The design practice of micropiles in Ontario soils under the ultimate limit state was improved through both statistical and reliability analyses of a database of 40 micropile load tests. Micropile design is extremely dependent on engineering experience and judgement due to the lack of an accurate estimation of the bond strength. The FHWA manual of micropiles only provides wide ranges of bond strength in different ground conditions. Micropile load tests were conducted by Keller Foundations Ltd and collected for this study. From a statistical analysis, Fuller and Hoy’s method was selected as the best method to estimate the failure load from non-failed tests. Adjusted parameters were given to predict the bond strength of micropiles. A method was proposed to estimate the contributions from the cased length and the tip to the total resistance. In the end, a reliability analysis was conducted and the resistance factors were recalibrated.