What does The Explanatory-Depth Razor conclude, in numbers, about “Reading the Classics: the Xunzi on Dispelling Obsession, Chapter Seventy-One of the Laozi and Descartes's Four Rules”?
Answer: The true overall repeat rate is 0.3250. The reader of one corner has an error of 0.2494 with a sample of 20, and with a sample of 1,280 the error has fallen only to 0.2244; it barely moves. The stratified reader's error is 0.0984 at 20 and falls to 0.0124 at 1,280. The last line gives the bias floor for the reader of one corner: 0.2250, exactly the gap between the loyal customers' repeat rate and the overall rate.
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.
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: 0.0124, 0.0984, 0.2244, 0.2250, 0.2494, 0.3250
| true overall repeat rate: 0.3250 | ||
| sample n | one-corner RMSE | stratified RMSE |
| 20 | 0.2494 | 0.0984 |
| 80 | 0.2310 | 0.0499 |
| 320 | 0.2261 | 0.0253 |
| 1,280 | 0.2244 | 0.0124 |
| bias floor of the one-corner reader: 0.2250 |
Show the full program
# Chapter 7. p71: "blinded by one corner" as a bias that more data keeps.
# A buying service sells to five buyer groups with different repeat
# rates. A decision maker who only reads messages from the group she
# already serves samples one corner; a stratified reader samples every
# group in proportion. We compare the root mean square error of the
# estimated overall repeat rate over 1,000 repetitions as the sample
# grows. One-corner error stops falling at its bias. Illustrative rates.
import math
import random
rng = random.Random(39671)
GROUPS = [("loyal", 0.25, 0.55), ("gift", 0.20, 0.30), ("trial", 0.25, 0.15),
("bulk", 0.15, 0.40), ("abroad", 0.15, 0.20)]
TRUE = math.fsum(w * r for _, w, r in GROUPS)
REPS = 1000
def one_corner(n):
r = GROUPS[0][2]
return len([1 for _ in range(n) if rng.random() < r]) / n
def stratified(n):
parts = []
for _, w, r in GROUPS:
k = max(1, round(w * n))
parts.append(w * len([1 for _ in range(k) if rng.random() < r]) / k)
return math.fsum(parts)
def rmse(fn, n):
return math.sqrt(math.fsum((fn(n) - TRUE) ** 2
for _ in range(REPS)) / REPS)
print("true overall repeat rate: {:.4f}".format(TRUE))
print("sample n one-corner RMSE stratified RMSE")
for n in (20, 80, 320, 1280):
print(" {:5,} {:.4f} {:.4f}".format(
n, rmse(one_corner, n), rmse(stratified, n)))
print("bias floor of the one-corner reader: {:.4f}".format(
GROUPS[0][2] - TRUE))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 7: Reading the Classics: the Xunzi on Dispelling Obsession, Chapter Seventy-One of the Laozi and Descartes's Four Rules. 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
- Chapter 1 · The Boundary of the Illusion of Explanatory Depth: Strongest for Mechanisms, Weakest for Facts
- Chapter 2 · Calibration, Not Confidence: the Brier Score, Murphy's Decomposition and the Statistical Artefact in the Dunning-Kruger Chart
- Chapter 3 · The Four-Step Razor: Causal Chain, Order of Magnitude, Failure Boundary and Counterexample
- Chapter 4 · Scoring the Assumption List and the Value of Information: Which Assumption to Test First
- Chapter 5 · Reasons Against, the Premortem and Sealing the Ledger against Hindsight
- Chapter 6 · Turning "Fatal" and "Stripping" into Intervals: a Band of Decision Loss, a Small Effect and the Conditions for Refutation
- Chapter 8 · Field Tools and Checker Specifications: the False-Knowledge Register, the Four-Step Razor Scorer and a Pre-Registered Test