Dozens of planets are now discovered with large orbital obliquity, and have become the proof for the dynamical evolution of planetary orbits. In the current samples, there is an apparent clustering of planets around 90°, and also an absence of planets around 180° although the latter is expected by some theories. Statistical extrapolation using Hierarchical Bayesian Analysis have recently refuted the significant clustering around 90° and suggested that the distribution may actually be broader. In this work, the symmetric TESS transit light curve of KELT-19Ab is analysed using gravity darkening to measure its true obliquity. Its large sky projected obliquity $\lambda = -179.7^{\circ +3.7^\circ }_{\, \, -3.8^\circ }$ makes KELT-19Ab the only currently known planet with obliquity potentially close to 180°. We apply spectroscopic constraints on vsini and λ as well as theoretical constraints on the limb-darkening coefficients to find that the KELT-19Ab's obliquity is $\psi = 155^{\circ +17^\circ }_{\, \, -21^\circ }$, in favour of a flipped orbit. The result is consistent with the statistically inferred uniformity of obliquity distribution, and also highlights the applicability of the gravity darkening technique to symmetric light curves.