What does The Explanatory-Depth Razor conclude, in numbers, about “Reasons Against, the Premortem and Sealing the Ledger against Hindsight”?

Answer: Writing a reason against can shrink overconfidence, but it can improve only a thin layer of the Brier score; in this chapter's illustrative model, at most 0.00931. A premortem's coverage of failure causes saturates with the number of writers, five to seven are enough, and what matters most is that the writers bring different backgrounds. Rewriting even a tenth of the original confidences after the outcome creates paper progress far larger than any real improvement, so the ledger must be sealed.

Scope of the numbers: Parameters are illustrative, all data are synthetic, and every calculation is generated by the book's programs. Chapter 1 measures, in synthetic data, a drop of 1.283 points in self-rated understanding of devices against an interval for facts from -0.263 to 0.063 that includes zero.

中文版

Data table

The numbers below are printed verbatim by the chapter's Python program and can be reproduced by running it. standard-library Python onlyprinted values that match the answer: 0.00931

deltasRELBSBS gainshare of REL removed
0.01.80.009310.191810.000000.000
0.11.70.007680.190180.001630.175
0.21.60.006100.188600.003220.345
0.41.40.003180.185680.006130.658
0.81.00.000000.182500.009311.000
irreducible part E[q(1-q)] = 0.18250
Show the full program
# Chapter 5. p51: what writing one reason against buys in calibration.
# The judge reports logit(p) = s logit(q) with s = 1.8 (overconfident).
# Koriat, Lichtenstein & Fischhoff (1980) found that listing contradicting
# reasons improved the fit of confidence; we do not know by how much in a
# given team, so the intervention shrinks s by an unknown delta. Because
# p is monotone in q, the reliability term is E[(p - q)^2] exactly and
# BS = REL + E[q(1 - q)]. q is uniform on (0.05, 0.95); we integrate on a
# grid of 9,000 midpoints. Delta values are illustrative.
import math

M = 9000
QS = [0.05 + 0.9 * (i + 0.5) / M for i in range(M)]
UNC = math.fsum(q * (1.0 - q) for q in QS) / M


def logit(x):
    return math.log(x / (1.0 - x))


def rel(s):
    return math.fsum((1.0 / (1.0 + math.exp(-s * logit(q))) - q) ** 2
                     for q in QS) / M


base = rel(1.8)
print("delta   s     REL      BS       BS gain   share of REL removed")
for d in (0.0, 0.1, 0.2, 0.4, 0.8):
    r = rel(1.8 - d)
    print(" {:.1f}   {:.1f}   {:.5f}  {:.5f}  {:.5f}   {:6.3f}".format(
        d, 1.8 - d, r, r + UNC, base - r, (base - r) / base))
print("irreducible part E[q(1-q)] = {:.5f}".format(UNC))

Single source and citation

Fan, Y. (2026). Shattering the Illusion of Explanatory Depth, Epistemic Calibration and the Razor Separating True Knowledge from Assumed Familiarity (Version 1.0.0), Chapter 5: Reasons Against, the Premortem and Sealing the Ledger against Hindsight. 5D Library. DOI pending

DOI pending · monograph page goes live after publication · 5DL-B-396 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026396,
  author    = {Fan, Yafu},
  title     = {Shattering the Illusion of Explanatory Depth, Epistemic Calibration and the Razor Separating True Knowledge from Assumed Familiarity},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

Other chapter answers in this book