Human skeletal muscle fiber type composition varies greatly along the muscle, so one biopsy may not accurately represent the whole muscle. Recommendations on the number of biopsies and fiber counts using immunohistochemistry and whether these findings can be extrapolated to other muscles are lacking. We assessed fiber type composition in the vastus lateralis and gastrocnemius medialis muscles of 40 individuals. Per muscle, we took four biopsy samples from one incision, collecting two samples each from a proximally and distally directed needle. Based on another dataset involving 10 vastus lateralis biopsies per participant (N=7), we calculated 95% limits of agreement for subsets of biopsies and fiber counts compared to the 10-biopsy average. Average absolute differences in type I fiber proportions between proximal and distal, and between within-needle samples were 6.9 and 4.5 percentage points in the vastus lateralis, and 5.5 and 4.4 percentage points in the gastrocnemius medialis, respectively. The 95% limits of agreement narrowed to ±10 percentage points when 200 fibers from at least three biopsies were analyzed, with minimal improvements with greater fiber counts. Type I fiber proportions in the vastus lateralis and gastrocnemius medialis showed a moderate positive association (r²=0.22; p=0.006; at least 200 fibers in each of three to four samples per muscle). In conclusion, three biopsies with a minimum of 200 counted fibers are required to estimate vastus lateralis fiber type composition within ±10 percentage points. Even when using these standards, researchers should be cautious when extrapolating muscle fiber type proportions from one muscle to another.