A sharp rebuttal to Lisa Feldman Barrett's NYT op-ed 'Psychology Is Not In Crisis' (which framed failed replications as benign context-dependence, like Newton→quantum). Scott's rebuttals: it ignores publication bias (many replication failures are just false positives); experimenter effects (the 'remarkable coincidence' that experimenters find what they believe); and the 'it is more likely' phrase ignores base rates (Ioannidis — a positive study on a low-prior hypothesis is more likely a false positive than a real-but-context-dependent effect; the dart-at-metabolic-pathways / Uri-Geller-Randi analogies). And even granting genuine context-dependence, it's STILL a crisis: if priming works with one wallpaper color but not another and we can't predict which, we have 'the practice of doing physics experiments' but not 'a collection of results that describe reality' (the physicist who says 'apples sometimes fall up, sometimes down, can't predict which, but no crisis').

Quality 74 · Strong. High-Strong: a sharp, cited replication-crisis rebuttal with a portable central reframe ('if you can't predict, you're still in crisis') plus the memorable physics analogy and the base-rate/Ioannidis point — alongside Chocolate-Study (74)/Streetlight (73). 74.

Claude’s paradigm shift 56 · Moderate. A strong, original rebuttal + the predict-or-crisis framing. B56.

Real-world impact 2 · Minor. Influential within the replication-crisis discourse; no material reach. RWI2.

If You Can’t Make Predictions, You’re Still In A Crisis