We study the dynamics of Abell 370 ( A370 ) , a highly massive Hubble Frontier Fields galaxy cluster , using self-consistent three-dimensional N -body/hydrodynamical simulations . Our simulations are constrained by X-ray , optical spectroscopic and gravitational lensing , and Sunyaev–Zel ’ dovich ( SZ ) effect observations . Analyzing archival Chandra observations of A370 and comparing the X-ray morphology to the latest gravitational lensing mass reconstruction , we find offsets of \sim 30 kpc and \sim 100 kpc between the two X-ray surface brightness peaks and their nearest mass surface density peaks , suggesting that it is a merging system , in agreement with previous studies . Based on our dedicated binary cluster merger simulations , we find that initial conditions of the two progenitors with virial masses of 1.7 \times 10 ^ { 15 } M _ { \odot } and 1.6 \times 10 ^ { 15 } M _ { \odot } , an infall velocity of 3500 km s ^ { -1 } , and an impact parameter of 100 kpc can explain the positions and the offsets between the peaks of the X-ray emission and mass surface density , the amplitude of the integrated SZ signal , and the observed relative line-of-sight velocity . Moreover , our best model reproduces the observed velocity dispersion of cluster member galaxies , which supports the large total mass of A370 derived from weak lensing . Our simulations strongly suggest that A370 is a post major merger after the second core passage in the infalling phase , just before the third core passage . In this phase , the gas has not settled down in the gravitational potential well of the cluster , which explains why A370 does not follow closely the galaxy cluster scaling relations .