Experimental product distribution of Fischer-Tropsch synthesis frequently presents notable deviations from the typical double-alpha Anderson-Schulz-Flory pattern: bump or dip around the breaking carbon number, positive or negative deviation for heavy hydrocarbons. These irregularities were studied experimentally in a fixed-bed reactor over an industrial Fe/Mn catalyst, and theoretically by a product separation model based on Aspen Plus software. First, it was found that the unsteady state of reaction condition or improper gas chromatograph procedure could lead to deviation for heavy hydrocarbon distribution. Second, the bump near the breaking carbon number could be attributed to the accumulation of water in hot trap, which leads to an inaccurate measurement of the wax amount. This irregularity can be eliminated by selecting either a higher temperature or a lower pressure of the hot trap. Third, vaporization or flash loss of the oil sample during product collection could result in dip in light hydrocarbon distribution. High syngas conversion levels should be avoided for accurate data acquirement.