What does Short-Video Traffic Pools conclude, in numbers, about “Closing the Loop: the Weekly Sheet, the Promotion Calculator and a Pre-registered Refutation Plan”?
Answer: Chapter 8 gives the weekly procedure, the Traffic-Pool Promotion Probability Calculator and a pre-registered refutation plan: the product model predicts a top-third rate of 0.3449 at the third level against 0.1451 under an unchanged ratio, testable with 72 videos per third over 18 weeks. Marked boxes separate published evidence, runnable Python calculations (18 programs), classical sources and the author's hypotheses.
Scope of the numbers: Chapter 3 shows that a hard threshold is a ramp in the data, whose width narrows from 0.124 to 0.023 as the batch grows from 100 to 3,000 viewers; the illustrative base video breaks out with probability 0.03585, with the bottleneck at the fourth gate. Marked boxes separate published evidence, runnable Python calculations (18 programs), classical sources and the author's hypotheses.
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: 0.1451, 0.3449, 18, 72
| gate | bottom | middle | top | top/middle |
|---|---|---|---|---|
| 1 | 0.2106 | 0.5000 | 0.7894 | 1.58 |
| 2 | 0.0634 | 0.2392 | 0.5660 | 2.37 |
| 3 | 0.0162 | 0.0919 | 0.3449 | 3.75 |
| 4 | 0.0031 | 0.0265 | 0.1720 | 6.50 |
| gate 3, top third: flat-ratio null 0.1451 vs chain 0.3449 | ||||
| videos per third needed: 72, weeks at 4 per third: 18 |
Show the full program
# Chapter 8. p82: a pre-registered test of the multiplicative chain.
# Prediction of the gate model (p11, illustrative; cohort q ~ N(0, 1)):
# if promotion is a product of gates, the ratio of reach rates between
# the top and the middle third of videos must grow with every gate; if
# promotion were one draw after gate 1, it would stay flat after gate 1.
# The creator cannot see q, only a gate-1 score q + e with e ~ N(0, 1)
# (reliability 0.5, illustrative), and splits videos into thirds by
# that score. We compute the reach rates by third, the ratio by gate,
# and the videos per third needed to tell the chain from the flat
# ratio at gate 3 (two proportions, alpha 0.05, power 0.80). At 12
# videos a week, 4 fall into each third.
import math
T = [0.0, 0.5, 1.0, 1.5]
S = 0.3
ZA, ZB = 1.959964, 0.841621
CUT = 0.430727 * math.sqrt(2.0)
def reach(q, k):
return math.prod(1.0 / (1.0 + math.exp(-(q - t) / S)) for t in T[:k])
def dens(x, mu, sd):
z = (x - mu) / sd
return math.exp(-z * z / 2.0) / (sd * math.sqrt(2.0 * math.pi))
HQ = 0.01
QG = [-6.0 + HQ * i for i in range(1201)]
R = {k: [reach(q, k) for q in QG] for k in range(1, 5)}
HO = 0.02
OG = [-8.0 + HO * i for i in range(801)]
COND = {}
for o in OG:
w = [HQ * dens(q, o / 2.0, math.sqrt(0.5)) for q in QG]
COND[o] = [math.fsum(a * b for a, b in zip(w, R[k]))
for k in range(1, 5)]
def third(k, lo, hi):
pts = [o for o in OG if lo <= o < hi]
wo = [HO * dens(o, 0.0, math.sqrt(2.0)) for o in pts]
num = math.fsum(w * COND[o][k - 1] for w, o in zip(wo, pts))
return num / math.fsum(wo)
print("gate bottom middle top top/middle")
rates = {}
for k in range(1, 5):
b, m, t = third(k, -99, -CUT), third(k, -CUT, CUT), third(k, CUT, 99)
rates[k] = (m, t)
print("%4d %6.4f %7.4f %6.4f %9.2f" % (k, b, m, t, t / m))
(m1, t1), (m3, t3) = rates[1], rates[3]
null = m3 * t1 / m1
pb = (null + t3) / 2.0
top = ZA * math.sqrt(2 * pb * (1 - pb)) \
+ ZB * math.sqrt(null * (1 - null) + t3 * (1 - t3))
n = math.ceil(top * top / (t3 - null) ** 2)
print("gate 3, top third: flat-ratio null %.4f vs chain %.4f" % (null, t3))
print("videos per third needed: %d, weeks at 4 per third: %d"
% (n, math.ceil(n / 4.0)))Single source and citation
Fan, Y. (2026). Cold-Start Dynamics of Short-Video Traffic Pools: Step Thresholds in Completion, Engagement and Replay and a Zero-Ad-Spend Breakout Test Matrix (Version 1.0.0), Chapter 8: Closing the Loop: the Weekly Sheet, the Promotion Calculator and a Pre-registered Refutation Plan. 5D Library. DOI pending
DOI pending · monograph page goes live after publication · 5DL-B-273 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026273,
author = {Fan, Yafu},
title = {Cold-Start Dynamics of Short-Video Traffic Pools: Step Thresholds in Completion, Engagement and Replay and a Zero-Ad-Spend Breakout Test Matrix},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}Other chapter answers in this book
- Chapter 1 · Promotion as a Product: Four Threshold Gates and the Nonlinear Return to Quality
- Chapter 2 · Branching Processes: Galton–Watson Cascades, Criticality and Overdispersion
- Chapter 3 · Why the Step Becomes a Ramp: Finite Batches, Logistic Promotion and the Bottleneck Gate
- Chapter 4 · Completion, Engagement, Replay: the Statistics of Three Rulers
- Chapter 5 · More Posts or Better Posts: the Linear Ledger and the Multiplicative Ledger
- Chapter 6 · The Zero-Ad-Spend Test Matrix: a Twelve-Run Plackett–Burman Design and Its Sample Size
- Chapter 7 · Classical Sources East and West: Sunzi's "Disposition" and Galton–Watson