The 26-Sigma Mirage

This podcast episode was generated using Notebook LM as a conversational summary of the final pre-registered report. While the transcript has been substantially corrected for accuracy, listeners should refer to the official Zenodo document (DOI: 10.5281/zenodo.20258203) for the complete, authoritative findings.The Hook: A Decade-Old Mystery Meets the BlockchainFor over a decade, independent researcher Mario Buildreps has curated a massive database of 1,159 ancient pyramids, temples, and megalithic sites. His claim is the kind that keeps orthodox geologists awake at night: these structures weren’t built to face our current North Pole. Instead, they supposedly align with a series of “paleopoles”, former positions of Earth’s rotational axis, sliding down the 47°W meridian.In 2026, independent researcher Salah-Eddin Gherbi launched a forensic audit to determine if this was a revolutionary discovery or a statistical hallucination. To ensure the highest level of scientific rigor, Gherbi utilized “open science” protocols, pre-registering his full study on Zenodo and locking his analysis plan into the Bitcoin blockchain. This ensured that no one, including Gherbi, could “massage” the data once the numbers began to talk.The result is a startling mixed verdict. While Gherbi successfully dismantled a massive statistical artifact, he also confirmed a core of data so robust it survived every “hostile” test he could throw at it.The Trap of the “26-Sigma” ResultWhen Gherbi first ran the pre-registered test on the Buildreps database, he received a “red alert” result. The primary test statistic showed a significance of p = 0.0001—a 26-sigma effect that would normally signal a discovery of Nobel proportions.However, an investigative skeptic knows that a 26-sigma result in archaeological data should be approached with suspicion, not celebration. Gherbi flagged this as a “hemisphere-selection” artifact. The database was pre-filtered to include only structures that pointed toward the northern hemisphere along the 47°W meridian.To explain the error, think of it this way: if you only count people currently standing in a kitchen, you shouldn’t be surprised when your data “proves” that 100% of the population prefers kitchens. When Gherbi ran an “unconditional null”, comparing the real sites to random points that could fall anywhere on the globe, the real data looked impossibly perfect because it stayed in the “kitchen” (the north) while the random data wandered into the “living room” (the south).When he switched to a “conditional null”, limiting the comparison only to other northern orientations, the massive 26-sigma signal evaporated.The Survivors: Why Poles II and III Refuse to VanishDespite resolving the 26-sigma artefact, Gherbi found something that genuinely shocked him. Two specific locations, Pole II (76.0°N) and Pole III (72.2°N), refused to vanish.Gherbi subjected these sites to a “Block-Conditional” test—a statistically hostile gauntlet that shuffles orientations only within regional geographic blocks. This test was designed to rule out “regional traditions” or “The Maya doing Maya things.” If the clustering was just a cultural habit, this test would have killed the signal.Instead, Poles II and III survived with Šidák-corrected p-values of 0.0015 and 0.0005. The data revealed that roughly 24% of the structures (about 234 sites) point to these two spots far more than chance would allow. Most notably, there were 50 more structures clustered at these poles than the most stringent models predicted. This isn’t background noise; it’s a real feature of the data that survived a forensic takedown.The Meridian Shift: Hunting for the Real AttractorWhile the original theory tethers these poles to the 47°W meridian, Gherbi’s “longitude scan” suggests the data has a mind of its own. When he looked for the “natural attractor”, the line where the structures cluster most tightly, the 47°W meridian was only Rank 10 out of 72 scanned longitudes.The true peak of the data lies further east. Here is how the meridians actually rank for clustering:* 20°W (Rank 1 - The Natural Attractor)* 25°W (Rank 2)* 30°W (Rank 3) ...* 45°W (Rank 8 - The Pre-registered Band) ...…* 47°W (Rank 10 - Buildreps’ Target)This discrepancy shows that, while the clustering is undeniable, the theory’s proposed “path” of the poles is slightly off. The real signal is pulling toward the Atlantic between West Africa and Brazil.The Silent Current Pole: A Surprising Lack of SignalThe most counter-intuitive finding of Gherbi’s audit involved “Pole I”, our current geographic North Pole (9