Abstract: The purpose of this research is to analyse delivery prediction deviation compared to the realdelivery date. Research design was descriptive explorative with retrospective approach. The populationwere all of the gravides who delivery at BPM and clinic at Malang Raya. Number of sample were279 with random sampling technique. The result was the < 20-years-old women had deviation -5,3148+14,227, male baby -3,716666667 + 9,01463 had deviation, and female baby -2,875 + 12,9317 haddeviation. Babies who weigh < 2500 g had deviation of -12 + 14,992. G IV mothers had -9 + 8,14248deviation. In conclusion, gravide mothers can deliver the baby 2 weeks earlier or later.