Found this while googling for swim drag... the tri bodies might find it useful.
Facial Hydrodynamics and Front Crawl
D Drexler et al, UC Berkeley Institute of Sports Science
Athletes have long known the importance of correct body position and head orientation when performing the front crawl. With advances in computer modeling and materials science, it has been possible to develop swimsuits with remarkably low drag coefficients (e.g. Speedo 'Sharkskin') and goggles whose shape produces minimal disturbance of the water flow around an athlete's face. We have applied these computer modeling techniques to a neglected are of the athlete's body: facial hair.
It is generally assumed that a smooth skin surface is optimal for a swimming athlete, but as the 'sharkskin' suit and, on a different Reynolds scale the dimples in golf balls show, this is not necessarily true: better is to capture a layer of the fluid medium against the surface and allow this to move with the athlete (or ball); the drag of water-on-water or air-on-air is significantly lower than surface-in-air drag.
The lower face of an athlete is dominated by the transverse topological features of the mouth, then the inline features of the jawline. We investigate:
1) The possibility that facial hair may mediate water flow around the 'braking' effect of the mouth.
2) The possibility that fluid turbulence occurring after the jawline may be mediated.
3) The relative drag coefficients of facial hair styles: is there a 'magic length' which minimizes drag?
Three-dimensional computer models of competition-successful swimmers were constructed and, using standard forensic pore density models, populated with a range of stubble, beards, sideboards and mustaches. These models were put through simulated swim tests and the drag coefficients and overall times predicted are presented in the full paper here:
http://tinyurl.com/2pgxprIn summary:
1) "Bar" mustaches had no noticeable effect on drag. They brought the braking surface of the mouth forward, but without statistically-significant drag or time effects.
2) Full beards almost invariably increased drag by a significant amount. Only beards below 6mm in average hair length did not. Two exceptions arise: very dense beards are more slippery in the water; and vandyked beards reduced drag over conventional beards due to a vortex-spoiler effect at the tip of the stroke-side beard.
3) Stubble of an average 1.2mm was marginally more slippery than smooth skin. Shorter stubble failed to trap the water; longer stubble encountered the "beard drag" effect of (2).
4) Goatees where pointed and dense significantly improved drag where the facial model had a deeply cleft chin: the water flow was normalized around this flow-breaking structure.
4) Long, drooping mustaches of a "Mexican" style acted to guide water flow around the mouth, reducing drag over the lower face by as much as an order of magnitude.
5) "Dali" mustaches collapsed under testing.
Recommendations:
From this exercise we conclude that an athlete with optimal head positioning would gain significantly from the following facial hair styles, expecting a performance improvement of 3%-5% over a clean-shaven athlete of similar ability:
* A dense, trimmed, vandyked beard.
* A full 'Mexican' mustache allowed to descend below the lips.
* Not shaving on the morning of competition.
