Why is the scientific success argument sometimes called the ultimate argument for realism?

The "ultimate argument for realism" is so named because it posits that the success of scientific theories would be a "miracle" or a "coincidence of cosmic proportions" if those theories were not at least approximately true.1

Proponents of this argument, including J.J.C. Smart, Hilary Putnam, and Richard Boyd, observed that science consistently produces systematic successes across broad domains—such as accurately computing reaction energies or manufacturing organisms with precise traits. They argued that the most reasonable explanation for this sustained effectiveness is that the theories correctly describe the world, including its unobservable entities.1

By framing the truth of scientific theories as the only explanation for their success, realists attempt to move beyond mere instrumentalism, which views theories only as tools for prediction. However, this argument remains a subject of intense philosophical debate; critics like Larry Laudan have challenged it by pointing to the history of science, noting that many past theories were once considered successful yet are now known to be false, a position often called the "pessimistic induction".213