Over on Fittish, Jon Gugala ran through why fancy workout t-shirts tend to smell worse once they're loaded up with sweat (the bacteria love the material). The one thing he left out, though, seems like a crucial detail. The way the study was conducted literally just had seven people smelling bags of disgusting t-shirts. From the paper's methodology:

Odor assessment

Individual T-shirts in the plastic bags were presented to a panel of seven selected and screened human assessors. Assessors were selected by means of sensitivity to dilutions of n-butanol and wastewater, and by means of the triangle test (17). Each member of the panel was presented three flasks, two of which were the same but the third contained a different odor. The flask was shaken, the stopper was removed, after which the vapors were sniffed. The panelists had to correctly identify the different flask. The triangle test was repeated three times, with a minimum of two days in between each measurement. The room in which the tests were conducted, was free from extraneous odor stimuli e.g. caused by smoking, eating, soaps, perfume, etc. A representative team of odor assessors was chosen from the pool of assessors. The odor assessors were familiar with the olfactometric procedures and met the following conditions: (i) older than 16 years and willing to follow the instructions; (ii) no smoking, eating, drinking (except water) or using chewing gum or sweets 30 min before olfactometric measurement; (iii) free from colds, allergies or other infections; (iv) no interference by perfumes, deodorants, body lotions, cosmetics or personal body odor; (v) no communication during odor assessment. The samples were assessed by seven odor characteristics: hedonic value (between -4 and +4), intensity (scale 0 to 6), musty (scale 0 to 10), ammonia (scale 0 to 10), strongness (scale 0 to 10), sweatiness (scale 0 to 10), sourness (scale 0 to 10). A blank odor measurement, a clean cotton T-shirt with random number, was served to the odor panel together with the other samples.

Yeah, gross.