Contrary to recent claims, we argue that the orientation of the massive binary system Eta Carinae is such that the secondary star is closer to us at periastron passage, and it is on the far side during most of the time of the eccentric orbit. The binary orientation we dispute is based on problematic interpretations of recent observations. Among these observations are the radial velocity of the absorption component of He I P-Cyg lines, of the He II $lambda4686$ emission line, and of the Br$gamma$ line emitted by clumps close to the binary system. We also base our orientation on observations of asymmetric molecular clumps that were recently observed by ALMA around the binary system, and were claimed to compose a torus with a missing segment. The orientation has implications for the modeling of the binary interaction during the nineteenth century Great Eruption (GE) of Eta Carinae that occurred close to periastron passage. The orientation where the secondary is closer to us at periastron leads us to suggest that the mass-missing side of the molecular clumps is a result of accretion onto the secondary star during the periastron passage when the clumps were ejected, probably during the GE. The secondary star accreted a few solar masses during the GE and the energy from the accretion process consists the majority of the GE energy. This in turn strengthens the more general model according to which many intermediate-luminosity optical transients (ILOTs) are powered by accretion onto a secondary star.