a), (b), (c) and (f), the uncertainty in the measurement shall not exceed 0.1% (i.e. 42m for the Marathon) and the length of the course should have been certified in advance by an IAAF approved course measurer."
"In events on roads the course shall be measured
along the shortest possible route that an athlete could follow within the section of the road permitted for use in the race."
For world record purposes page 224:
"The overall decrease in elevation between the start and finish shall not exceed 1:1000, i.e. 1m per km."
Many other rules - extremely strict for track (and timing regulations) such as:
"The lateral inclination of tracks should not exceed 1:100 and the overall inclination in the running direction shall not exceed 1:1000 downwards ...
The length of a standard running track shall be 400m ...
The measurement shall be taken 30cm outward from the kerb or, where no kerb exists, 20cm from the line marking the inside of the
track"
LONG SUMMARY:
Tris and duathlons can be 5% short, and the run leg or any other leg could possibly be even shorter. Also start/finish of run leg can often be at different points allowing for net elevation loss, therefore absolute times in these races do not have any meaning but times relative to fellow competitors do. Also questions raised by others such as placement of timing mats, actual course on race day (i notice many of these races have out and back sections and the turnaround on these could be placed in the wrong spot on race day) etc ...
Running courses much more accurate (+/- 0.1%) and no net elevation loss for record purposes. This is probably why in running we can compare current runners to the past but in tri this is not really possible. I wouldn't mind seeing science of sport doing a post on this topic.
SHORT SUMMARY: triathlon sucks