Abstract We calculate the microlensing event rate and typical time-scales for the free-floating planet (FFP) population that is predicted by the core accretion theory of planet formation. The event rate is found to be ∼1.8 × 10−3 of that for the stellar population. While the stellar microlensing event time-scale peaks at around 20 d, the median time-scale for FFP events (∼0.1 d) is much shorter. Our values for the event rate and the median time-scale are significantly smaller than those required to explain the Sumi et al. result, by factors of ∼13 and ∼16, respectively. The inclusion of planets at wide separations does not change the results significantly. This discrepancy may be too significant for standard versions of both the core accretion theory and the gravitational instability model to explain satisfactorily. Therefore, either a modification to the planet formation theory is required or other explanations to the excess of short-time-scale microlensing events are needed. Our predictions can be tested by ongoing microlensing experiment such as Korean Microlensing Telescope Network, and by future satellite missions such as WFIRST and Euclid.