an equilibrium point
An equilibrium point is a place in the phase plane where the system, once there, stays forever. It is a point of perfect balance: a pendulum hanging straight down, a population at exactly its carrying capacity, two species at numbers where neither grows nor shrinks. Nothing is pushing the state in any direction, so it simply rests. In the wind-map picture, it is a point of dead calm where every arrow shrinks to zero.
Mathematically, for the system x' = f(x, y), y' = g(x, y), an equilibrium is a point (x0, y0) where BOTH rates vanish at once: f(x0, y0) = 0 and g(x0, y0) = 0. You find them by solving those two equations simultaneously. There the velocity field is exactly the zero vector, so a state starting at the equilibrium has zero velocity and never leaves — the constant function (x(t), y(t)) = (x0, y0) is a genuine solution, a trajectory that is a single point.
Equilibria are the skeleton of every phase portrait. You always locate them first, because the long-term behaviour of the whole system is organised around them: trajectories are pulled toward stable equilibria, repelled from unstable ones, and steered past saddles. The next question — what the nearby trajectories DO — is the classification problem (node, saddle, spiral, center), and it is decided by the system's behaviour in the immediate neighbourhood of the equilibrium.
For x' = x(1 - x - y), y' = y(0.75 - y - 0.5x), one equilibrium is (0, 0) (both species absent); others come from solving 1 - x - y = 0 and 0.75 - y - 0.5x = 0 together for a coexistence point. At each, both right-hand sides are zero.
Equilibria are exactly the simultaneous roots of f = 0 and g = 0.
Finding an equilibrium tells you a constant solution exists; it does NOT tell you whether nearby states are drawn in or pushed away. That separate question — its stability and type — needs the local analysis (eigenvalues of the linearization), not just the location.