KMT-2019-BLG-2073: Fourth Free-Floating-Planet Candidate with θ_rm E < 10 rmμ as

We analyze the very short Einstein timescale ($t_\rm E \simeq 7\,{\rm hr}$) event KMT-2019-BLG-2073. Making use of the pronounced finite-source effects generated by the clump-giant source, we measure the Einstein radius $\theta_\rm E \simeq 4.8\,\rm \mu as$, and so infer a mass $M = 59\,M_\oplus (\pi_\rm{rel}/16 \,\rm \mu as)^{-1}$, where $\pi_\rm{rel}$ is the lens-source relative parallax. We find no significant evidence for a host of this planetary mass object, though one could be present at sufficiently wide separation. If so, it would be detectable after about 10 years. This is the fourth isolated microlens with a measured Einstein radius $\theta_\rm{E}<10\,\rm \mu as$, which we argue is a useful threshold for a "likely free-floating planet (FFP)" candidate. We outline a new approach to constructing a homogeneous sample of giant-star finite-source/point-lens (FSPL) events, within which the subsample of FFP candidates can be statistically analyzed. We illustrate this approach using 2019 KMTNet data and show that there is a large $\theta_\rm{E}$ gap between the two FFP candidates and the 11 other FSPL events. We argue that such sharp features are more identifiable in a sample selected on $\theta_\rm{E}$ compared to the traditional approach of identifying candidates based on short $t_\rm{E}$.