The discussion centers around P hacking, the manipulation of data to achieve statistically significant results. Users comment on the perils of relying on null hypothesis significance testing, equating it to gambling—where results can be skewed by repeated testing. There's a strong sentiment that the focus should be on improving the transparency of research practices and reassessing the incentives that guide tenure, promotion, and grant decisions in academia. Concerns are raised about the common occurrence of poorly formulated ("dogshit") null hypotheses, which readily succumb to manipulation when sufficient data is gathered. The consensus points towards a need for a shift in research attitudes, advocating for more stringent falsification practices rather than confirmationalist approaches, where researchers tend to defend their favorite hypotheses instead of rigorously testing them. This highlights the need for a more robust scientific methodology that values reproducibility and sound hypothesis testing.