I Got My Body Fat Tested Every Way- Which Is Most Accurate?
In a Nutshell
A man with visible muscle definition tested eight body composition methods and found underwater weighing gave the lowest readings (3.2%), Bod Pod the next lowest (4.4%), and skinfold measurements varied widely by protocol (4.1–8.3%). BIA devices produced absurdly high results (20–33%), while DEXA scans had previously swung between 6% and 14%. His final estimate settled on 7–8% body fat, with high bone density likely skewing density-based methods downward; he recommends tracking waist circumference for progress rather than relying on any single absolute value.
These notes were generated by AI and may contain inaccuracies.
The testing was conducted at Cornerstone University in Michigan after previous DEXA scans produced inconsistent results, with one showing 14% body fat and another showing 6% body fat after a fat loss phase. The subject weighed 217 lbs on the day of testing and 222 lbs during the DEXA scan. The university has a well-equipped kinesiology lab with Dr. James Sackett, PhD leading the program. The goal was to compare all body composition testing methods except MRI.
The Bod Pod operates on a two-compartment model using body density calculations. It measures body volume by having the subject sit inside the chamber, then calculates density by combining volume with mass from weighing. The result was 4.4% body fat. The subject noted this seemed optimistic compared to the recent 6% DEXA result, though more accurate than the previous 14% reading.
The subject performed multiple underwater weighing trials, exhaling maximally before submerging and remaining still for 5 seconds. The platform scale measures weight underwater wirelessly. Results were 3.9%, 2.9%, and 3.4% body fat across trials, averaging approximately 3.2%. This method is considered the gold standard in research literature, though it showed the lowest body fat estimates.
Nine-site, seven-site, and three-site skinfold protocols were performed. The nine-site measurement yielded 5.9% body fat, the seven-site (most validated) gave 8.3%, and the three-site produced 4.13%. The seven-site method has the most research validation. Site selection affects results, with the seven-site excluding bicep and calf measurements, which were the subject's leanest areas.
Both handheld and standing scale bioelectrical impedance analysis devices were tested. The handheld BIA measured 20.5% body fat, while the standing scale measured 32.8% body fat. These results were considered highly inaccurate given the subject's visible muscle definition and vascularity. BIA devices often have "athlete mode" versus "regular mode" settings, which the subject noted as problematic since humans exist on a continuum rather than discrete categories.
An AI-based body composition system called Spren was tested but failed to produce usable results due to positioning issues with the camera setup.
Waist and hip circumference measurements were taken, yielding a 34-inch waist and 43-inch hips for a waist-to-hip ratio of 0.79. This method classified the subject as low health risk with a pear body shape. Using the Bergman et al. 2011 relative fat mass equation, this suggested 10-14% body fat.
The subject had previously received conflicting DEXA results: one scan showed 14% body fat with minimal fat loss despite significant weight reduction, while another brand showed 6% body fat at 222 lbs. This discrepancy prompted testing with multiple methods to establish ground truth rather than relying on a single measurement.
From least to most valid in the literature: waist-to-hip ratio (10-14% estimate), standing BIA (32.8%), handheld BIA (20.4%), three-site skinfolds (4.13%), nine-site skinfolds (5.9%), seven-site skinfolds (8.3%), Bod Pod (4.4%), DEXA (6%), and underwater weighing (3.2%).
The subject concluded their actual body fat was likely 7-8%. High bone density (five standard deviations above average) likely skewed density-based methods like Bod Pod and underwater weighing toward lower estimates. The seven-site skinfold result of 8.3% was considered most realistic given the method's research validation.
Waist circumference tracking every 3 weeks is recommended for monitoring fat loss progress or setting bulk limits, though not for absolute body fat estimation. BIA is unreliable for non-average demographics and functions essentially as a random number generator. Skinfolds should be performed by certified anthropometrists and used for tracking changes rather than absolute values. Bod Pod provides estimates within approximately 3% of actual body fat. DEXA has 1-3% error rates with classified algorithms that may produce erroneous results. Underwater weighing, while the gold standard, has 1-2% error from lung volume and bone density assumptions.
For skinfolds, professional administration is essential for validity. Bod Pod offers convenience without water submersion or X-ray exposure. DEXA should not be used too frequently as it measures noise when done every 3 weeks; consistency in hydration, feeding, and glycogen status is crucial for reliable lean body mass tracking. Underwater weighing accuracy improves with maximal exhalation, making the subject appear heavier and thus leaner to the machine.
The subject plans to diet from March to July to achieve greater leanness, potentially conducting additional measurements. Cornerstone University's kinesiology program was highly recommended for its exceptional body composition laboratory facilities.
Keep Renaissance Periodization in your library
Save the videos and channels worth coming back to, and find them again in one place.
