Purpose Various time-lapse studies have postulated embryo selection criteria based on early morphokinetic markers. However, late paternal effects are mostly not visible before embryonic genome activation. The primary objective of this retrospective study was to investigate whether those early morphokinetic algorithms investigated by time-lapse imaging are reliable enough to allow for the accurate selection of those embryos that develop into blastocysts, while of course taking into account the correlation with the type of injected spermatozoa. Methods During a period of 18 months, a total of 461 MII oocytes from 43 couples with severe male factor infertility and previous Bexternal^IVF failures after cleavage-stage embryo transfer (ET) were fertilized by intracytoplasmic morphologically selected sperm injection (IMSI). Thereof, 373 embryos were monitored in a time-lapse incubator until ET on day 5. Blastocyst outcome in combination with three previously postulated MKc (cc2: t3-t2, 5-12 h; t3, 35-40 h; t5, 48-56 h) and the morphology of the selected sperm were analyzed. Results A significant increase in the rate of blastocysts (54.0 vs. 36.3 %; P<0.01) and top blastocysts (25.3 vs. 10.8 %; P<0.001) was observed in the group of those meeting all three morphokinetic criteria (MKc3). However, MKc3 were only met in 23.3 % of all embryos. Moreover, TBR was influenced by the type of injected spermatozoa. In both groups, TBR decreased dramatically (MKc3, 35.0 vs. 17.0 %; MKc<3, 14.2 vs. 8.4 %) when class II/III sperm instead of class I were injected. Conclusion Early morphokinetic parameters might give some predictive information but fail to serve as a feasible selective tool for the prediction of blastocyst development given the influence of the type of spermatozoa injected.